quirky update - Page 751

Global Email Forwarding with EXIM?

- by Dexirian

Been trying to find a solution to this for a while without success so here i go : I was given the task to build a High-Availability Load-Balanced Network Cluster for our 2 linux servers. I did some workaround and managed to get a DNS + SQL + Web Folders + Mails synchronisation going between both. Now i would like my server 2 to only do mailing and server 1 to only do web hosting. I transfered all the accounts for 1 to 2 using the WHM built-in account transfert feature. I created 2 different rsync jobs that sync, update, and delete the files for mail and websites. Now i was able to successfully transfer 1 mail accounts from 1 to 2, and the server 2 works flawlessly. All i had to do was change the MX entries to point to the new server and bingo. Now my problem is, some clients have their mail softwares configured so that they point to oldserver.domain.com. I cant make the (A) entry of oldserver.domain.com point to the new server for obvious reasons. I thought of using .foward files and add them to the home directories of the concerned users but that would be very difficult. So my question is : Is there a way to configure exim so that it will only foward mails to the new server? I need to change all the users so they use their mail on server 2 without them doing anything. Thanks! EDIT : TO CLARIFY MY PROBLEM Some clients have their mail point to oldserver.xyz instead of mail.olderserver.xyz I want to know if i can do something to prevent modifying the clients configuration I would also like to know is there is a way to find out what clients aren't properly configured

Read the article

Accessing CIFS shares from an OS X machine incredibly slow

- by Aron Rotteveel

This is a longshot, because this issue seems over-reported and unanswered on the internet (see references below), but it is about time this issue is permanently solved. The facts: Server: Windows Server 2008, acting as a file server Client: OS X Lion 10.7.3. Method of connecting: directly via IP through Finder: smb://192.168.1.100/share The problem: The initial connection attempt takes about a minute. After the connection is made, it takes one more minute to show the directories in Finder. After navigating to any other directory, it takes several seconds/minutes to parse the directory, seemingly based on the size of the contents. Actually, my entire Finder has this problem after connecting. When using Finder to show my desktop, it can literally take up to a minute to load. Obviously, this is not right. I have no clue how to fix this and would appreciate any help I can get. I am unsure about other relevant information I can provide, but if there is any, please let me know so I can update the post. I seem to be not the only one having this problem: Most importantly, an apple.stackexchange.com entry. Unsolved and unanswered. Several users on the Apple support forums. Users on EduGeek.

Read the article

Installing Debian 7.6.0 on Lenovo Y50

- by Girauder

I was trying to install Debian on my new laptop: a Lenovo Y50 64bit running Windows 8. I got together with a friend and installed Debian in his computer first and had no problems. However I've tried to install Debian several times using the AMD64 KDE and netinst versions and accomplished nothing. First try: installed the KDE version. Grub would let me choose which operating system I wanted, but when I selected Debian it would only load the command line. Second try: Reinstalled but this time with the netinst version. I only got a black screen where I could type but nothing else. Third Try. Tried the netinst again. This time after making the partitions I got a message that said that no EFI partition was found. I ignored the message and this time it wouldn't even load Grub. only a command like interface with grub rescue or something. Not once did I get an error during the installation. What am I doing wrong? I assume the problem is I need to make an EFI partition or something like that. So why is it that during the first installations I didn't ask me for that. And if that is indeed the problem, How can I solve it? Update So the installation failed again... as predicted. Here you can find the Disk Management picture. http://postimg.org/image/433cpfkjz/ Please somebody help me. I keep getting the grub rescue thing. secure boot is disabled and legacy support is set first.

Read the article

how to remove location block from $uri in nginx configuration?

- by Jason

I have a rewrite in my ngix conf file that works properly except it seems to include the location block as part of the $uri variable. I only want the path after the location block. My current config code is: location /cargo { try_files $uri $uri/ /cargo/index.php?_REWRITE_COMMAND=$uri&args; } Using an example url of http://localhost/cargo/testpage the redirect works, however the value of the "_REWRITE_COMMAND" parameter received by my php file is "/cargo/testpage". I need to strip off the location block and just have "testpage" as the $uri I am pretty sure there is a regex syntax to split the $uri and assign it to a new variable using $1 $2 etc, but I can't find any example to do just a variable assignment using a regex that is not part of a rewrite statement. I've been looking and trying for hours and I just can't seem to get past this last step. I also know I could just strip this out on the application code, but the reason I want to try to fix it in the nginx conf is for compatibility reasons as it also runs on Apache. I also should say that I have figured out a really hacky way to do it, but it involves an "if" statement to check for file existance and the documentation specifically says not to do it that way. -- UPDATE: ANSWERED BY theuni: The regex goes in the location block definition. one note of caution is that php handler location needs to be ABOVE this location, otherwise you will get a server error because it goes into an infinite redirect loop location ~ ^/cargo/(.*) { try_files $1 /cargo/$1/ /cargo/index.php?_REWRITE_COMMAND=$1&args; }

Read the article

where are the default ulimit values set? (linux, centos)

- by nomercysir

I have two CentOS (5) servers with nearly identical specs. When I login and do: ulimit -u on one machine, I get unlimited and on the other, 77824 When I run a cron like * * * * * ulimit -u > ulimit.txt I get the same results (unlimited, 77824). I am trying to determine where these are set so that I can alter them. They are not set in any of my profiles (.bashrc, /etc/profile, etc .. these wouldn't affect cron anyway) nor in /etc/security/limits.conf (which is empty). I have scoured google and even gone so far as to do grep -Ir 77824 / but nothing has turned up so far. I don't understand how these machines could have come preset with different limits. I am actually wondering not for these machines, but for a different (CentOS 6) machine which has a limit of 1024, which is far too small. I need to run cron jobs with a higher limit and the only way I know how to set that is in the cron job itself. That's ok, but I'd rather set it system wide so it's not as hacky. Thanks for any help. This seems like it should be easy (NOT) EDIT -- SOLVED Ok, I figured this out. It seems to be an issue either with CentOS 6 or perhaps my machine configuration. On the CentOS 5 configuration, I can set in /etc/security/limits.conf: * - nproc unlimited and that would effectively update the accounts and cron limits. However, this does not work in my CentOS 6 box. Instead, I must do: myname1 - nproc unlimited myname2 - nproc unlimited ... And things work as expected. Maybe the UID specification works to, but the wildcard (*) definitely DOES NOT here. Oddly, wildcards DO work for the 'nofile' limit. I still would love to know where the default values are actually coming from, because by default, this file is empty and I couldn't see why I had different defaults for the two CentOS boxes, which had identical hardware and were from the same provider.

Read the article

Ethernet/8P8C crimp contacts bent

- by Fire Lancer

(if anyone knows correct terminology please correct). Ive got a (fairly large) number of existing Ethernet cables that over the years many have got damaged connector clips, so got a crimp tool and some new connectors for them. However out of all 4 attempts I have tried, on crimping 2+ of the little copper contacts that bite into the wires have instead just bent to one side, and so gone between the gaps in in the crimp tool... Unless this really is me doing something wrong (what?) I am inclined to blame the hardware, but is this the crimper or the new connectors I got? I tried to take a picture, as you can just about see looking from the left 3rd, 6th, 7th and 8th pins didn't get pushed in, and so don't form a connector. Unfortunately my camera was barely able to focus on it and then this website converted it to a JPEG... Update: Connectors/Cable/Tools: The wires are stranded (looks about 6 and no evidence of being aluminum/not copper), and the pins(?) have 2 little flat spikes lengthways along the cables (I understand to dig into it, while solid core connectors would have like 2 plates designed to go around the core?). Crimper was http://www.amazon.co.uk/gp/product/B0013EXTKK/ref=oh_aui_detailpage_o02_s00?ie=UTF8&psc=1 (seemed to be highly rated, I already had tools for cutting/stripping). Update2: Picture of crimp "prongs" (?) Update3: Side picture of connector Update4: Comparison with old connector. The top (used) connector is one from a few years back (different tool and connectors), the thing that concerns me that it might not be the tool I need to replace is just how thin the pins are on the new one that maybe a tool could legitimately bend some into a gap rather than pushing them in fully? In fact I can move individual pins to the sides significantly with my fingernail, is that normal?

Read the article

IE8 page reload hangs

- by Rod

When a 7mb HTM file is first opened (by clicking on the file icon or using Open With), both IE8 and Firefox display this browser file quickly. After the file is closed, Firefox will reopen this file quickly, but IE8 appears to hang during the reopen. Clearing the IE cache does not help. However, IE will reopen the file quickly again only if the File/Open/Browse feature of the menu bar is used (clicking on the file icon can be used only once between computer reboots). Testing suggests that the problem relates to the number of HTML hyperlinks pointing to another part of the file. There are many hyperlinks, but they are not a problem during the first load of the document (between computer reboots). What needs to be fixed to avoid use of the workaround? Using Windows XP SP3 Update 6/23/12 - Controlled testing shows that the number of hyperlinks is not the problem. The way this large file is opened is the difference: 1) from the IE menu bar, File/Open/Browse is consistent and fast (but not as fast as FF). 2) clicking on the file name in the folder (even when IE is the default program for this file type) causes a much delayed load of the file. Creating a smaller file demonstrates the delayed load, but verifies that the load eventually occurs.

Read the article

[deb-5.0] Setup DNS on my server so I can put my IPs in as nameservers of my domain provider

- by Maurycy Zarzycki

Basically, my unmanaged VPS provider doesn't supply me with nameserver which I can use with my domain provider to route domain to my server. As I've been told: You need to configure the custom DNS server in your VPS, to setup the custom nameservers. Please refer the following article that would help: http://www.linuxhomenetworking.com/wiki/index.php/Quick_HOWTO_:_Ch18_:_Configuring_DNS Once you configure the nameserver records, please update the domain registrar panel with the custom nameserver details. I tried to follow this guide but it seems to be a bit outdated, and I am complete newb with non-windows systems. I also scanned the google for other articles which could help me with this problem but, alas, nothing I found was of any value for someone who doesn't know this stuff better than his own pockets. I realize this is quite a complex thing to do, but maybe there is some way to automate it? Or a better solution, like a paid service which would act as my nameservers (this one would be interesting), or even hoped to find some company which "rents" people to do stuff like that. Blah, any help will be appreciated, I am at a complete loss here. I can follow some of these steps, but then I soon find that half of the files which are mentioned in the article are somehow not existing anywhere on the server which confuses me, and once we get to the point of creating Zone I can't really decipher all the things written there :/. As per title, my system is Debian 5.0.

Read the article

[deb-5.0] Setup DNS on my server so I can put my IPs in as nameservers of my domain provider

- by Maurycy Zarzycki

Basically, my unmanaged VPS provider doesn't supply me with nameserver which I can use with my domain provider to route domain to my server. As I've been told: You need to configure the custom DNS server in your VPS, to setup the custom nameservers. Please refer the following article that would help: http://www.linuxhomenetworking.com/wiki/index.php/Quick_HOWTO_:_Ch18_:_Configuring_DNS Once you configure the nameserver records, please update the domain registrar panel with the custom nameserver details. I tried to follow this guide but it seems to be a bit outdated, and I am complete newb with non-windows systems. I also scanned the google for other articles which could help me with this problem but, alas, nothing I found was of any value for someone who doesn't know this stuff better than his own pockets. I realize this is quite a complex thing to do, but maybe there is some way to automate it? Or a better solution, like a paid service which would act as my nameservers (this one would be interesting), or even hoped to find some company which "rents" people to do stuff like that. Blah, any help will be appreciated, I am at a complete loss here. I can follow some of these steps, but then I soon find that half of the files which are mentioned in the article are somehow not existing anywhere on the server which confuses me, and once we get to the point of creating Zone I can't really decipher all the things written there :/. As per title, my system is Debian 5.0.

Read the article

Help - since adding an elastic load balancer to my EC2 web application I cannot connect with the MySQL database (not in AWS)

- by undefined

I have a web application that uses an EC2 instance to receive uploaded images, resize and store on S3 and update my MySQL database with the image record. This database is hosted outside Amazon Web Services and so obviously involves communication between the EC2 instance and the database. Images are posted to the upload server from a Flash client which receives the IP address of the upload server when it is loaded and so sends images to 1.12.23.34/resize_script.php This has worked great .. until i started to try and include a load balancer. Since the ELBs do not use an IP address but a DNS address I am now passing this to Flash. Now when I upload images I get the following response from the server - Could not connect to MySQL: Lost connection to MySQL server at 'reading initial communication packet', system error: 111 What might be causing the lost connection to MySQL server. Is there any additional steps I need to take to allow my upload servers to be load balanced? I have set the host property of my MySQL privileges for this user to % any pointers greatly appreciated thanks.

Read the article

WINDOWS 7: Make the contents of two folders appear in one

- by big_smile

In Windows 7, I have three folders: "Images", "assets" and "all". I want the contents of "Images " and "assets" to appear in "all" automatically without copying those files into that folder (e.g. I don't want to duplicate the files). I also only want the contents to be copied over and not the folders themslves (The reason for this is that if the folders are copied over, they will become sub-directories. I am using a printing hot folder that access "all" but it can't see any subdirectories in "all"). When Images and Assets are updated (e.g. with files being added or deleted), "all" should automatically update as well. How can I do this? This is what I have tried: Libraries: This is a feature built into Windows. It works exactly as I want. However, the print hot folder cannot recognise the library as a folder. Sym Link Extension: I can use this to make the "images" and "assets" folders appear as a sub directories of "all". However, I want the contents of "images"/"assets" to appear in the "all" folder (I don't want the directories to appear as sub directories, because as stated, the print hot folder cannot access sub directories).

Read the article

Building a small server farm

- by RayQuang

Hi, I am planning to set up a Tech startup company that will provide web application solutions. Eventually we hope to diversify into different areas such as possibly social media or other services. For now we plan on running a high demand (from 1000 to 10,000 users in the first year) website running the application. This includes a MySQL database backend, email, and development servers. My question is then, what type of server arrangement will work best, that is ti say should i have a small cluster of ultra high power machines (E.G. Top of the range Xeons, with 12GB RAM) or will it be better to have more less powerful servers load balanced? Should I go for 1 - 2 u servers rack mounted ot would it be beter for it just to be tower servers for maintainability? Finally I would also like to know what kind of Internet and router i would need, I currently have 10mbit down and barely 1 mbit up, but soon our area will have a fiber optic connection with international speeds of up to 25 mbit / sec. Thanks in advance, RayQuang UPDATE: sorry I forgot to mention it, the platform that I will be using is PHP with the APC code cache, Probably running Debian.

Read the article

How to manage enterprise network of Linux machines?

- by killy9999

I work at the university. In my institute we have six computer laboratories used for teaching. Each lab has almost 20 computers, which gives over 100 machines total. Computers have either Windows XP or Windows 7 Eneterprise operating system. We use Symantec Ghost to manage all the computers. Each computer has a Ghost client installed, which allows to control computers over network. Every six months we restore a master image on one of the computers in a lab, update that image and distribute it over the network to all computers in a laboratory. Thanks to Ghost client this is done automatically with just a few clicks. Recently I suggested that it would be good to have Linux installed in the laboratories. The administrators were concerned that we would not be able to manage that many computers if each would have to be updated manually. The question is: how to manage such a huge network of Linux machines in an automated way? To make the description of our network more complete I'll add that all students have their accounts (about few thousand users) on a central server. These are accessed via LDAP. To use a computer in laboratory each student has to log in using his own account.

Read the article

Fedora 13 becomes unresponsive when too many applications running.

