Search Results

Search found 1013 results on 41 pages for 'recommendation'.

Page 27/41 | < Previous Page | 23 24 25 26 27 28 29 30 31 32 33 34  | Next Page >

  • SQLAuthority News – Free Trip on SQL Cruise

    - by pinaldave
    Everybody wants to go cruising.  I want to relax in a cruise as well, of course! (Anybody who wants to be my sponsor? Just kidding!) My family wants to go to a cruise, too. Even though I really want go to a cruise, I always wonder about one thing: what happens if I get bored on the cruise because I’d just look at the water most of the time? The best recommendation to avoid boredom on board is to travel with friends. How many friends usually accompany you when travelling? I have several good friends going on a cruise, and this is the reason why I want to go to SQL Cruise. One of them is Brent, who I consider as my friend. (Tim, you are my friend, too!) Now, we all have an opportunity to travel for free. Idera is offering a trip to SQL Cruise for FREE. To win a FREE SQL Cruise trip, you have to to do a very simple thing: just talk about How you saved the day You can tell your story via a video, photo, poem, or interpretive dance. If you refer to superheroes and Idera product, you will gain more credits to win. WHAT YOU CAN WIN: A 5-day cruise for two from Miami to Grand Cayman and Cozumel 1 seat in the SQLcruise training Airfare for two to Miami (up to $1000) Please read for further details over here. Make sure you participate and submit your entry within January 5 up to 21, 2011. Reference: Pinal Dave (http://blog.SQLAuthority.com) Filed under: PostADay, SQL, SQL Authority, SQL Query, SQL Server, SQL Tips and Tricks, T SQL, Technology

    Read the article

  • Oracle ADF at Oracle OpenWorld 2012

    - by Shay Shmeltzer
    This year is going to be very busy for Oracle ADF developers who'll attend Oracle Open World. Check out the list of Oracle ADF related sessions, labs, demos and other Oracle ADF activities.  This list will help you not to miss any ADF related activity. We have over 50 ADF related sessions, multiple labs including new ones on ADF Mobile, Application Life Cycle Management and ADF in Eclipse, we'll have several demo booths where you can meet product managers, and we'll be featured in several keynotes as well. While we have several "beginners" sessions, you'll find that we have a lot of in-depth technical sessions and sessions that cover best-practices too. Of course, it is not just us product managers presenting about Oracle ADF, there are a lot of Oracle ADF sessions presented by customers, Oracle ACEs, and other developers. So you can learn from the experience of real life implementations. Note that the ADF content starts early on Sunday with a full set of Oracle ADF sessions arranged for you by the Oracle ADF Enterprise Methodology Group - so plan your trip accordingly and be there early Sunday morning. First thing on Monday morning, don't miss the keynote for Oracle ADF developers at 10:45 at the Marriott Marquis - Salon 8 - "The Future of Development for Oracle Fusion—From Desktop to Mobile to Cloud". We are also arranging a meet-up of developers using Oracle ADF at the OTN Lounge on Wed at 4:30pm - and we would love to meet you there - this will also give you an opportunity to meet other Oracle ADF users and members of the community. And after that we can all head over to the big Wed party to see Pearl Jam and Kings of Leon. One recommendation for those who are already registered - start planning your schedule and booking your place in the sessions now through the schedule builder. This will guarantee that you won't be left out of sessions you want to attend due room size limitations. Oracle OpenWorld 2013 will be a must attend event for serious Oracle ADF developers - don't miss it.