- by user61766

I am using Fedora 13 64bit on Dell Vostro full 4GB RAM system with default Gnome GUI. This is very annoying problem that I don't how to fix except by rebooting the whole PC. When I have too many applications running ( like browser windows), the system start acting sluggish. The fist symptoms appear in Eclipse IDE which becomes so terrible it just becomes frozen for sometime one whole minute after I try to edit something in the editor. Then Firefox seems like it has crashed. Google Chrome becomes very unresponsive as well. All GUI applications including File manager becomes unresponsive. When I check System Monitor, the CPU is still around 20% and memory is at 80% but the system seems getting fried up. This progressively becomes worse until I soft reboot it or if I dont do it evetually the whole system is fried, no response to any keyboard key or mouse and I have to hit the hardware turn off button. I regularly yum update the system but this makes no difference. Please don't tell not to run too many applications because I need those for my work. I thought Linux is well designed Operating System but I am very disappointed so far. Can some one here help ?

Read the article

Best practice ACLs to prepare for auditors?

- by Nic

An auditor will be visiting our office soon, and they will require read-only access to our data. I have already created a domain user account and placed them into a group called "Auditors". We have a single fileserver (Windows Server 2008) with about ten shared folders. All of the shares are set up to allow full access to authenticated users, and access restrictions are implemented with NTFS ACL's. Most folders allow full access to the "Domain Users" group, but the auditor won't need to make any changes. It takes several hours to update NTFS ACL's since we have about one million files. Here are the options that I am currently considering. Create a "staff" group to assign read/write instead of "Domain Users" at the share level Create a "staff" group to assign read/write instead of "Domain Users" at the NTFS level Deny access to the "Auditors" group at the share level Deny access to the "Auditors" group at the NTFS level Accept the status quo and trust the auditor. I will probably need to configure similar users in the future, as some of our contractors require a domain account but shouldn't be able to modify our client data. Is there a best practice for this?

Read the article

How to ensure local file is up-to-date or ahead (dropbox sync) before truecrypt auto-mount it?

- by user620965

There are a lot tutorials out there that states that dropbox build-in encryption is not secure enought. That tutorials recommands to sync a truecrypt container file to have all files in it securely encrypted. This setup is know to be limited. You can NOT have that truecrypt container file mounted on the same time on more than one location - if you have inserted changes to the contents of the container in more then one location at a time then this setup produces a conflict on the container file in the dropbox system - resulting in one container file for each location. In my case that issue is not relevant - i do not use my data on more than one location at a time. I want to use the auto-mount feature of truecrypt on startup of windows 7 to have a zero configuration environment - and start working right away. But i want to ensure that the local truecrypt container file is up-to-date before truecrypt mounts it automatically - imagine you updated the contents of the container on your primary location and your secondary location was off for a long time. In that case it can take "a long time" till dropbox sync is complete (e.g. depending on your internet connection and the size of the container file). There is a option in truecrypt that ensures that truecrypt do not update the timestamp of the container file - which speeds up the sync, because dropbox client is doing a differential sync then instead of a time consuming full-sync. That is an improvement to that setup, but this do not fix my issue. The question is how to make the auto-mount function wait for the container file to be up-to-date (updated by dropbox)? In contrast: if the file was changed local, but remote file (in the dropbox cloud system) is still old (not jet updated by the sync process / or process is progress), should not make truecrypt to wait for the sync. Suggestions?

Read the article

Mysql: Working With 192 Trillion Records... (Yes, 192 Trillion)

- by Sarah

Here's the question... Considering 192 trillion records, what should my considerations be? My main concern is speed. Here's the table... CREATE TABLE `ref` ( `id` INTEGER(13) AUTO_INCREMENT DEFAULT NOT NULL, `rel_id` INTEGER(13) NOT NULL, `p1` INTEGER(13) NOT NULL, `p2` INTEGER(13) DEFAULT NULL, `p3` INTEGER(13) DEFAULT NULL, `s` INTEGER(13) NOT NULL, `p4` INTEGER(13) DEFAULT NULL, `p5` INTEGER(13) DEFAULT NULL, `p6` INTEGER(13) DEFAULT NULL, PRIMARY KEY (`id`), KEY (`s`), KEY (`rel_id`), KEY (`p3`), KEY (`p4`) ); Here's the queries... SELECT id, s FROM ref WHERE red_id="$rel_id" AND p3="$p3" AND p4="$p4" SELECT rel_id, p1, p2, p3, p4, p5, p6 FROM ref WHERE id="$id" INSERT INTO rel (rel_id, p1, p2, p3, s, p4, p5, p6) VALUES ("$rel_id", "$p1", "$p2", "$p3", "$s", "$p4", "$p5", "$p6") Here's some notes... The SELECT's will be done much more frequently than the INSERT. However, occasionally I want to add a few hundred records at a time. Load-wise, there will be nothing for hours then maybe a few thousand queries all at once. Don't think I can normalize any more (need the p values in a combination) The database as a whole is very relational. This will be the largest table by far (next largest is about 900k) UPDATE (08/11/2010) Interestingly, I've been given a second option... Instead of 192 trillion I could store 2.6*10^16 (15 zeros, meaning 26 Quadrillion)... But in this second option I would only need to store one bigint(18) as the index in a table. That's it - just the one column. So I would just be checking for the existence of a value. Occasionally adding records, never deleting them. So that makes me think there must be a better solution then mysql for simply storing numbers... Given this second option, should I take it or stick with the first... [edit] Just got news of some testing that's been done - 100 million rows with this setup returns the query in 0.0004 seconds [/edit]

Read the article

Provide a user with service start/stop permissions

- by slakr007

I have a very basic domain that I use for development. I want to create a GPO that provides users in the Backup Operators group with start/stop permissions for two specific services on a specific server. I have read several articles about this, and they all indicate that this is very easy. Create a GPO, give the user start/stop permissions to the services under Computer Configuration Policies Windows Settings Security Settings System Services, and voila. Done. Not so much, but I have to be doing something wrong. My install is pretty much the default. The domain controller is in the Domain Controllers OU, the Backup Operators group is under Builtin, and I created a user called Backup under Users. I created a GPO and linked it to the Domain Controllers OU. In the GPO I give the Backup user permission to start/stop two specific services on the server. I forced an update with gpupdate. I used Group Policy Results to verify that my GPO is the winning GPO giving the user the permission to start/stop the two services. However, the user is still unable to start/stop the services. I attempted different loopback settings on the GPO to no avail. I'm sort of at a loss here.

Read the article

Windows 7 Upgrade Fail from Home Premium to Ultimate Professional

- by Michael S-B

I had a hard drive crash, which meant I had to install a new HDD in my Dell 64-Bit XPS 1350 (lovely computer). I had previously been running Windows 7 Ultimate Professional which I had upgraded from the OEM Win 7 Home Premium by means of a disk I purchased from my university. Using the Recovery disk from Dell I installed Windows 7 Home Premium successfully on the new hard drive, but when I have tried to upgrade via my disk to Ultimate it installs the whole thing, says its complete, but when I reboot, tells me: "This version of Windows could not be installed. Your previous version of Windows has been restored, and you can continue to use it." I've installed the drivers from Dell's driver disk, but still to no avail. I've also used Driver Robot to update all my drivers. I can't find a .dmp file anywhere under C:\$WINDOWS.~BT\Sources but I did find this file under C:\$WINDOWS.~BT\Sources\Panther. setupact.log https://www.dropbox.com/s/yzy7fhkxlzc235y/setupact.log If anyone could please advise me what I need to do to fix Windows so it will upgrade properly, I would greatly appreciate it.

Read the article

How many VPS do I need for my website? [duplicate]

- by michael

This question already has an answer here: How do you do load testing and capacity planning for web sites? 3 answers I made a website which aims at simulating a trading market. There are a list of prices and corresponding volumes that people want to purchase. Users can purchase at any price any time. My website retrieves the prices and volumes from my database every 2 seconds (I have to update the user's browser frequently to allow them to see the current market). Users' database INSERT query can be sent any time if they purchase. I used ajax to post or get data from my database (sometimes nested ajax calls). So, every 2 seconds, each user will send or retrieve data by using more than 20 database queries (in order to show a users the current prices and volumes). Also, I may have 200 users at a time. I was not using VPS before, and I got banned because of using too much CPU resources on my host. Now, I've purchased VPS*2 from a hosting servers. I have: CPU Speed: 2000 Mhz Memory: 2048 MB Disk Space: 20000 MB Bandwidth: 2000 GB Connection: 40 Mb/s Dedicated IP's 2 IP's Is this enough for my 200 users? Also, which VPS OS is suitable for me? Thank you.

Read the article

100% CPU use when new usb device plugged in - services.exe / Windows Server 2003

- by Will3265

On my server I am trying to install a new usb drive but all that happens is that the system starts using huge amounts of processor cycles with services.exe. On closer inspection with process explorer there is a thread called umpnpmgr.dll using most of the services.exe processor time. I left it for a half hour and still nothing happened. Rebooted and tried again, same result. Tried a different usb drive, then a flash drive but still same issue. Tried updating driver but it said the update function was already in action. I have used process explorer to kill the thread now so the server can still perform its intended functions. Any device that was previously installed before this began happening will still work but any device new to the system will not. My question(s) is/are: Is there a way to manually install the device into the registry so Windows thinks it is a previously installed device? Or can this problem be repaired through anything other than a reinstall? To do a reinstall would mean backing up large amounts of data which is hard with a usb drive and insufficient space on all other network machines. Any help would be greatly appreciated. William

Read the article

A free standing ASP.NET Pager Web Control

- by Rick Strahl

Paging in ASP.NET has been relatively easy with stock controls supporting basic paging functionality. However, recently I built an MVC application and one of the things I ran into was that I HAD TO build manual paging support into a few of my pages. Dealing with list controls and rendering markup is easy enough, but doing paging is a little more involved. I ended up with a small but flexible component that can be dropped anywhere. As it turns out the task of creating a semi-generic Pager control for MVC was fairly easily. Now I’m back to working in Web Forms and thought to myself that the way I created the pager in MVC actually would also work in ASP.NET – in fact quite a bit easier since the whole thing can be conveniently wrapped up into an easily reusable control. A standalone pager would provider easier reuse in various pages and a more consistent pager display regardless of what kind of 'control’ the pager is associated with. Why a Pager Control? At first blush it might sound silly to create a new pager control – after all Web Forms has pretty decent paging support, doesn’t it? Well, sort of. Yes the GridView control has automatic paging built in and the ListView control has the related DataPager control. The built in ASP.NET paging has several issues though: Postback and JavaScript requirements If you look at paging links in ASP.NET they are always postback links with javascript:__doPostback() calls that go back to the server. While that works fine and actually has some benefit like the fact that paging saves changes to the page and post them back, it’s not very SEO friendly. Basically if you use javascript based navigation nosearch engine will follow the paging links which effectively cuts off list content on the first page. The DataPager control does support GET based links via the QueryStringParameter property, but the control is effectively tied to the ListView control (which is the only control that implements IPageableItemContainer). DataSource Controls required for Efficient Data Paging Retrieval The only way you can get paging to work efficiently where only the few records you display on the page are queried for and retrieved from the database you have to use a DataSource control - only the Linq and Entity DataSource controls support this natively. While you can retrieve this data yourself manually, there’s no way to just assign the page number and render the pager based on this custom subset. Other than that default paging requires a full resultset for ASP.NET to filter the data and display only a subset which can be very resource intensive and wasteful if you’re dealing with largish resultsets (although I’m a firm believer in returning actually usable sets :-}). If you use your own business layer that doesn’t fit an ObjectDataSource you’re SOL. That’s a real shame too because with LINQ based querying it’s real easy to retrieve a subset of data that is just the data you want to display but the native Pager functionality doesn’t support just setting properties to display just the subset AFAIK. DataPager is not Free Standing The DataPager control is the closest thing to a decent Pager implementation that ASP.NET has, but alas it’s not a free standing component – it works off a related control and the only one that it effectively supports from the stock ASP.NET controls is the ListView control. This means you can’t use the same data pager formatting for a grid and a list view or vice versa and you’re always tied to the control. Paging Events In order to handle paging you have to deal with paging events. The events fire at specific time instances in the page pipeline and because of this you often have to handle data binding in a way to work around the paging events or else end up double binding your data sources based on paging. Yuk. Styling The GridView pager is a royal pain to beat into submission for styled rendering. The DataPager control has many more options and template layout and it renders somewhat cleaner, but it too is not exactly easy to get a decent display for. Not a Generic Solution The problem with the ASP.NET controls too is that it’s not generic. GridView, DataGrid use their own internal paging, ListView can use a DataPager and if you want to manually create data layout – well you’re on your own. IOW, depending on what you use you likely have very different looking Paging experiences. So, I figured I’ve struggled with this once too many and finally sat down and built a Pager control. The Pager Control My goal was to create a totally free standing control that has no dependencies on other controls and certainly no requirements for using DataSource controls. The idea is that you should be able to use this pager control without any sort of data requirements at all – you should just be able to set properties and be able to display a pager. The Pager control I ended up with has the following features: Completely free standing Pager control – no control or data dependencies Complete manual control – Pager can render without any data dependency Easy to use: Only need to set PageSize, ActivePage and TotalItems Supports optional filtering of IQueryable for efficient queries and Pager rendering Supports optional full set filtering of IEnumerable<T> and DataTable Page links are plain HTTP GET href Links Control automatically picks up Page links on the URL and assigns them (automatic page detection no page index changing events to hookup) Full CSS Styling support On the downside there’s no templating support for the control so the layout of the pager is relatively fixed. All elements however are stylable and there are options to control the text, and layout options such as whether to display first and last pages and the previous/next buttons and so on. To give you an idea what the pager looks like, here are two differently styled examples (all via CSS): The markup for these two pagers looks like this: <ww:Pager runat="server" id="ItemPager" PageSize="5" PageLinkCssClass="gridpagerbutton" SelectedPageCssClass="gridpagerbutton-selected" PagesTextCssClass="gridpagertext" CssClass="gridpager" RenderContainerDiv="true" ContainerDivCssClass="gridpagercontainer" MaxPagesToDisplay="6" PagesText="Item Pages:" NextText="next" PreviousText="previous" /> <ww:Pager runat="server" id="ItemPager2" PageSize="5" RenderContainerDiv="true" MaxPagesToDisplay="6" /> The latter example uses default style settings so it there’s not much to set. The first example on the other hand explicitly assigns custom styles and overrides a few of the formatting options. Styling The styling is based on a number of CSS classes of which the the main pager, pagerbutton and pagerbutton-selected classes are the important ones. Other styles like pagerbutton-next/prev/first/last are based on the pagerbutton style. The default styling shown for the red outlined pager looks like this: .pagercontainer { margin: 20px 0; background: whitesmoke; padding: 5px; } .pager { float: right; font-size: 10pt; text-align: left; } .pagerbutton,.pagerbutton-selected,.pagertext { display: block; float: left; text-align: center; border: solid 2px maroon; min-width: 18px; margin-left: 3px; text-decoration: none; padding: 4px; } .pagerbutton-selected { font-size: 130%; font-weight: bold; color: maroon; border-width: 0px; background: khaki; } .pagerbutton-first { margin-right: 12px; } .pagerbutton-last,.pagerbutton-prev { margin-left: 12px; } .pagertext { border: none; margin-left: 30px; font-weight: bold; } .pagerbutton a { text-decoration: none; } .pagerbutton:hover { background-color: maroon; color: cornsilk; } .pagerbutton-prev { background-image: url(images/prev.png); background-position: 2px center; background-repeat: no-repeat; width: 35px; padding-left: 20px; } .pagerbutton-next { background-image: url(images/next.png); background-position: 40px center; background-repeat: no-repeat; width: 35px; padding-right: 20px; margin-right: 0px; } Yup that’s a lot of styling settings although not all of them are required. The key ones are pagerbutton, pager and pager selection. The others (which are implicitly created by the control based on the pagerbutton style) are for custom markup of the ‘special’ buttons. In my apps I tend to have two kinds of pages: Those that are associated with typical ‘grid’ displays that display purely tabular data and those that have a more looser list like layout. The two pagers shown above represent these two views and the pager and gridpager styles in my standard style sheet reflect these two styles. Configuring the Pager with Code Finally lets look at what it takes to hook up the pager. As mentioned in the highlights the Pager control is completely independent of other controls so if you just want to display a pager on its own it’s as simple as dropping the control and assigning the PageSize, ActivePage and either TotalPages or TotalItems. So for this markup: <ww:Pager runat="server" id="ItemPagerManual" PageSize="5" MaxPagesToDisplay="6" /> I can use code as simple as: ItemPagerManual.PageSize = 3; ItemPagerManual.ActivePage = 4;ItemPagerManual.TotalItems = 20; Note that ActivePage is not required - it will automatically use any Page=x query string value and assign it, although you can override it as I did above. TotalItems can be any value that you retrieve from a result set or manually assign as I did above. A more realistic scenario based on a LINQ to SQL IQueryable result is even easier. In this example, I have a UserControl that contains a ListView control that renders IQueryable data. I use a User Control here because there are different views the user can choose from with each view being a different user control. This incidentally also highlights one of the nice features of the pager: Because the pager is independent of the control I can put the pager on the host page instead of into each of the user controls. IOW, there’s only one Pager control, but there are potentially many user controls/listviews that hold the actual display data. The following code demonstrates how to use the Pager with an IQueryable that loads only the records it displays: protected voidPage_Load(objectsender, EventArgs e) { Category = Request.Params["Category"] ?? string.Empty; IQueryable<wws_Item> ItemList = ItemRepository.GetItemsByCategory(Category); // Update the page and filter the list down ItemList = ItemPager.FilterIQueryable<wws_Item>(ItemList); // Render user control with a list view Control ulItemList = LoadControl("~/usercontrols/" + App.Configuration.ItemListType + ".ascx"); ((IInventoryItemListControl)ulItemList).InventoryItemList = ItemList; phItemList.Controls.Add(ulItemList); // placeholder } The code uses a business object to retrieve Items by category as an IQueryable which means that the result is only an expression tree that hasn’t execute SQL yet and can be further filtered. I then pass this IQueryable to the FilterIQueryable() helper method of the control which does two main things: Filters the IQueryable to retrieve only the data displayed on the active page Sets the Totaltems property and calculates TotalPages on the Pager and that’s it! When the Pager renders it uses those values, plus the PageSize and ActivePage properties to render the Pager. In addition to IQueryable there are also filter methods for IEnumerable<T> and DataTable, but these versions just filter the data by removing rows/items from the entire already retrieved data. Output Generated and Paging Links The output generated creates pager links as plain href links. Here’s what the output looks like: <div id="ItemPager" class="pagercontainer"> <div class="pager"> <span class="pagertext">Pages: </span><a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=1" class="pagerbutton" />1</a> <a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=2" class="pagerbutton" />2</a> <a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=3" class="pagerbutton" />3</a> <span class="pagerbutton-selected">4</span> <a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=5" class="pagerbutton" />5</a> <a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=6" class="pagerbutton" />6</a> <a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=20" class="pagerbutton pagerbutton-last" />20</a> <a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=3" class="pagerbutton pagerbutton-prev" />Prev</a> <a href="http://localhost/WestWindWebStore/itemlist.aspx?Page=5" class="pagerbutton pagerbutton-next" />Next</a></div> <br clear="all" /> </div> </div> The links point back to the current page and simply append a Page= page link into the page. When the page gets reloaded with the new page number the pager automatically detects the page number and automatically assigns the ActivePage property which results in the appropriate page to be displayed. The code shown in the previous section is all that’s needed to handle paging. Note that HTTP GET based paging is different than the Postback paging ASP.NET uses by default. Postback paging preserves modified page content when clicking on pager buttons, but this control will simply load a new page – no page preservation at this time. The advantage of not using Postback paging is that the URLs generated are plain HTML links that a search engine can follow where __doPostback() links are not. Pager with a Grid The pager also works in combination with grid controls so it’s easy to bypass the grid control’s paging features if desired. In the following example I use a gridView control and binds it to a DataTable result which is also filterable by the Pager control. The very basic plain vanilla ASP.NET grid markup looks like this: <div style="width: 600px; margin: 0 auto;padding: 20px; "> <asp:DataGrid runat="server" AutoGenerateColumns="True" ID="gdItems" CssClass="blackborder" style="width: 600px;"> <AlternatingItemStyle CssClass="gridalternate" /> <HeaderStyle CssClass="gridheader" /> </asp:DataGrid> <ww:Pager runat="server" ID="Pager" CssClass="gridpager" ContainerDivCssClass="gridpagercontainer" PageLinkCssClass="gridpagerbutton" SelectedPageCssClass="gridpagerbutton-selected" PageSize="8" RenderContainerDiv="true" MaxPagesToDisplay="6" /> </div> and looks like this when rendered: using custom set of CSS styles. The code behind for this code is also very simple: protected void Page_Load(object sender, EventArgs e) { string category = Request.Params["category"] ?? ""; busItem itemRep = WebStoreFactory.GetItem(); var items = itemRep.GetItemsByCategory(category) .Select(itm => new {Sku = itm.Sku, Description = itm.Description}); // run query into a DataTable for demonstration DataTable dt = itemRep.Converter.ToDataTable(items,"TItems"); // Remove all items not on the current page dt = Pager.FilterDataTable(dt,0); // bind and display gdItems.DataSource = dt; gdItems.DataBind(); } A little contrived I suppose since the list could already be bound from the list of elements, but this is to demonstrate that you can also bind against a DataTable if your business layer returns those. Unfortunately there’s no way to filter a DataReader as it’s a one way forward only reader and the reader is required by the DataSource to perform the bindings. However, you can still use a DataReader as long as your business logic filters the data prior to rendering and provides a total item count (most likely as a second query). Control Creation The control itself is a pretty brute force ASP.NET control. Nothing clever about this other than some basic rendering logic and some simple calculations and update routines to determine which buttons need to be shown. You can take a look at the full code from the West Wind Web Toolkit’s Repository (note there are a few dependencies). To give you an idea how the control works here is the Render() method: /// <summary> /// overridden to handle custom pager rendering for runtime and design time /// </summary> /// <param name="writer"></param> protected override void Render(HtmlTextWriter writer) { base.Render(writer); if (TotalPages == 0 && TotalItems > 0) TotalPages = CalculateTotalPagesFromTotalItems(); if (DesignMode) TotalPages = 10; // don't render pager if there's only one page if (TotalPages < 2) return; if (RenderContainerDiv) { if (!string.IsNullOrEmpty(ContainerDivCssClass)) writer.AddAttribute("class", ContainerDivCssClass); writer.RenderBeginTag("div"); } // main pager wrapper writer.WriteBeginTag("div"); writer.AddAttribute("id", this.ClientID); if (!string.IsNullOrEmpty(CssClass)) writer.WriteAttribute("class", this.CssClass); writer.Write(HtmlTextWriter.TagRightChar + "\r\n"); // Pages Text writer.WriteBeginTag("span"); if (!string.IsNullOrEmpty(PagesTextCssClass)) writer.WriteAttribute("class", PagesTextCssClass); writer.Write(HtmlTextWriter.TagRightChar); writer.Write(this.PagesText); writer.WriteEndTag("span"); // if the base url is empty use the current URL FixupBaseUrl(); // set _startPage and _endPage ConfigurePagesToRender(); // write out first page link if (ShowFirstAndLastPageLinks && _startPage != 1) { writer.WriteBeginTag("a"); string pageUrl = StringUtils.SetUrlEncodedKey(BaseUrl, QueryStringPageField, (1).ToString()); writer.WriteAttribute("href", pageUrl); if (!string.IsNullOrEmpty(PageLinkCssClass)) writer.WriteAttribute("class", PageLinkCssClass + " " + PageLinkCssClass + "-first"); writer.Write(HtmlTextWriter.SelfClosingTagEnd); writer.Write("1"); writer.WriteEndTag("a"); writer.Write(" "); } // write out all the page links for (int i = _startPage; i < _endPage + 1; i++) { if (i == ActivePage) { writer.WriteBeginTag("span"); if (!string.IsNullOrEmpty(SelectedPageCssClass)) writer.WriteAttribute("class", SelectedPageCssClass); writer.Write(HtmlTextWriter.TagRightChar); writer.Write(i.ToString()); writer.WriteEndTag("span"); } else { writer.WriteBeginTag("a"); string pageUrl = StringUtils.SetUrlEncodedKey(BaseUrl, QueryStringPageField, i.ToString()).TrimEnd('&'); writer.WriteAttribute("href", pageUrl); if (!string.IsNullOrEmpty(PageLinkCssClass)) writer.WriteAttribute("class", PageLinkCssClass); writer.Write(HtmlTextWriter.SelfClosingTagEnd); writer.Write(i.ToString()); writer.WriteEndTag("a"); } writer.Write("\r\n"); } // write out last page link if (ShowFirstAndLastPageLinks && _endPage < TotalPages) { writer.WriteBeginTag("a"); string pageUrl = StringUtils.SetUrlEncodedKey(BaseUrl, QueryStringPageField, TotalPages.ToString()); writer.WriteAttribute("href", pageUrl); if (!string.IsNullOrEmpty(PageLinkCssClass)) writer.WriteAttribute("class", PageLinkCssClass + " " + PageLinkCssClass + "-last"); writer.Write(HtmlTextWriter.SelfClosingTagEnd); writer.Write(TotalPages.ToString()); writer.WriteEndTag("a"); } // Previous link if (ShowPreviousNextLinks && !string.IsNullOrEmpty(PreviousText) && ActivePage > 1) { writer.Write(" "); writer.WriteBeginTag("a"); string pageUrl = StringUtils.SetUrlEncodedKey(BaseUrl, QueryStringPageField, (ActivePage - 1).ToString()); writer.WriteAttribute("href", pageUrl); if (!string.IsNullOrEmpty(PageLinkCssClass)) writer.WriteAttribute("class", PageLinkCssClass + " " + PageLinkCssClass + "-prev"); writer.Write(HtmlTextWriter.SelfClosingTagEnd); writer.Write(PreviousText); writer.WriteEndTag("a"); } // Next link if (ShowPreviousNextLinks && !string.IsNullOrEmpty(NextText) && ActivePage < TotalPages) { writer.Write(" "); writer.WriteBeginTag("a"); string pageUrl = StringUtils.SetUrlEncodedKey(BaseUrl, QueryStringPageField, (ActivePage + 1).ToString()); writer.WriteAttribute("href", pageUrl); if (!string.IsNullOrEmpty(PageLinkCssClass)) writer.WriteAttribute("class", PageLinkCssClass + " " + PageLinkCssClass + "-next"); writer.Write(HtmlTextWriter.SelfClosingTagEnd); writer.Write(NextText); writer.WriteEndTag("a"); } writer.WriteEndTag("div"); if (RenderContainerDiv) { if (RenderContainerDivBreak) writer.Write("<br clear=\"all\" />\r\n"); writer.WriteEndTag("div"); } } As I said pretty much brute force rendering based on the control’s property settings of which there are quite a few: You can also see the pager in the designer above. unfortunately the VS designer (both 2010 and 2008) fails to render the float: left CSS styles properly and starts wrapping after margins are applied in the special buttons. Not a big deal since VS does at least respect the spacing (the floated elements overlay). Then again I’m not using the designer anyway :-}. Filtering Data What makes the Pager easy to use is the filter methods built into the control. While this functionality is clearly not the most politically correct design choice as it violates separation of concerns, it’s very useful for typical pager operation. While I actually have filter methods that do something similar in my business layer, having it exposed on the control makes the control a lot more useful for typical databinding scenarios. Of course these methods are optional – if you have a business layer that can provide filtered page queries for you can use that instead and assign the TotalItems property manually. There are three filter method types available for IQueryable, IEnumerable and for DataTable which tend to be the most common use cases in my apps old and new. The IQueryable version is pretty simple as it can simply rely on on .Skip() and .Take() with LINQ: /// <summary> /// <summary> /// Queries the database for the ActivePage applied manually /// or from the Request["page"] variable. This routine /// figures out and sets TotalPages, ActivePage and /// returns a filtered subset IQueryable that contains /// only the items from the ActivePage. /// </summary> /// <param name="query"></param> /// <param name="activePage"> /// The page you want to display. Sets the ActivePage property when passed. /// Pass 0 or smaller to use ActivePage setting. /// </param> /// <returns></returns> public IQueryable<T> FilterIQueryable<T>(IQueryable<T> query, int activePage) where T : class, new() { ActivePage = activePage < 1 ? ActivePage : activePage; if (ActivePage < 1) ActivePage = 1; TotalItems = query.Count(); if (TotalItems <= PageSize) { ActivePage = 1; TotalPages = 1; return query; } int skip = ActivePage - 1; if (skip > 0) query = query.Skip(skip * PageSize); _TotalPages = CalculateTotalPagesFromTotalItems(); return query.Take(PageSize); } The IEnumerable<T> version simply converts the IEnumerable to an IQuerable and calls back into this method for filtering. The DataTable version requires a little more work to manually parse and filter records (I didn’t want to add the Linq DataSetExtensions assembly just for this): /// <summary> /// Filters a data table for an ActivePage. /// /// Note: Modifies the data set permanently by remove DataRows /// </summary> /// <param name="dt">Full result DataTable</param> /// <param name="activePage">Page to display. 0 to use ActivePage property </param> /// <returns></returns> public DataTable FilterDataTable(DataTable dt, int activePage) { ActivePage = activePage < 1 ? ActivePage : activePage; if (ActivePage < 1) ActivePage = 1; TotalItems = dt.Rows.Count; if (TotalItems <= PageSize) { ActivePage = 1; TotalPages = 1; return dt; } int skip = ActivePage - 1; if (skip > 0) { for (int i = 0; i < skip * PageSize; i++ ) dt.Rows.RemoveAt(0); } while(dt.Rows.Count > PageSize) dt.Rows.RemoveAt(PageSize); return dt; } Using the Pager Control The pager as it is is a first cut I built a couple of weeks ago and since then have been tweaking a little as part of an internal project I’m working on. I’ve replaced a bunch of pagers on various older pages with this pager without any issues and have what now feels like a more consistent user interface where paging looks and feels the same across different controls. As a bonus I’m only loading the data from the database that I need to display a single page. With the preset class tags applied too adding a pager is now as easy as dropping the control and adding the style sheet for styling to be consistent – no fuss, no muss. Schweet. Hopefully some of you may find this as useful as I have or at least as a baseline to build ontop of… Resources The Pager is part of the West Wind Web & Ajax Toolkit Pager.cs Source Code (some toolkit dependencies) Westwind.css base stylesheet with .pager and .gridpager styles Pager Example Page © Rick Strahl, West Wind Technologies, 2005-2010Posted in ASP.NET

Read the article

Upload File to Windows Azure Blob in Chunks through ASP.NET MVC, JavaScript and HTML5

- by Shaun

Originally posted on: http://geekswithblogs.net/shaunxu/archive/2013/07/01/upload-file-to-windows-azure-blob-in-chunks-through-asp.net.aspxMany people are using Windows Azure Blob Storage to store their data in the cloud. Blob storage provides 99.9% availability with easy-to-use API through .NET SDK and HTTP REST. For example, we can store JavaScript files, images, documents in blob storage when we are building an ASP.NET web application on a Web Role in Windows Azure. Or we can store our VHD files in blob and mount it as a hard drive in our cloud service. If you are familiar with Windows Azure, you should know that there are two kinds of blob: page blob and block blob. The page blob is optimized for random read and write, which is very useful when you need to store VHD files. The block blob is optimized for sequential/chunk read and write, which has more common usage. Since we can upload block blob in blocks through BlockBlob.PutBlock, and them commit them as a whole blob with invoking the BlockBlob.PutBlockList, it is very powerful to upload large files, as we can upload blocks in parallel, and provide pause-resume feature. There are many documents, articles and blog posts described on how to upload a block blob. Most of them are focus on the server side, which means when you had received a big file, stream or binaries, how to upload them into blob storage in blocks through .NET SDK. But the problem is, how can we upload these large files from client side, for example, a browser. This questioned to me when I was working with a Chinese customer to help them build a network disk production on top of azure. The end users upload their files from the web portal, and then the files will be stored in blob storage from the Web Role. My goal is to find the best way to transform the file from client (end user’s machine) to the server (Web Role) through browser. In this post I will demonstrate and describe what I had done, to upload large file in chunks with high speed, and save them as blocks into Windows Azure Blob Storage. Traditional Upload, Works with Limitation The simplest way to implement this requirement is to create a web page with a form that contains a file input element and a submit button. 1: @using (Html.BeginForm("About", "Index", FormMethod.Post, new { enctype = "multipart/form-data" })) 2: { 3: <input type="file" name="file" /> 4: <input type="submit" value="upload" /> 5: } And then in the backend controller, we retrieve the whole content of this file and upload it in to the blob storage through .NET SDK. We can split the file in blocks and upload them in parallel and commit. The code had been well blogged in the community. 1: [HttpPost] 2: public ActionResult About(HttpPostedFileBase file) 3: { 4: var container = _client.GetContainerReference("test"); 5: container.CreateIfNotExists(); 6: var blob = container.GetBlockBlobReference(file.FileName); 7: var blockDataList = new Dictionary<string, byte[]>(); 8: using (var stream = file.InputStream) 9: { 10: var blockSizeInKB = 1024; 11: var offset = 0; 12: var index = 0; 13: while (offset < stream.Length) 14: { 15: var readLength = Math.Min(1024 * blockSizeInKB, (int)stream.Length - offset); 16: var blockData = new byte[readLength]; 17: offset += stream.Read(blockData, 0, readLength); 18: blockDataList.Add(Convert.ToBase64String(BitConverter.GetBytes(index)), blockData); 19: 20: index++; 21: } 22: } 23: 24: Parallel.ForEach(blockDataList, (bi) => 25: { 26: blob.PutBlock(bi.Key, new MemoryStream(bi.Value), null); 27: }); 28: blob.PutBlockList(blockDataList.Select(b => b.Key).ToArray()); 29: 30: return RedirectToAction("About"); 31: } This works perfect if we selected an image, a music or a small video to upload. But if I selected a large file, let’s say a 6GB HD-movie, after upload for about few minutes the page will be shown as below and the upload will be terminated. In ASP.NET there is a limitation of request length and the maximized request length is defined in the web.config file. It’s a number which less than about 4GB. So if we want to upload a really big file, we cannot simply implement in this way. Also, in Windows Azure, a cloud service network load balancer will terminate the connection if exceed the timeout period. From my test the timeout looks like 2 - 3 minutes. Hence, when we need to upload a large file we cannot just use the basic HTML elements. Besides the limitation mentioned above, the simple HTML file upload cannot provide rich upload experience such as chunk upload, pause and pause-resume. So we need to find a better way to upload large file from the client to the server. Upload in Chunks through HTML5 and JavaScript In order to break those limitation mentioned above we will try to upload the large file in chunks. This takes some benefit to us such as - No request size limitation: Since we upload in chunks, we can define the request size for each chunks regardless how big the entire file is. - No timeout problem: The size of chunks are controlled by us, which means we should be able to make sure request for each chunk upload will not exceed the timeout period of both ASP.NET and Windows Azure load balancer. It was a big challenge to upload big file in chunks until we have HTML5. There are some new features and improvements introduced in HTML5 and we will use them to implement our solution. In HTML5, the File interface had been improved with a new method called “slice”. It can be used to read part of the file by specifying the start byte index and the end byte index. For example if the entire file was 1024 bytes, file.slice(512, 768) will read the part of this file from the 512nd byte to 768th byte, and return a new object of interface called "Blob”, which you can treat as an array of bytes. In fact, a Blob object represents a file-like object of immutable, raw data. The File interface is based on Blob, inheriting blob functionality and expanding it to support files on the user's system. For more information about the Blob please refer here. File and Blob is very useful to implement the chunk upload. We will use File interface to represent the file the user selected from the browser and then use File.slice to read the file in chunks in the size we wanted. For example, if we wanted to upload a 10MB file with 512KB chunks, then we can read it in 512KB blobs by using File.slice in a loop. Assuming we have a web page as below. User can select a file, an input box to specify the block size in KB and a button to start upload. 1: <div> 2: <input type="file" id="upload_files" name="files[]" /><br /> 3: Block Size: <input type="number" id="block_size" value="512" name="block_size" />KB<br /> 4: <input type="button" id="upload_button_blob" name="upload" value="upload (blob)" /> 5: </div> Then we can have the JavaScript function to upload the file in chunks when user clicked the button. 1: <script type="text/javascript"> 1: 2: $(function () { 3: $("#upload_button_blob").click(function () { 4: }); 5: });</script> Firstly we need to ensure the client browser supports the interfaces we are going to use. Just try to invoke the File, Blob and FormData from the “window” object. If any of them is “undefined” the condition result will be “false” which means your browser doesn’t support these premium feature and it’s time for you to get your browser updated. FormData is another new feature we are going to use in the future. It could generate a temporary form for us. We will use this interface to create a form with chunk and associated metadata when invoked the service through ajax. 1: $("#upload_button_blob").click(function () { 2: // assert the browser support html5 3: if (window.File && window.Blob && window.FormData) { 4: alert("Your brwoser is awesome, let's rock!"); 5: } 6: else { 7: alert("Oh man plz update to a modern browser before try is cool stuff out."); 8: return; 9: } 10: }); Each browser supports these interfaces by their own implementation and currently the Blob, File and File.slice are supported by Chrome 21, FireFox 13, IE 10, Opera 12 and Safari 5.1 or higher. After that we worked on the files the user selected one by one since in HTML5, user can select multiple files in one file input box. 1: var files = $("#upload_files")[0].files; 2: for (var i = 0; i < files.length; i++) { 3: var file = files[i]; 4: var fileSize = file.size; 5: var fileName = file.name; 6: } Next, we calculated the start index and end index for each chunks based on the size the user specified from the browser. We put them into an array with the file name and the index, which will be used when we upload chunks into Windows Azure Blob Storage as blocks since we need to specify the target blob name and the block index. At the same time we will store the list of all indexes into another variant which will be used to commit blocks into blob in Azure Storage once all chunks had been uploaded successfully. 1: $("#upload_button_blob").click(function () { 2: // assert the browser support html5 3: ... ... 4: // start to upload each files in chunks 5: var files = $("#upload_files")[0].files; 6: for (var i = 0; i < files.length; i++) { 7: var file = files[i]; 8: var fileSize = file.size; 9: var fileName = file.name; 10: 11: // calculate the start and end byte index for each blocks(chunks) 12: // with the index, file name and index list for future using 13: var blockSizeInKB = $("#block_size").val(); 14: var blockSize = blockSizeInKB * 1024; 15: var blocks = []; 16: var offset = 0; 17: var index = 0; 18: var list = ""; 19: while (offset < fileSize) { 20: var start = offset; 21: var end = Math.min(offset + blockSize, fileSize); 22: 23: blocks.push({ 24: name: fileName, 25: index: index, 26: start: start, 27: end: end 28: }); 29: list += index + ","; 30: 31: offset = end; 32: index++; 33: } 34: } 35: }); Now we have all chunks’ information ready. The next step should be upload them one by one to the server side, and at the server side when received a chunk it will upload as a block into Blob Storage, and finally commit them with the index list through BlockBlobClient.PutBlockList. But since all these invokes are ajax calling, which means not synchronized call. So we need to introduce a new JavaScript library to help us coordinate the asynchronize operation, which named “async.js”. You can download this JavaScript library here, and you can find the document here. I will not explain this library too much in this post. We will put all procedures we want to execute as a function array, and pass into the proper function defined in async.js to let it help us to control the execution sequence, in series or in parallel. Hence we will define an array and put the function for chunk upload into this array. 1: $("#upload_button_blob").click(function () { 2: // assert the browser support html5 3: ... ... 4: 5: // start to upload each files in chunks 6: var files = $("#upload_files")[0].files; 7: for (var i = 0; i < files.length; i++) { 8: var file = files[i]; 9: var fileSize = file.size; 10: var fileName = file.name; 11: // calculate the start and end byte index for each blocks(chunks) 12: // with the index, file name and index list for future using 13: ... ... 14: 15: // define the function array and push all chunk upload operation into this array 16: blocks.forEach(function (block) { 17: putBlocks.push(function (callback) { 18: }); 19: }); 20: } 21: }); 22: }); As you can see, I used File.slice method to read each chunks based on the start and end byte index we calculated previously, and constructed a temporary HTML form with the file name, chunk index and chunk data through another new feature in HTML5 named FormData. Then post this form to the backend server through jQuery.ajax. This is the key part of our solution. 1: $("#upload_button_blob").click(function () { 2: // assert the browser support html5 3: ... ... 4: // start to upload each files in chunks 5: var files = $("#upload_files")[0].files; 6: for (var i = 0; i < files.length; i++) { 7: var file = files[i]; 8: var fileSize = file.size; 9: var fileName = file.name; 10: // calculate the start and end byte index for each blocks(chunks) 11: // with the index, file name and index list for future using 12: ... ... 13: // define the function array and push all chunk upload operation into this array 14: blocks.forEach(function (block) { 15: putBlocks.push(function (callback) { 16: // load blob based on the start and end index for each chunks 17: var blob = file.slice(block.start, block.end); 18: // put the file name, index and blob into a temporary from 19: var fd = new FormData(); 20: fd.append("name", block.name); 21: fd.append("index", block.index); 22: fd.append("file", blob); 23: // post the form to backend service (asp.net mvc controller action) 24: $.ajax({ 25: url: "/Home/UploadInFormData", 26: data: fd, 27: processData: false, 28: contentType: "multipart/form-data", 29: type: "POST", 30: success: function (result) { 31: if (!result.success) { 32: alert(result.error); 33: } 34: callback(null, block.index); 35: } 36: }); 37: }); 38: }); 39: } 40: }); Then we will invoke these functions one by one by using the async.js. And once all functions had been executed successfully I invoked another ajax call to the backend service to commit all these chunks (blocks) as the blob in Windows Azure Storage. 1: $("#upload_button_blob").click(function () { 2: // assert the browser support html5 3: ... ... 4: // start to upload each files in chunks 5: var files = $("#upload_files")[0].files; 6: for (var i = 0; i < files.length; i++) { 7: var file = files[i]; 8: var fileSize = file.size; 9: var fileName = file.name; 10: // calculate the start and end byte index for each blocks(chunks) 11: // with the index, file name and index list for future using 12: ... ... 13: // define the function array and push all chunk upload operation into this array 14: ... ... 15: // invoke the functions one by one 16: // then invoke the commit ajax call to put blocks into blob in azure storage 17: async.series(putBlocks, function (error, result) { 18: var data = { 19: name: fileName, 20: list: list 21: }; 22: $.post("/Home/Commit", data, function (result) { 23: if (!result.success) { 24: alert(result.error); 25: } 26: else { 27: alert("done!"); 28: } 29: }); 30: }); 31: } 32: }); That’s all in the client side. The outline of our logic would be - Calculate the start and end byte index for each chunks based on the block size. - Defined the functions of reading the chunk form file and upload the content to the backend service through ajax. - Execute the functions defined in previous step with “async.js”. - Commit the chunks by invoking the backend service in Windows Azure Storage finally. Save Chunks as Blocks into Blob Storage In above we finished the client size JavaScript code. It uploaded the file in chunks to the backend service which we are going to implement in this step. We will use ASP.NET MVC as our backend service, and it will receive the chunks, upload into Windows Azure Bob Storage in blocks, then finally commit as one blob. As in the client side we uploaded chunks by invoking the ajax call to the URL "/Home/UploadInFormData", I created a new action under the Index controller and it only accepts HTTP POST request. 1: [HttpPost] 2: public JsonResult UploadInFormData() 3: { 4: var error = string.Empty; 5: try 6: { 7: } 8: catch (Exception e) 9: { 10: error = e.ToString(); 11: } 12: 13: return new JsonResult() 14: { 15: Data = new 16: { 17: success = string.IsNullOrWhiteSpace(error), 18: error = error 19: } 20: }; 21: } Then I retrieved the file name, index and the chunk content from the Request.Form object, which was passed from our client side. And then, used the Windows Azure SDK to create a blob container (in this case we will use the container named “test”.) and create a blob reference with the blob name (same as the file name). Then uploaded the chunk as a block of this blob with the index, since in Blob Storage each block must have an index (ID) associated with so that finally we can put all blocks as one blob by specifying their block ID list. 1: [HttpPost] 2: public JsonResult UploadInFormData() 3: { 4: var error = string.Empty; 5: try 6: { 7: var name = Request.Form["name"]; 8: var index = int.Parse(Request.Form["index"]); 9: var file = Request.Files[0]; 10: var id = Convert.ToBase64String(BitConverter.GetBytes(index)); 11: 12: var container = _client.GetContainerReference("test"); 13: container.CreateIfNotExists(); 14: var blob = container.GetBlockBlobReference(name); 15: blob.PutBlock(id, file.InputStream, null); 16: } 17: catch (Exception e) 18: { 19: error = e.ToString(); 20: } 21: 22: return new JsonResult() 23: { 24: Data = new 25: { 26: success = string.IsNullOrWhiteSpace(error), 27: error = error 28: } 29: }; 30: } Next, I created another action to commit the blocks into blob once all chunks had been uploaded. Similarly, I retrieved the blob name from the Request.Form. I also retrieved the chunks ID list, which is the block ID list from the Request.Form in a string format, split them as a list, then invoked the BlockBlob.PutBlockList method. After that our blob will be shown in the container and ready to be download. 1: [HttpPost] 2: public JsonResult Commit() 3: { 4: var error = string.Empty; 5: try 6: { 7: var name = Request.Form["name"]; 8: var list = Request.Form["list"]; 9: var ids = list 10: .Split(',') 11: .Where(id => !string.IsNullOrWhiteSpace(id)) 12: .Select(id => Convert.ToBase64String(BitConverter.GetBytes(int.Parse(id)))) 13: .ToArray(); 14: 15: var container = _client.GetContainerReference("test"); 16: container.CreateIfNotExists(); 17: var blob = container.GetBlockBlobReference(name); 18: blob.PutBlockList(ids); 19: } 20: catch (Exception e) 21: { 22: error = e.ToString(); 23: } 24: 25: return new JsonResult() 26: { 27: Data = new 28: { 29: success = string.IsNullOrWhiteSpace(error), 30: error = error 31: } 32: }; 33: } Now we finished all code we need. The whole process of uploading would be like this below. Below is the full client side JavaScript code. 1: <script type="text/javascript" src="~/Scripts/async.js"></script> 2: <script type="text/javascript"> 3: $(function () { 4: $("#upload_button_blob").click(function () { 5: // assert the browser support html5 6: if (window.File && window.Blob && window.FormData) { 7: alert("Your brwoser is awesome, let's rock!"); 8: } 9: else { 10: alert("Oh man plz update to a modern browser before try is cool stuff out."); 11: return; 12: } 13: 14: // start to upload each files in chunks 15: var files = $("#upload_files")[0].files; 16: for (var i = 0; i < files.length; i++) { 17: var file = files[i]; 18: var fileSize = file.size; 19: var fileName = file.name; 20: 21: // calculate the start and end byte index for each blocks(chunks) 22: // with the index, file name and index list for future using 23: var blockSizeInKB = $("#block_size").val(); 24: var blockSize = blockSizeInKB * 1024; 25: var blocks = []; 26: var offset = 0; 27: var index = 0; 28: var list = ""; 29: while (offset < fileSize) { 30: var start = offset; 31: var end = Math.min(offset + blockSize, fileSize); 32: 33: blocks.push({ 34: name: fileName, 35: index: index, 36: start: start, 37: end: end 38: }); 39: list += index + ","; 40: 41: offset = end; 42: index++; 43: } 44: 45: // define the function array and push all chunk upload operation into this array 46: var putBlocks = []; 47: blocks.forEach(function (block) { 48: putBlocks.push(function (callback) { 49: // load blob based on the start and end index for each chunks 50: var blob = file.slice(block.start, block.end); 51: // put the file name, index and blob into a temporary from 52: var fd = new FormData(); 53: fd.append("name", block.name); 54: fd.append("index", block.index); 55: fd.append("file", blob); 56: // post the form to backend service (asp.net mvc controller action) 57: $.ajax({ 58: url: "/Home/UploadInFormData", 59: data: fd, 60: processData: false, 61: contentType: "multipart/form-data", 62: type: "POST", 63: success: function (result) { 64: if (!result.success) { 65: alert(result.error); 66: } 67: callback(null, block.index); 68: } 69: }); 70: }); 71: }); 72: 73: // invoke the functions one by one 74: // then invoke the commit ajax call to put blocks into blob in azure storage 75: async.series(putBlocks, function (error, result) { 76: var data = { 77: name: fileName, 78: list: list 79: }; 80: $.post("/Home/Commit", data, function (result) { 81: if (!result.success) { 82: alert(result.error); 83: } 84: else { 85: alert("done!"); 86: } 87: }); 88: }); 89: } 90: }); 91: }); 92: </script> And below is the full ASP.NET MVC controller code. 1: public class HomeController : Controller 2: { 3: private CloudStorageAccount _account; 4: private CloudBlobClient _client; 5: 6: public HomeController() 7: : base() 8: { 9: _account = CloudStorageAccount.Parse(CloudConfigurationManager.GetSetting("DataConnectionString")); 10: _client = _account.CreateCloudBlobClient(); 11: } 12: 13: public ActionResult Index() 14: { 15: ViewBag.Message = "Modify this template to jump-start your ASP.NET MVC application."; 16: 17: return View(); 18: } 19: 20: [HttpPost] 21: public JsonResult UploadInFormData() 22: { 23: var error = string.Empty; 24: try 25: { 26: var name = Request.Form["name"]; 27: var index = int.Parse(Request.Form["index"]); 28: var file = Request.Files[0]; 29: var id = Convert.ToBase64String(BitConverter.GetBytes(index)); 30: 31: var container = _client.GetContainerReference("test"); 32: container.CreateIfNotExists(); 33: var blob = container.GetBlockBlobReference(name); 34: blob.PutBlock(id, file.InputStream, null); 35: } 36: catch (Exception e) 37: { 38: error = e.ToString(); 39: } 40: 41: return new JsonResult() 42: { 43: Data = new 44: { 45: success = string.IsNullOrWhiteSpace(error), 46: error = error 47: } 48: }; 49: } 50: 51: [HttpPost] 52: public JsonResult Commit() 53: { 54: var error = string.Empty; 55: try 56: { 57: var name = Request.Form["name"]; 58: var list = Request.Form["list"]; 59: var ids = list 60: .Split(',') 61: .Where(id => !string.IsNullOrWhiteSpace(id)) 62: .Select(id => Convert.ToBase64String(BitConverter.GetBytes(int.Parse(id)))) 63: .ToArray(); 64: 65: var container = _client.GetContainerReference("test"); 66: container.CreateIfNotExists(); 67: var blob = container.GetBlockBlobReference(name); 68: blob.PutBlockList(ids); 69: } 70: catch (Exception e) 71: { 72: error = e.ToString(); 73: } 74: 75: return new JsonResult() 76: { 77: Data = new 78: { 79: success = string.IsNullOrWhiteSpace(error), 80: error = error 81: } 82: }; 83: } 84: } And if we selected a file from the browser we will see our application will upload chunks in the size we specified to the server through ajax call in background, and then commit all chunks in one blob. Then we can find the blob in our Windows Azure Blob Storage. Optimized by Parallel Upload In previous example we just uploaded our file in chunks. This solved the problem that ASP.NET MVC request content size limitation as well as the Windows Azure load balancer timeout. But it might introduce the performance problem since we uploaded chunks in sequence. In order to improve the upload performance we could modify our client side code a bit to make the upload operation invoked in parallel. The good news is that, “async.js” library provides the parallel execution function. If you remembered the code we invoke the service to upload chunks, it utilized “async.series” which means all functions will be executed in sequence. Now we will change this code to “async.parallel”. This will invoke all functions in parallel. 1: $("#upload_button_blob").click(function () { 2: // assert the browser support html5 3: ... ... 4: // start to upload each files in chunks 5: var files = $("#upload_files")[0].files; 6: for (var i = 0; i < files.length; i++) { 7: var file = files[i]; 8: var fileSize = file.size; 9: var fileName = file.name; 10: // calculate the start and end byte index for each blocks(chunks) 11: // with the index, file name and index list for future using 12: ... ... 13: // define the function array and push all chunk upload operation into this array 14: ... ... 15: // invoke the functions one by one 16: // then invoke the commit ajax call to put blocks into blob in azure storage 17: async.parallel(putBlocks, function (error, result) { 18: var data = { 19: name: fileName, 20: list: list 21: }; 22: $.post("/Home/Commit", data, function (result) { 23: if (!result.success) { 24: alert(result.error); 25: } 26: else { 27: alert("done!"); 28: } 29: }); 30: }); 31: } 32: }); In this way all chunks will be uploaded to the server side at the same time to maximize the bandwidth usage. This should work if the file was not very large and the chunk size was not very small. But for large file this might introduce another problem that too many ajax calls are sent to the server at the same time. So the best solution should be, upload the chunks in parallel with maximum concurrency limitation. The code below specified the concurrency limitation to 4, which means at the most only 4 ajax calls could be invoked at the same time. 1: $("#upload_button_blob").click(function () { 2: // assert the browser support html5 3: ... ... 4: // start to upload each files in chunks 5: var files = $("#upload_files")[0].files; 6: for (var i = 0; i < files.length; i++) { 7: var file = files[i]; 8: var fileSize = file.size; 9: var fileName = file.name; 10: // calculate the start and end byte index for each blocks(chunks) 11: // with the index, file name and index list for future using 12: ... ... 13: // define the function array and push all chunk upload operation into this array 14: ... ... 15: // invoke the functions one by one 16: // then invoke the commit ajax call to put blocks into blob in azure storage 17: async.parallelLimit(putBlocks, 4, function (error, result) { 18: var data = { 19: name: fileName, 20: list: list 21: }; 22: $.post("/Home/Commit", data, function (result) { 23: if (!result.success) { 24: alert(result.error); 25: } 26: else { 27: alert("done!"); 28: } 29: }); 30: }); 31: } 32: }); Summary In this post we discussed how to upload files in chunks to the backend service and then upload them into Windows Azure Blob Storage in blocks. We focused on the frontend side and leverage three new feature introduced in HTML 5 which are - File.slice: Read part of the file by specifying the start and end byte index. - Blob: File-like interface which contains the part of the file content. - FormData: Temporary form element that we can pass the chunk alone with some metadata to the backend service. Then we discussed the performance consideration of chunk uploading. Sequence upload cannot provide maximized upload speed, but the unlimited parallel upload might crash the browser and server if too many chunks. So we finally came up with the solution to upload chunks in parallel with the concurrency limitation. We also demonstrated how to utilize “async.js” JavaScript library to help us control the asynchronize call and the parallel limitation. Regarding the chunk size and the parallel limitation value there is no “best” value. You need to test vary composition and find out the best one for your particular scenario. It depends on the local bandwidth, client machine cores and the server side (Windows Azure Cloud Service Virtual Machine) cores, memory and bandwidth. Below is one of my performance test result. The client machine was Windows 8 IE 10 with 4 cores. I was using Microsoft Cooperation Network. The web site was hosted on Windows Azure China North data center (in Beijing) with one small web role (1.7GB 1 core CPU, 1.75GB memory with 100Mbps bandwidth). The test cases were - Chunk size: 512KB, 1MB, 2MB, 4MB. - Upload Mode: Sequence, parallel (unlimited), parallel with limit (4 threads, 8 threads). - Chunk Format: base64 string, binaries. - Target file: 100MB. - Each case was tested 3 times. Below is the test result chart. Some thoughts, but not guidance or best practice: - Parallel gets better performance than series. - No significant performance improvement between parallel 4 threads and 8 threads. - Transform with binaries provides better performance than base64. - In all cases, chunk size in 1MB - 2MB gets better performance. Hope this helps, Shaun All documents and related graphics, codes are provided "AS IS" without warranty of any kind. Copyright © Shaun Ziyan Xu. This work is licensed under the Creative Commons License.

Read the article

Design by Contract with Microsoft .Net Code Contract

- by Fredrik N

I have done some talks on different events and summits about Defensive Programming and Design by Contract, last time was at Cornerstone’s Developer Summit 2010. Next time will be at SweNug (Sweden .Net User Group). I decided to write a blog post about of some stuffs I was talking about. Users are a terrible thing! Protect your self from them ”Human users have a gift for doing the worst possible thing at the worst possible time.” – Michael T. Nygard, Release It! The kind of users Michael T. Nygard are talking about is the users of a system. We also have users that uses our code, the users I’m going to focus on is the users of our code. Me and you and another developers. “Any fool can write code that a computer can understand. Good programmers write code that humans can understand.” – Martin Fowler Good programmers also writes code that humans know how to use, good programmers also make sure software behave in a predictable manner despise inputs or user actions. Design by Contract Design by Contract (DbC) is a way for us to make a contract between us (the code writer) and the users of our code. It’s about “If you give me this, I promise to give you this”. It’s not about business validations, that is something completely different that should be part of the domain model. DbC is to make sure the users of our code uses it in a correct way, and that we can rely on the contract and write code in a way where we know that the users will follow the contract. It will make it much easier for us to write code with a contract specified. Something like the following code is something we may see often: public void DoSomething(Object value) { value.DoIKnowThatICanDoThis(); } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } Where “value” can be uses directly or passed to other methods and later be used. What some of us can easily forget here is that the “value” can be “null”. We will probably not passing a null value, but someone else that uses our code maybe will do it. I think most of you (including me) have passed “null” into a method because you don’t know if the argument need to be specified to a valid value etc. I bet most of you also have got the “Null reference exception”. Sometimes this “Null reference exception” can be hard and take time to fix, because we need to search among our code to see where the “null” value was passed in etc. Wouldn’t it be much better if we can as early as possible specify that the value can’t not be null, so the users of our code also know it when the users starts to use our code, and before run time execution of the code? This is where DbC comes into the picture. We can use DbC to specify what we need, and by doing so we can rely on the contract when we write our code. So the code above can actually use the DoIKnowThatICanDoThis() method on the value object without being worried that the “value” can be null. The contract between the users of the code and us writing the code, says that the “value” can’t be null. Pre- and Postconditions When working with DbC we are specifying pre- and postconditions. Precondition is a condition that should be met before a query or command is executed. An example of a precondition is: “The Value argument of the method can’t be null”, and we make sure the “value” isn’t null before the method is called. Postcondition is a condition that should be met when a command or query is completed, a postcondition will make sure the result is correct. An example of a postconditon is “The method will return a list with at least 1 item”. Commands an Quires When using DbC, we need to know what a Command and a Query is, because some principles that can be good to follow are based on commands and queries. A Command is something that will not return anything, like the SQL’s CREATE, UPDATE and DELETE. There are two kinds of Commands when using DbC, the Creation commands (for example a Constructor), and Others. Others can for example be a Command to add a value to a list, remove or update a value etc. //Creation commands public Stack(int size) //Other commands public void Push(object value); public void Remove(); .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } A Query, is something that will return something, for example an Attribute, Property or a Function, like the SQL’s SELECT. There are two kinds of Queries, the Basic Queries (Quires that aren’t based on another queries), and the Derived Queries, queries that is based on another queries. Here is an example of queries of a Stack: //Basic Queries public int Count; public object this[int index] { get; } //Derived Queries //Is related to Count Query public bool IsEmpty() { return Count == 0; } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } To understand about some principles that are good to follow when using DbC, we need to know about the Commands and different Queries. The 6 Principles When working with DbC, it’s advisable to follow some principles to make it easier to define and use contracts. The following DbC principles are: Separate commands and queries. Separate basic queries from derived queries. For each derived query, write a postcondition that specifies what result will be returned, in terms of one or more basic queries. For each command, write a postcondition that specifies the value of every basic query. For every query and command, decide on a suitable precondition. Write invariants to define unchanging properties of objects. Before I will write about each of them I want you to now that I’m going to use .Net 4.0 Code Contract. I will in the rest of the post uses a simple Stack (Yes I know, .Net already have a Stack class) to give you the basic understanding about using DbC. A Stack is a data structure where the first item in, will be the first item out. Here is a basic implementation of a Stack where not contract is specified yet: public class Stack { private object[] _array; //Basic Queries public uint Count; public object this[uint index] { get { return _array[index]; } set { _array[index] = value; } } //Derived Queries //Is related to Count Query public bool IsEmpty() { return Count == 0; } //Is related to Count and this[] Query public object Top() { return this[Count]; } //Creation commands public Stack(uint size) { Count = 0; _array = new object[size]; } //Other commands public void Push(object value) { this[++Count] = value; } public void Remove() { this[Count] = null; Count--; } } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } Note: The Stack is implemented in a way to demonstrate the use of Code Contract in a simple way, the implementation may not look like how you would implement it, so don’t think this is the perfect Stack implementation, only used for demonstration. Before I will go deeper into the principles I will simply mention how we can use the .Net Code Contract. I mention before about pre- and postcondition, is about “Require” something and to “Ensure” something. When using Code Contract, we will use a static class called “Contract” and is located in he “System.Diagnostics.Contracts” namespace. The contract must be specified at the top or our member statement block. To specify a precondition with Code Contract we uses the Contract.Requires method, and to specify a postcondition, we uses the Contract.Ensure method. Here is an example where both a pre- and postcondition are used: public object Top() { Contract.Requires(Count > 0, "Stack is empty"); Contract.Ensures(Contract.Result<object>() == this[Count]); return this[Count]; } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } The contract above requires that the Count is greater than 0, if not we can’t get the item at the Top of a Stack. We also Ensures that the results (By using the Contract.Result method, we can specify a postcondition that will check if the value returned from a method is correct) of the Top query is equal to this[Count]. 1. Separate Commands and Queries When working with DbC, it’s important to separate Command and Quires. A method should either be a command that performs an Action, or returning information to the caller, not both. By asking a question the answer shouldn’t be changed. The following is an example of a Command and a Query of a Stack: public void Push(object value) public object Top() .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } The Push is a command and will not return anything, just add a value to the Stack, the Top is a query to get the item at the top of the stack. 2. Separate basic queries from derived queries There are two different kinds of queries, the basic queries that doesn’t rely on another queries, and derived queries that uses a basic query. The “Separate basic queries from derived queries” principle is about about that derived queries can be specified in terms of basic queries. So this principles is more about recognizing that a query is a derived query or a basic query. It will then make is much easier to follow the other principles. The following code shows a basic query and a derived query: //Basic Queries public uint Count; //Derived Queries //Is related to Count Query public bool IsEmpty() { return Count == 0; } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } We can see that IsEmpty will use the Count query, and that makes the IsEmpty a Derived query. 3. For each derived query, write a postcondition that specifies what result will be returned, in terms of one or more basic queries. When the derived query is recognize we can follow the 3ed principle. For each derived query, we can create a postcondition that specifies what result our derived query will return in terms of one or more basic queries. Remember that DbC is about contracts between the users of the code and us writing the code. So we can’t use demand that the users will pass in a valid value, we must also ensure that we will give the users what the users wants, when the user is following our contract. The IsEmpty query of the Stack will use a Count query and that will make the IsEmpty a Derived query, so we should now write a postcondition that specified what results will be returned, in terms of using a basic query and in this case the Count query, //Basic Queries public uint Count; //Derived Queries public bool IsEmpty() { Contract.Ensures(Contract.Result<bool>() == (Count == 0)); return Count == 0; } The Contract.Ensures is used to create a postcondition. The above code will make sure that the results of the IsEmpty (by using the Contract.Result to get the result of the IsEmpty method) is correct, that will say that the IsEmpty will be either true or false based on Count is equal to 0 or not. The postcondition are using a basic query, so the IsEmpty is now following the 3ed principle. We also have another Derived Query, the Top query, it will also need a postcondition and it uses all basic queries. The Result of the Top method must be the same value as the this[] query returns. //Basic Queries public uint Count; public object this[uint index] { get { return _array[index]; } set { _array[index] = value; } } //Derived Queries //Is related to Count and this[] Query public object Top() { Contract.Ensures(Contract.Result<object>() == this[Count]); return this[Count]; } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } 4. For each command, write a postcondition that specifies the value of every basic query. For each command we will create a postconditon that specifies the value of basic queries. If we look at the Stack implementation we will have three Commands, one Creation command, the Constructor, and two others commands, Push and Remove. Those commands need a postcondition and they should include basic query to follow the 4th principle. //Creation commands public Stack(uint size) { Contract.Ensures(Count == 0); Count = 0; _array = new object[size]; } //Other commands public void Push(object value) { Contract.Ensures(Count == Contract.OldValue<uint>(Count) + 1); Contract.Ensures(this[Count] == value); this[++Count] = value; } public void Remove() { Contract.Ensures(Count == Contract.OldValue<uint>(Count) - 1); this[Count] = null; Count--; } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } As you can see the Create command will Ensures that Count will be 0 when the Stack is created, when a Stack is created there shouldn’t be any items in the stack. The Push command will take a value and put it into the Stack, when an item is pushed into the Stack, the Count need to be increased to know the number of items added to the Stack, and we must also make sure the item is really added to the Stack. The postconditon of the Push method will make sure the that old value of the Count (by using the Contract.OldValue we can get the value a Query has before the method is called) plus 1 will be equal to the Count query, this is the way we can ensure that the Push will increase the Count with one. We also make sure the this[] query will now contain the item we pushed into the Stack. The Remove method must make sure the Count is decreased by one when the top item is removed from the Stack. The Commands is now following the 4th principle, where each command now have a postcondition that used the value of basic queries. Note: The principle says every basic Query, the Remove only used one Query the Count, it’s because this command can’t use the this[] query because an item is removed, so the only way to make sure an item is removed is to just use the Count query, so the Remove will still follow the principle. 5. For every query and command, decide on a suitable precondition. We have now focused only on postcondition, now time for some preconditons. The 5th principle is about deciding a suitable preconditon for every query and command. If we starts to look at one of our basic queries (will not go through all Queries and commands here, just some of them) the this[] query, we can’t pass an index that is lower then 1 (.Net arrays and list are zero based, but not the stack in this blog post ;)) and the index can’t be lesser than the number of items in the stack. So here we will need a preconditon. public object this[uint index] { get { Contract.Requires(index >= 1); Contract.Requires(index <= Count); return _array[index]; } } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } Think about the Contract as an documentation about how to use the code in a correct way, so if the contract could be specified elsewhere (not part of the method body), we could simply write “return _array[index]” and there is no need to check if index is greater or lesser than Count, because that is specified in a “contract”. The implementation of Code Contract, requires that the contract is specified in the code. As a developer I would rather have this contract elsewhere (Like Spec#) or implemented in a way Eiffel uses it as part of the language. Now when we have looked at one Query, we can also look at one command, the Remove command (You can see the whole implementation of the Stack at the end of this blog post, where precondition is added to more queries and commands then what I’m going to show in this section). We can only Remove an item if the Count is greater than 0. So we can write a precondition that will require that Count must be greater than 0. public void Remove() { Contract.Requires(Count > 0); Contract.Ensures(Count == Contract.OldValue<uint>(Count) - 1); this[Count] = null; Count--; } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } 6. Write invariants to define unchanging properties of objects. The last principle is about making sure the object are feeling great! This is done by using invariants. When using Code Contract we can specify invariants by adding a method with the attribute ContractInvariantMethod, the method must be private or public and can only contains calls to Contract.Invariant. To make sure the Stack feels great, the Stack must have 0 or more items, the Count can’t never be a negative value to make sure each command and queries can be used of the Stack. Here is our invariant for the Stack object: [ContractInvariantMethod] private void ObjectInvariant() { Contract.Invariant(Count >= 0); } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } Note: The ObjectInvariant method will be called every time after a Query or Commands is called. Here is the full example using Code Contract: public class Stack { private object[] _array; //Basic Queries public uint Count; public object this[uint index] { get { Contract.Requires(index >= 1); Contract.Requires(index <= Count); return _array[index]; } set { Contract.Requires(index >= 1); Contract.Requires(index <= Count); _array[index] = value; } } //Derived Queries //Is related to Count Query public bool IsEmpty() { Contract.Ensures(Contract.Result<bool>() == (Count == 0)); return Count == 0; } //Is related to Count and this[] Query public object Top() { Contract.Requires(Count > 0, "Stack is empty"); Contract.Ensures(Contract.Result<object>() == this[Count]); return this[Count]; } //Creation commands public Stack(uint size) { Contract.Requires(size > 0); Contract.Ensures(Count == 0); Count = 0; _array = new object[size]; } //Other commands public void Push(object value) { Contract.Requires(value != null); Contract.Ensures(Count == Contract.OldValue<uint>(Count) + 1); Contract.Ensures(this[Count] == value); this[++Count] = value; } public void Remove() { Contract.Requires(Count > 0); Contract.Ensures(Count == Contract.OldValue<uint>(Count) - 1); this[Count] = null; Count--; } [ContractInvariantMethod] private void ObjectInvariant() { Contract.Invariant(Count >= 0); } } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } Summary By using Design By Contract we can make sure the users are using our code in a correct way, and we must also make sure the users will get the expected results when they uses our code. This can be done by specifying contracts. To make it easy to use Design By Contract, some principles may be good to follow like the separation of commands an queries. With .Net 4.0 we can use the Code Contract feature to specify contracts.

Read the article

SQL Server 2012 - AlwaysOn

- by Claus Jandausch

Ich war nicht nur irritiert, ich war sogar regelrecht schockiert - und für einen kurzen Moment sprachlos (was nur selten der Fall ist). Gerade eben hatte mich jemand gefragt "Wann Oracle denn etwas Vergleichbares wie AlwaysOn bieten würde - und ob überhaupt?" War ich hier im falschen Film gelandet? Ich konnte nicht anders, als meinen Unmut kundzutun und zu erklären, dass die Fragestellung normalerweise anders herum läuft. Zugegeben - es mag vielleicht strittige Punkte geben im Vergleich zwischen Oracle und SQL Server - bei denen nicht unbedingt immer Oracle die Nase vorn haben muss - aber das Thema Clustering für Hochverfügbarkeit (HA), Disaster Recovery (DR) und Skalierbarkeit gehört mit Sicherheit nicht dazu. Dieses Erlebnis hakte ich am Nachgang als Einzelfall ab, der so nie wieder vorkommen würde. Bis ich kurz darauf eines Besseren belehrt wurde und genau die selbe Frage erneut zu hören bekam. Diesmal sogar im Exadata-Umfeld und einem Oracle Stretch Cluster. Einmal ist keinmal, doch zweimal ist einmal zu viel... Getreu diesem alten Motto war mir klar, dass man das so nicht länger stehen lassen konnte. Ich habe keine Ahnung, wie die Microsoft Marketing Abteilung es geschafft hat, unter dem AlwaysOn Brading eine innovative Technologie vermuten zu lassen - aber sie hat ihren Job scheinbar gut gemacht. Doch abgesehen von einem guten Marketing, stellt sich natürlich die Frage, was wirklich dahinter steckt und wie sich das Ganze mit Oracle vergleichen lässt - und ob überhaupt? Damit wären wir wieder bei der ursprünglichen Frage angelangt. So viel zum Hintergrund dieses Blogbeitrags - von meiner Antwort handelt der restliche Blog. "Windows was the God ..." Um den wahren Unterschied zwischen Oracle und Microsoft verstehen zu können, muss man zunächst das bedeutendste Microsoft Dogma kennen. Es lässt sich schlicht und einfach auf den Punkt bringen: "Alles muss auf Windows basieren." Die Überschrift dieses Absatzes ist kein von mir erfundener Ausspruch, sondern ein Zitat. Konkret stammt es aus einem längeren Artikel von Kurt Eichenwald in der Vanity Fair aus dem August 2012. Er lautet Microsoft's Lost Decade und sei jedem ans Herz gelegt, der die "Microsoft-Maschinerie" unter Steve Ballmer und einige ihrer Kuriositäten besser verstehen möchte. "YOU TALKING TO ME?" Microsoft C.E.O. Steve Ballmer bei seiner Keynote auf der 2012 International Consumer Electronics Show in Las Vegas am 9. Januar Manche Dinge in diesem Artikel mögen überspitzt dargestellt erscheinen - sind sie aber nicht. Vieles davon kannte ich bereits aus eigener Erfahrung und kann es nur bestätigen. Anderes hat sich mir erst so richtig erschlossen. Insbesondere die folgenden Passagen führten zum Aha-Erlebnis: “Windows was the god—everything had to work with Windows,” said Stone... “Every little thing you want to write has to build off of Windows (or other existing roducts),” one software engineer said. “It can be very confusing, …” Ich habe immer schon darauf hingewiesen, dass in einem SQL Server Failover Cluster die Microsoft Datenbank eigentlich nichts Nenneswertes zum Geschehen beiträgt, sondern sich voll und ganz auf das Windows Betriebssystem verlässt. Deshalb muss man auch die Windows Server Enterprise Edition installieren, soll ein Failover Cluster für den SQL Server eingerichtet werden. Denn hier werden die Cluster Services geliefert - nicht mit dem SQL Server. Er ist nur lediglich ein weiteres Server Produkt, für das Windows in Ausfallszenarien genutzt werden kann - so wie Microsoft Exchange beispielsweise, oder Microsoft SharePoint, oder irgendein anderes Server Produkt das auf Windows gehostet wird. Auch Oracle kann damit genutzt werden. Das Stichwort lautet hier: Oracle Failsafe. Nur - warum sollte man das tun, wenn gleichzeitig eine überlegene Technologie wie die Oracle Real Application Clusters (RAC) zur Verfügung steht, die dann auch keine Windows Enterprise Edition voraussetzen, da Oracle die eigene Clusterware liefert. Welche darüber hinaus für kürzere Failover-Zeiten sorgt, da diese Cluster-Technologie Datenbank-integriert ist und sich nicht auf "Dritte" verlässt. Wenn man sich also schon keine technischen Vorteile mit einem SQL Server Failover Cluster erkauft, sondern zusätzlich noch versteckte Lizenzkosten durch die Lizenzierung der Windows Server Enterprise Edition einhandelt, warum hat Microsoft dann in den vergangenen Jahren seit SQL Server 2000 nicht ebenfalls an einer neuen und innovativen Lösung gearbeitet, die mit Oracle RAC mithalten kann? Entwickler hat Microsoft genügend? Am Geld kann es auch nicht liegen? Lesen Sie einfach noch einmal die beiden obenstehenden Zitate und sie werden den Grund verstehen. Anders lässt es sich ja auch gar nicht mehr erklären, dass AlwaysOn aus zwei unterschiedlichen Technologien besteht, die beide jedoch wiederum auf dem Windows Server Failover Clustering (WSFC) basieren. Denn daraus ergeben sich klare Nachteile - aber dazu später mehr. Um AlwaysOn zu verstehen, sollte man sich zunächst kurz in Erinnerung rufen, was Microsoft bisher an HA/DR (High Availability/Desaster Recovery) Lösungen für SQL Server zur Verfügung gestellt hat. Replikation Basiert auf logischer Replikation und Pubisher/Subscriber Architektur Transactional Replication Merge Replication Snapshot Replication Microsoft's Replikation ist vergleichbar mit Oracle GoldenGate. Oracle GoldenGate stellt jedoch die umfassendere Technologie dar und bietet High Performance. Log Shipping Microsoft's Log Shipping stellt eine einfache Technologie dar, die vergleichbar ist mit Oracle Managed Recovery in Oracle Version 7. Das Log Shipping besitzt folgende Merkmale: Transaction Log Backups werden von Primary nach Secondary/ies geschickt Einarbeitung (z.B. Restore) auf jedem Secondary individuell Optionale dritte Server Instanz (Monitor Server) für Überwachung und Alarm Log Restore Unterbrechung möglich für Read-Only Modus (Secondary) Keine Unterstützung von Automatic Failover Database Mirroring Microsoft's Database Mirroring wurde verfügbar mit SQL Server 2005, sah aus wie Oracle Data Guard in Oracle 9i, war funktional jedoch nicht so umfassend. Für ein HA/DR Paar besteht eine 1:1 Beziehung, um die produktive Datenbank (Principle DB) abzusichern. Auf der Standby Datenbank (Mirrored DB) werden alle Insert-, Update- und Delete-Operationen nachgezogen. Modi Synchron (High-Safety Modus) Asynchron (High-Performance Modus) Automatic Failover Unterstützt im High-Safety Modus (synchron) Witness Server vorausgesetzt Zur Frage der Kontinuität Es stellt sich die Frage, wie es um diesen Technologien nun im Zusammenhang mit SQL Server 2012 bestellt ist. Unter Fanfaren seinerzeit eingeführt, war Database Mirroring das erklärte Mittel der Wahl. Ich bin kein Produkt Manager bei Microsoft und kann hierzu nur meine Meinung äußern, aber zieht man den SQL AlwaysOn Team Blog heran, so sieht es nicht gut aus für das Database Mirroring - zumindest nicht langfristig. "Does AlwaysOn Availability Group replace Database Mirroring going forward?” “The short answer is we recommend that you migrate from the mirroring configuration or even mirroring and log shipping configuration to using Availability Group. Database Mirroring will still be available in the Denali release but will be phased out over subsequent releases. Log Shipping will continue to be available in future releases.” Damit wären wir endlich beim eigentlichen Thema angelangt. Was ist eine sogenannte Availability Group und was genau hat es mit der vielversprechend klingenden Bezeichnung AlwaysOn auf sich? SQL Server 2012 - AlwaysOn Zwei HA-Features verstekcne sich hinter dem “AlwaysOn”-Branding. Einmal das AlwaysOn Failover Clustering aka SQL Server Failover Cluster Instances (FCI) - zum Anderen die AlwaysOn Availability Groups. Failover Cluster Instances (FCI) Entspricht ungefähr dem Stretch Cluster Konzept von Oracle Setzt auf Windows Server Failover Clustering (WSFC) auf Bietet HA auf Instanz-Ebene AlwaysOn Availability Groups (Verfügbarkeitsgruppen) Ähnlich der Idee von Consistency Groups, wie in Storage-Level Replikations-Software von z.B. EMC SRDF Abhängigkeiten zu Windows Server Failover Clustering (WSFC) Bietet HA auf Datenbank-Ebene Hinweis: Verwechseln Sie nicht eine SQL Server Datenbank mit einer Oracle Datenbank. Und auch nicht eine Oracle Instanz mit einer SQL Server Instanz. Die gleichen Begriffe haben hier eine andere Bedeutung - nicht selten ein Grund, weshalb Oracle- und Microsoft DBAs schnell aneinander vorbei reden. Denken Sie bei einer SQL Server Datenbank eher an ein Oracle Schema, das kommt der Sache näher. So etwas wie die SQL Server Northwind Datenbank ist vergleichbar mit dem Oracle Scott Schema. Wenn Sie die genauen Unterschiede kennen möchten, finden Sie eine detaillierte Beschreibung in meinem Buch "Oracle10g Release 2 für Windows und .NET", erhältich bei Lehmanns, Amazon, etc. Windows Server Failover Clustering (WSFC) Wie man sieht, basieren beide AlwaysOn Technologien wiederum auf dem Windows Server Failover Clustering (WSFC), um einerseits Hochverfügbarkeit auf Ebene der Instanz zu gewährleisten und andererseits auf der Datenbank-Ebene. Deshalb nun eine kurze Beschreibung der WSFC. Die WSFC sind ein mit dem Windows Betriebssystem geliefertes Infrastruktur-Feature, um HA für Server Anwendungen, wie Microsoft Exchange, SharePoint, SQL Server, etc. zu bieten. So wie jeder andere Cluster, besteht ein WSFC Cluster aus einer Gruppe unabhängiger Server, die zusammenarbeiten, um die Verfügbarkeit einer Applikation oder eines Service zu erhöhen. Falls ein Cluster-Knoten oder -Service ausfällt, kann der auf diesem Knoten bisher gehostete Service automatisch oder manuell auf einen anderen im Cluster verfügbaren Knoten transferriert werden - was allgemein als Failover bekannt ist. Unter SQL Server 2012 verwenden sowohl die AlwaysOn Avalability Groups, als auch die AlwaysOn Failover Cluster Instances die WSFC als Plattformtechnologie, um Komponenten als WSFC Cluster-Ressourcen zu registrieren. Verwandte Ressourcen werden in eine Ressource Group zusammengefasst, die in Abhängigkeit zu anderen WSFC Cluster-Ressourcen gebracht werden kann. Der WSFC Cluster Service kann jetzt die Notwendigkeit zum Neustart der SQL Server Instanz erfassen oder einen automatischen Failover zu einem anderen Server-Knoten im WSFC Cluster auslösen. Failover Cluster Instances (FCI) Eine SQL Server Failover Cluster Instanz (FCI) ist eine einzelne SQL Server Instanz, die in einem Failover Cluster betrieben wird, der aus mehreren Windows Server Failover Clustering (WSFC) Knoten besteht und so HA (High Availability) auf Ebene der Instanz bietet. Unter Verwendung von Multi-Subnet FCI kann auch Remote DR (Disaster Recovery) unterstützt werden. Eine weitere Option für Remote DR besteht darin, eine unter FCI gehostete Datenbank in einer Availability Group zu betreiben. Hierzu später mehr. FCI und WSFC Basis FCI, das für lokale Hochverfügbarkeit der Instanzen genutzt wird, ähnelt der veralteten Architektur eines kalten Cluster (Aktiv-Passiv). Unter SQL Server 2008 wurde diese Technologie SQL Server 2008 Failover Clustering genannt. Sie nutzte den Windows Server Failover Cluster. In SQL Server 2012 hat Microsoft diese Basistechnologie unter der Bezeichnung AlwaysOn zusammengefasst. Es handelt sich aber nach wie vor um die klassische Aktiv-Passiv-Konfiguration. Der Ablauf im Failover-Fall ist wie folgt: Solange kein Hardware-oder System-Fehler auftritt, werden alle Dirty Pages im Buffer Cache auf Platte geschrieben Alle entsprechenden SQL Server Services (Dienste) in der Ressource Gruppe werden auf dem aktiven Knoten gestoppt Die Ownership der Ressource Gruppe wird auf einen anderen Knoten der FCI transferriert Der neue Owner (Besitzer) der Ressource Gruppe startet seine SQL Server Services (Dienste) Die Connection-Anforderungen einer Client-Applikation werden automatisch auf den neuen aktiven Knoten mit dem selben Virtuellen Network Namen (VNN) umgeleitet Abhängig vom Zeitpunkt des letzten Checkpoints, kann die Anzahl der Dirty Pages im Buffer Cache, die noch auf Platte geschrieben werden müssen, zu unvorhersehbar langen Failover-Zeiten führen. Um diese Anzahl zu drosseln, besitzt der SQL Server 2012 eine neue Fähigkeit, die Indirect Checkpoints genannt wird. Indirect Checkpoints ähnelt dem Fast-Start MTTR Target Feature der Oracle Datenbank, das bereits mit Oracle9i verfügbar war. SQL Server Multi-Subnet Clustering Ein SQL Server Multi-Subnet Failover Cluster entspricht vom Konzept her einem Oracle RAC Stretch Cluster. Doch dies ist nur auf den ersten Blick der Fall. Im Gegensatz zu RAC ist in einem lokalen SQL Server Failover Cluster jeweils nur ein Knoten aktiv für eine Datenbank. Für die Datenreplikation zwischen geografisch entfernten Sites verlässt sich Microsoft auf 3rd Party Lösungen für das Storage Mirroring. Die Verbesserung dieses Szenario mit einer SQL Server 2012 Implementierung besteht schlicht darin, dass eine VLAN-Konfiguration (Virtual Local Area Network) nun nicht mehr benötigt wird, so wie dies bisher der Fall war. Das folgende Diagramm stellt dar, wie der Ablauf mit SQL Server 2012 gehandhabt wird. In Site A und Site B wird HA jeweils durch einen lokalen Aktiv-Passiv-Cluster sichergestellt. Besondere Aufmerksamkeit muss hier der Konfiguration und dem Tuning geschenkt werden, da ansonsten völlig inakzeptable Failover-Zeiten resultieren. Dies liegt darin begründet, weil die Downtime auf Client-Seite nun nicht mehr nur von der reinen Failover-Zeit abhängt, sondern zusätzlich von der Dauer der DNS Replikation zwischen den DNS Servern. (Rufen Sie sich in Erinnerung, dass wir gerade von Multi-Subnet Clustering sprechen). Außerdem ist zu berücksichtigen, wie schnell die Clients die aktualisierten DNS Informationen abfragen. Spezielle Konfigurationen für Node Heartbeat, HostRecordTTL (Host Record Time-to-Live) und Intersite Replication Frequeny für Active Directory Sites und Services werden notwendig. Default TTL für Windows Server 2008 R2: 20 Minuten Empfohlene Einstellung: 1 Minute DNS Update Replication Frequency in Windows Umgebung: 180 Minuten Empfohlene Einstellung: 15 Minuten (minimaler Wert) Betrachtet man diese Werte, muss man feststellen, dass selbst eine optimale Konfiguration die rigiden SLAs (Service Level Agreements) heutiger geschäftskritischer Anwendungen für HA und DR nicht erfüllen kann. Denn dies impliziert eine auf der Client-Seite erlebte Failover-Zeit von insgesamt 16 Minuten. Hierzu ein Auszug aus der SQL Server 2012 Online Dokumentation: Cons: If a cross-subnet failover occurs, the client recovery time could be 15 minutes or longer, depending on your HostRecordTTL setting and the setting of your cross-site DNS/AD replication schedule. Wir sind hier an einem Punkt unserer Überlegungen angelangt, an dem sich erklärt, weshalb ich zuvor das "Windows was the God ..." Zitat verwendet habe. Die unbedingte Abhängigkeit zu Windows wird zunehmend zum Problem, da sie die Komplexität einer Microsoft-basierenden Lösung erhöht, anstelle sie zu reduzieren. Und Komplexität ist das Letzte, was sich CIOs heutzutage wünschen. Zur Ehrenrettung des SQL Server 2012 und AlwaysOn muss man sagen, dass derart lange Failover-Zeiten kein unbedingtes "Muss" darstellen, sondern ein "Kann". Doch auch ein "Kann" kann im unpassenden Moment unvorhersehbare und kostspielige Folgen haben. Die Unabsehbarkeit ist wiederum Ursache vieler an der Implementierung beteiligten Komponenten und deren Abhängigkeiten, wie beispielsweise drei Cluster-Lösungen (zwei von Microsoft, eine 3rd Party Lösung). Wie man die Sache auch dreht und wendet, kommt man an diesem Fakt also nicht vorbei - ganz unabhängig von der Dauer einer Downtime oder Failover-Zeiten. Im Gegensatz zu AlwaysOn und der hier vorgestellten Version eines Stretch-Clusters, vermeidet eine entsprechende Oracle Implementierung eine derartige Komplexität, hervorgerufen duch multiple Abhängigkeiten. Den Unterschied machen Datenbank-integrierte Mechanismen, wie Fast Application Notification (FAN) und Fast Connection Failover (FCF). Für Oracle MAA Konfigurationen (Maximum Availability Architecture) sind Inter-Site Failover-Zeiten im Bereich von Sekunden keine Seltenheit. Wenn Sie dem Link zur Oracle MAA folgen, finden Sie außerdem eine Reihe an Customer Case Studies. Auch dies ist ein wichtiges Unterscheidungsmerkmal zu AlwaysOn, denn die Oracle Technologie hat sich bereits zigfach in höchst kritischen Umgebungen bewährt. Availability Groups (Verfügbarkeitsgruppen) Die sogenannten Availability Groups (Verfügbarkeitsgruppen) sind - neben FCI - der weitere Baustein von AlwaysOn. Hinweis: Bevor wir uns näher damit beschäftigen, sollten Sie sich noch einmal ins Gedächtnis rufen, dass eine SQL Server Datenbank nicht die gleiche Bedeutung besitzt, wie eine Oracle Datenbank, sondern eher einem Oracle Schema entspricht. So etwas wie die SQL Server Northwind Datenbank ist vergleichbar mit dem Oracle Scott Schema. Eine Verfügbarkeitsgruppe setzt sich zusammen aus einem Set mehrerer Benutzer-Datenbanken, die im Falle eines Failover gemeinsam als Gruppe behandelt werden. Eine Verfügbarkeitsgruppe unterstützt ein Set an primären Datenbanken (primäres Replikat) und einem bis vier Sets von entsprechenden sekundären Datenbanken (sekundäre Replikate). Es können jedoch nicht alle SQL Server Datenbanken einer AlwaysOn Verfügbarkeitsgruppe zugeordnet werden. Der SQL Server Spezialist Michael Otey zählt in seinem SQL Server Pro Artikel folgende Anforderungen auf: Verfügbarkeitsgruppen müssen mit Benutzer-Datenbanken erstellt werden. System-Datenbanken können nicht verwendet werden Die Datenbanken müssen sich im Read-Write Modus befinden. Read-Only Datenbanken werden nicht unterstützt Die Datenbanken in einer Verfügbarkeitsgruppe müssen Multiuser Datenbanken sein Sie dürfen nicht das AUTO_CLOSE Feature verwenden Sie müssen das Full Recovery Modell nutzen und es muss ein vollständiges Backup vorhanden sein Eine gegebene Datenbank kann sich nur in einer einzigen Verfügbarkeitsgruppe befinden und diese Datenbank düerfen nicht für Database Mirroring konfiguriert sein Microsoft empfiehl außerdem, dass der Verzeichnispfad einer Datenbank auf dem primären und sekundären Server identisch sein sollte Wie man sieht, eignen sich Verfügbarkeitsgruppen nicht, um HA und DR vollständig abzubilden. Die Unterscheidung zwischen der Instanzen-Ebene (FCI) und Datenbank-Ebene (Availability Groups) ist von hoher Bedeutung. Vor kurzem wurde mir gesagt, dass man mit den Verfügbarkeitsgruppen auf Shared Storage verzichten könne und dadurch Kosten spart. So weit so gut ... Man kann natürlich eine Installation rein mit Verfügbarkeitsgruppen und ohne FCI durchführen - aber man sollte sich dann darüber bewusst sein, was man dadurch alles nicht abgesichert hat - und dies wiederum für Desaster Recovery (DR) und SLAs (Service Level Agreements) bedeutet. Kurzum, um die Kombination aus beiden AlwaysOn Produkten und der damit verbundene Komplexität kommt man wohl in der Praxis nicht herum. Availability Groups und WSFC AlwaysOn hängt von Windows Server Failover Clustering (WSFC) ab, um die aktuellen Rollen der Verfügbarkeitsreplikate einer Verfügbarkeitsgruppe zu überwachen und zu verwalten, und darüber zu entscheiden, wie ein Failover-Ereignis die Verfügbarkeitsreplikate betrifft. Das folgende Diagramm zeigt de Beziehung zwischen Verfügbarkeitsgruppen und WSFC: Der Verfügbarkeitsmodus ist eine Eigenschaft jedes Verfügbarkeitsreplikats. Synychron und Asynchron können also gemischt werden: Availability Modus (Verfügbarkeitsmodus) Asynchroner Commit-Modus Primäres replikat schließt Transaktionen ohne Warten auf Sekundäres Synchroner Commit-Modus Primäres Replikat wartet auf Commit von sekundärem Replikat Failover Typen Automatic Manual Forced (mit möglichem Datenverlust) Synchroner Commit-Modus Geplanter, manueller Failover ohne Datenverlust Automatischer Failover ohne Datenverlust Asynchroner Commit-Modus Nur Forced, manueller Failover mit möglichem Datenverlust Der SQL Server kennt keinen separaten Switchover Begriff wie in Oracle Data Guard. Für SQL Server werden alle Role Transitions als Failover bezeichnet. Tatsächlich unterstützt der SQL Server keinen Switchover für asynchrone Verbindungen. Es gibt nur die Form des Forced Failover mit möglichem Datenverlust. Eine ähnliche Fähigkeit wie der Switchover unter Oracle Data Guard ist so nicht gegeben. SQL Sever FCI mit Availability Groups (Verfügbarkeitsgruppen) Neben den Verfügbarkeitsgruppen kann eine zweite Failover-Ebene eingerichtet werden, indem SQL Server FCI (auf Shared Storage) mit WSFC implementiert wird. Ein Verfügbarkeitesreplikat kann dann auf einer Standalone Instanz gehostet werden, oder einer FCI Instanz. Zum Verständnis: Die Verfügbarkeitsgruppen selbst benötigen kein Shared Storage. Diese Kombination kann verwendet werden für lokale HA auf Ebene der Instanz und DR auf Datenbank-Ebene durch Verfügbarkeitsgruppen. Das folgende Diagramm zeigt dieses Szenario: Achtung! Hier handelt es sich nicht um ein Pendant zu Oracle RAC plus Data Guard, auch wenn das Bild diesen Eindruck vielleicht vermitteln mag - denn alle sekundären Knoten im FCI sind rein passiv. Es existiert außerdem eine weitere und ernsthafte Einschränkung: SQL Server Failover Cluster Instanzen (FCI) unterstützen nicht das automatische AlwaysOn Failover für Verfügbarkeitsgruppen. Jedes unter FCI gehostete Verfügbarkeitsreplikat kann nur für manuelles Failover konfiguriert werden. Lesbare Sekundäre Replikate Ein oder mehrere Verfügbarkeitsreplikate in einer Verfügbarkeitsgruppe können für den lesenden Zugriff konfiguriert werden, wenn sie als sekundäres Replikat laufen. Dies ähnelt Oracle Active Data Guard, jedoch gibt es Einschränkungen. Alle Abfragen gegen die sekundäre Datenbank werden automatisch auf das Snapshot Isolation Level abgebildet. Es handelt sich dabei um eine Versionierung der Rows. Microsoft versuchte hiermit die Oracle MVRC (Multi Version Read Consistency) nachzustellen. Tatsächlich muss man die SQL Server Snapshot Isolation eher mit Oracle Flashback vergleichen. Bei der Implementierung des Snapshot Isolation Levels handelt sich um ein nachträglich aufgesetztes Feature und nicht um einen inhärenten Teil des Datenbank-Kernels, wie im Falle Oracle. (Ich werde hierzu in Kürze einen weiteren Blogbeitrag verfassen, wenn ich mich mit der neuen SQL Server 2012 Core Lizenzierung beschäftige.) Für die Praxis entstehen aus der Abbildung auf das Snapshot Isolation Level ernsthafte Restriktionen, derer man sich für den Betrieb in der Praxis bereits vorab bewusst sein sollte: Sollte auf der primären Datenbank eine aktive Transaktion zu dem Zeitpunkt existieren, wenn ein lesbares sekundäres Replikat in die Verfügbarkeitsgruppe aufgenommen wird, werden die Row-Versionen auf der korrespondierenden sekundären Datenbank nicht sofort vollständig verfügbar sein. Eine aktive Transaktion auf dem primären Replikat muss zuerst abgeschlossen (Commit oder Rollback) und dieser Transaktions-Record auf dem sekundären Replikat verarbeitet werden. Bis dahin ist das Isolation Level Mapping auf der sekundären Datenbank unvollständig und Abfragen sind temporär geblockt. Microsoft sagt dazu: "This is needed to guarantee that row versions are available on the secondary replica before executing the query under snapshot isolation as all isolation levels are implicitly mapped to snapshot isolation." (SQL Storage Engine Blog: AlwaysOn: I just enabled Readable Secondary but my query is blocked?) Grundlegend bedeutet dies, dass ein aktives lesbares Replikat nicht in die Verfügbarkeitsgruppe aufgenommen werden kann, ohne das primäre Replikat vorübergehend stillzulegen. Da Leseoperationen auf das Snapshot Isolation Transaction Level abgebildet werden, kann die Bereinigung von Ghost Records auf dem primären Replikat durch Transaktionen auf einem oder mehreren sekundären Replikaten geblockt werden - z.B. durch eine lang laufende Abfrage auf dem sekundären Replikat. Diese Bereinigung wird auch blockiert, wenn die Verbindung zum sekundären Replikat abbricht oder der Datenaustausch unterbrochen wird. Auch die Log Truncation wird in diesem Zustant verhindert. Wenn dieser Zustand längere Zeit anhält, empfiehlt Microsoft das sekundäre Replikat aus der Verfügbarkeitsgruppe herauszunehmen - was ein ernsthaftes Downtime-Problem darstellt. Die Read-Only Workload auf den sekundären Replikaten kann eingehende DDL Änderungen blockieren. Obwohl die Leseoperationen aufgrund der Row-Versionierung keine Shared Locks halten, führen diese Operatioen zu Sch-S Locks (Schemastabilitätssperren). DDL-Änderungen durch Redo-Operationen können dadurch blockiert werden. Falls DDL aufgrund konkurrierender Lese-Workload blockiert wird und der Schwellenwert für 'Recovery Interval' (eine SQL Server Konfigurationsoption) überschritten wird, generiert der SQL Server das Ereignis sqlserver.lock_redo_blocked, welches Microsoft zum Kill der blockierenden Leser empfiehlt. Auf die Verfügbarkeit der Anwendung wird hierbei keinerlei Rücksicht genommen. Keine dieser Einschränkungen existiert mit Oracle Active Data Guard. Backups auf sekundären Replikaten Über die sekundären Replikate können Backups (BACKUP DATABASE via Transact-SQL) nur als copy-only Backups einer vollständigen Datenbank, Dateien und Dateigruppen erstellt werden. Das Erstellen inkrementeller Backups ist nicht unterstützt, was ein ernsthafter Rückstand ist gegenüber der Backup-Unterstützung physikalischer Standbys unter Oracle Data Guard. Hinweis: Ein möglicher Workaround via Snapshots, bleibt ein Workaround. Eine weitere Einschränkung dieses Features gegenüber Oracle Data Guard besteht darin, dass das Backup eines sekundären Replikats nicht ausgeführt werden kann, wenn es nicht mit dem primären Replikat kommunizieren kann. Darüber hinaus muss das sekundäre Replikat synchronisiert sein oder sich in der Synchronisation befinden, um das Beackup auf dem sekundären Replikat erstellen zu können. Vergleich von Microsoft AlwaysOn mit der Oracle MAA Ich komme wieder zurück auf die Eingangs erwähnte, mehrfach an mich gestellte Frage "Wann denn - und ob überhaupt - Oracle etwas Vergleichbares wie AlwaysOn bieten würde?" und meine damit verbundene (kurze) Irritation. Wenn Sie diesen Blogbeitrag bis hierher gelesen haben, dann kennen Sie jetzt meine darauf gegebene Antwort. Der eine oder andere Punkt traf dabei nicht immer auf Jeden zu, was auch nicht der tiefere Sinn und Zweck meiner Antwort war. Wenn beispielsweise kein Multi-Subnet mit im Spiel ist, sind alle diesbezüglichen Kritikpunkte zunächst obsolet. Was aber nicht bedeutet, dass sie nicht bereits morgen schon wieder zum Thema werden könnten (Sag niemals "Nie"). In manch anderes Fettnäpfchen tritt man wiederum nicht unbedingt in einer Testumgebung, sondern erst im laufenden Betrieb. Erst recht nicht dann, wenn man sich potenzieller Probleme nicht bewusst ist und keine dedizierten Tests startet. Und wer AlwaysOn erfolgreich positionieren möchte, wird auch gar kein Interesse daran haben, auf mögliche Schwachstellen und den besagten Teufel im Detail aufmerksam zu machen. Das ist keine Unterstellung - es ist nur menschlich. Außerdem ist es verständlich, dass man sich in erster Linie darauf konzentriert "was geht" und "was gut läuft", anstelle auf das "was zu Problemen führen kann" oder "nicht funktioniert". Wer will schon der Miesepeter sein? Für mich selbst gesprochen, kann ich nur sagen, dass ich lieber vorab von allen möglichen Einschränkungen wissen möchte, anstelle sie dann nach einer kurzen Zeit der heilen Welt schmerzhaft am eigenen Leib erfahren zu müssen. Ich bin davon überzeugt, dass es Ihnen nicht anders geht. Nachfolgend deshalb eine Zusammenfassung all jener Punkte, die ich im Vergleich zur Oracle MAA (Maximum Availability Architecture) als unbedingt Erwähnenswert betrachte, falls man eine Evaluierung von Microsoft AlwaysOn in Betracht zieht. 1. AlwaysOn ist eine komplexe Technologie Der SQL Server AlwaysOn Stack ist zusammengesetzt aus drei verschiedenen Technlogien: Windows Server Failover Clustering (WSFC) SQL Server Failover Cluster Instances (FCI) SQL Server Availability Groups (Verfügbarkeitsgruppen) Man kann eine derartige Lösung nicht als nahtlos bezeichnen, wofür auch die vielen von Microsoft dargestellten Einschränkungen sprechen. Während sich frühere SQL Server Versionen in Richtung eigener HA/DR Technologien entwickelten (wie Database Mirroring), empfiehlt Microsoft nun die Migration. Doch weshalb dieser Schwenk? Er führt nicht zu einem konsisten und robusten Angebot an HA/DR Technologie für geschäftskritische Umgebungen. Liegt die Antwort in meiner These begründet, nach der "Windows was the God ..." noch immer gilt und man die Nachteile der allzu engen Kopplung mit Windows nicht sehen möchte? Entscheiden Sie selbst ... 2. Failover Cluster Instanzen - Kein RAC-Pendant Die SQL Server und Windows Server Clustering Technologie basiert noch immer auf dem veralteten Aktiv-Passiv Modell und führt zu einer Verschwendung von Systemressourcen. In einer Betrachtung von lediglich zwei Knoten erschließt sich auf Anhieb noch nicht der volle Mehrwert eines Aktiv-Aktiv Clusters (wie den Real Application Clusters), wie er von Oracle bereits vor zehn Jahren entwickelt wurde. Doch kennt man die Vorzüge der Skalierbarkeit durch einfaches Hinzufügen weiterer Cluster-Knoten, die dann alle gemeinsam als ein einziges logisches System zusammenarbeiten, versteht man was hinter dem Motto "Pay-as-you-Grow" steckt. In einem Aktiv-Aktiv Cluster geht es zwar auch um Hochverfügbarkeit - und ein Failover erfolgt zudem schneller, als in einem Aktiv-Passiv Modell - aber es geht eben nicht nur darum. An dieser Stelle sei darauf hingewiesen, dass die Oracle 11g Standard Edition bereits die Nutzung von Oracle RAC bis zu vier Sockets kostenfrei beinhaltet. Möchten Sie dazu Windows nutzen, benötigen Sie keine Windows Server Enterprise Edition, da Oracle 11g die eigene Clusterware liefert. Sie kommen in den Genuss von Hochverfügbarkeit und Skalierbarkeit und können dazu die günstigere Windows Server Standard Edition nutzen. 3. SQL Server Multi-Subnet Clustering - Abhängigkeit zu 3rd Party Storage Mirroring Die SQL Server Multi-Subnet Clustering Architektur unterstützt den Aufbau eines Stretch Clusters, basiert dabei aber auf dem Aktiv-Passiv Modell. Das eigentlich Problematische ist jedoch, dass man sich zur Absicherung der Datenbank auf 3rd Party Storage Mirroring Technologie verlässt, ohne Integration zwischen dem Windows Server Failover Clustering (WSFC) und der darunterliegenden Mirroring Technologie. Wenn nun im Cluster ein Failover auf Instanzen-Ebene erfolgt, existiert keine Koordination mit einem möglichen Failover auf Ebene des Storage-Array. 4. Availability Groups (Verfügbarkeitsgruppen) - Vier, oder doch nur Zwei? Ein primäres Replikat erlaubt bis zu vier sekundäre Replikate innerhalb einer Verfügbarkeitsgruppe, jedoch nur zwei im Synchronen Commit Modus. Während dies zwar einen Vorteil gegenüber dem stringenten 1:1 Modell unter Database Mirroring darstellt, fällt der SQL Server 2012 damit immer noch weiter zurück hinter Oracle Data Guard mit bis zu 30 direkten Stanbdy Zielen - und vielen weiteren durch kaskadierende Ziele möglichen. Damit eignet sich Oracle Active Data Guard auch für die Bereitstellung einer Reader-Farm Skalierbarkeit für Internet-basierende Unternehmen. Mit AwaysOn Verfügbarkeitsgruppen ist dies nicht möglich. 5. Availability Groups (Verfügbarkeitsgruppen) - kein asynchrones Switchover Die Technologie der Verfügbarkeitsgruppen wird auch als geeignetes Mittel für administrative Aufgaben positioniert - wie Upgrades oder Wartungsarbeiten. Man muss sich jedoch einem gravierendem Defizit bewusst sein: Im asynchronen Verfügbarkeitsmodus besteht die einzige Möglichkeit für Role Transition im Forced Failover mit Datenverlust! Um den Verlust von Daten durch geplante Wartungsarbeiten zu vermeiden, muss man den synchronen Verfügbarkeitsmodus konfigurieren, was jedoch ernstzunehmende Auswirkungen auf WAN Deployments nach sich zieht. Spinnt man diesen Gedanken zu Ende, kommt man zu dem Schluss, dass die Technologie der Verfügbarkeitsgruppen für geplante Wartungsarbeiten in einem derartigen Umfeld nicht effektiv genutzt werden kann. 6. Automatisches Failover - Nicht immer möglich Sowohl die SQL Server FCI, als auch Verfügbarkeitsgruppen unterstützen automatisches Failover. Möchte man diese jedoch kombinieren, wird das Ergebnis kein automatisches Failover sein. Denn ihr Zusammentreffen im Failover-Fall führt zu Race Conditions (Wettlaufsituationen), weshalb diese Konfiguration nicht länger das automatische Failover zu einem Replikat in einer Verfügbarkeitsgruppe erlaubt. Auch hier bestätigt sich wieder die tiefere Problematik von AlwaysOn, mit einer Zusammensetzung aus unterschiedlichen Technologien und der Abhängigkeit zu Windows. 7. Problematische RTO (Recovery Time Objective) Microsoft postioniert die SQL Server Multi-Subnet Clustering Architektur als brauchbare HA/DR Architektur. Bedenkt man jedoch die Problematik im Zusammenhang mit DNS Replikation und den möglichen langen Wartezeiten auf Client-Seite von bis zu 16 Minuten, sind strenge RTO Anforderungen (Recovery Time Objectives) nicht erfüllbar. Im Gegensatz zu Oracle besitzt der SQL Server keine Datenbank-integrierten Technologien, wie Oracle Fast Application Notification (FAN) oder Oracle Fast Connection Failover (FCF). 8. Problematische RPO (Recovery Point Objective) SQL Server ermöglicht Forced Failover (erzwungenes Failover), bietet jedoch keine Möglichkeit zur automatischen Übertragung der letzten Datenbits von einem alten zu einem neuen primären Replikat, wenn der Verfügbarkeitsmodus asynchron war. Oracle Data Guard hingegen bietet diese Unterstützung durch das Flush Redo Feature. Dies sichert "Zero Data Loss" und beste RPO auch in erzwungenen Failover-Situationen. 9. Lesbare Sekundäre Replikate mit Einschränkungen Aufgrund des Snapshot Isolation Transaction Level für lesbare sekundäre Replikate, besitzen diese Einschränkungen mit Auswirkung auf die primäre Datenbank. Die Bereinigung von Ghost Records auf der primären Datenbank, wird beeinflusst von lang laufenden Abfragen auf der lesabaren sekundären Datenbank. Die lesbare sekundäre Datenbank kann nicht in die Verfügbarkeitsgruppe aufgenommen werden, wenn es aktive Transaktionen auf der primären Datenbank gibt. Zusätzlich können DLL Änderungen auf der primären Datenbank durch Abfragen auf der sekundären blockiert werden. Und imkrementelle Backups werden hier nicht unterstützt. Keine dieser Restriktionen existiert unter Oracle Data Guard.

Search Results

Search found 23664 results on 947 pages for 'quirky update'.

Page 751/947 | < Previous Page | 747 748 749 750 751 752 753 754 755 756 757 758 | Next Page >

- by Dexirian

- by Aron Rotteveel

- by Girauder

- by Jason

- by nomercysir

- by Fire Lancer

- by Rod

- by Maurycy Zarzycki

- by Maurycy Zarzycki

- by undefined

- by big_smile

- by RayQuang

- by killy9999

- by user61766

- by Nic

- by user620965

- by Sarah

- by slakr007

- by Michael S-B

- by michael

- by Will3265

- by Rick Strahl

- by Shaun

- by Fredrik N

- by Claus Jandausch

< Previous Page | 747 748 749 750 751 752 753 754 755 756 757 758 | Next Page >