    Read the article

  • The lifecycle of "cool"

    - by Dori
    I've been thinking lately about how some programming projects/products become "cool," and in particular, how that trend can later reverse. Here are two examples that might better explain my context: Textmate Whenever someone asks about text editors on OS X, the answer on the SE sites is an automatic "Textmate!" But looked at objectively: Textmate 1.0 shipped October 2004 Textmate 1.5 shipped January 2006 Textmate 2 was announced February 2006 As of September 2010, the currently shipping version is 1.5.9 In all of 2010, there have been a total of three posts on the Textmate blog At what point (if ever) do Textmate fans start thinking about switching to another text editor? When it breaks after some future Apple update? When alpha geeks they respect start recommending something else? Or? jQuery Whenever a JavaScript-related question is asked on the SE sites, the knee-jerk response is "jQuery!" I've seen it happen even when the question itself only required a single line of JavaScript. Or when the question could be better answered by using CSS. Do the answerers understand they're suggesting a blowtorch to light a candle? That they're recommending adding 70K or so of code to do something trivial? Or is it a symptom of "When you have a hammer, everything looks like a nail"—that is, jQuery is all they know how to do, so that's their recommendation? And do they understand that while they may know jQuery well, that doesn't necessarily mean that they know JavaScript? Is there a way to explain that learning JavaScript would make them better jQuery programmers? My bigger-picture questions: Is this niche focus primarily a trait of programmers? How do you get programmers to not immediately jump to recommending their personal favorites? What can motivate programmers to review their initial selection criteria and possibly modify their choice? Your thoughts?

    Read the article

  • Is ServerManager module available on Windows 7?

    - by ruslan
    Is ServerManager module available on Windows 7 Ultimate ? I think (but I'm not sure because I never looked at output before) the PS script that I have worked before but after some problems with IIS7 installation it stopped working. Following script import-module servermanager fails with error Import-Module : The specified module 'servermanager' was not loaded because no valid module file was found in any module directory. At line:1 char:14 + import-module <<<< servermanager + CategoryInfo : ResourceUnavailable: (servermanager:String) [Import-Module], FileNotFoundException + FullyQualifiedErrorId : Modules_ModuleNotFound,Microsoft.PowerShell.Commands.ImportModuleCommand I found recommendation to run Dism.exe /Online /Enable-Feature /FeatureName:ServerManager-PSH-Cmdlets on my machine but it also fails with error Feature name ServerManager-PSH-Cmdlets is unknown.

    Read the article

  • Monitoring Maximum Process Network Activity

    - by user125973
    We have a collocated server on which we run some OpenVZ hosts. Recently, we have had to pay a lot extra we keep exceeding our bandwidth quota. Our quota is 5 Mb/s but we have spike to almost 50. I looked at the graphs and there are some spikes happening at some intervals. I want to know which process is causing this so I need a tool that monitors the processes and gives me the one with the maximum instant traffic (It doesn't matter how much traffic we have as long as we don't exceed the 5Mb/s quota). Does anyone have a recommendation for this? My hosts are CentOS 5 with OpenVZ so I can see all the containter processes from the host, if that helps in any way.

    Read the article

  • How to restrict paddle movement using Farseer Physics engine 3.2

    - by brainydexter
    I am new to using Farseer Physics Engine 3.2(FPE), so please bear with my questions. Also, since FPE 3.2 is based on Box2D, I have been reading Box2D manual and pieces of code scattered in samples to better understand terminology and usage. Pong is usually my testbed whenever I try to do something new. Here is one of the issue I am running into: How can I restrict paddles to move only along Y axis, because the ball comes in and knocks off the paddles and everything floats in space afterwards ? (Box = Rectangle and ball = circle) I know MKS is the unit system, but is there a recommendation for sizes/position to be used ? I know this is a very generic question, but it would be good to know a simple set of values that one could use for making a game as simple as pong. Between box2d and FPE, I have some doubts: what is the recommended way of making a body in FPE ? world.CreateBody() does not exist in FPE Box2d manual recommends never to "new" body(since Box2D uses Small Object allocators), so is there a recommended way in Farseer to create a body (apart from factories) ? In box2d, it is recommended to keep a track of the body object, since it is also the parent to fixture(s). Why is it that in most of the examples, the fixture object is tracked ? Is there a reason why body is not tracked ? Thanks

    Read the article

  • Suggestion for auditd set-up

    - by JVerstry
    Hi, I am trying to learn about securing a Linux box (I am using Ubuntu). Auditd is recommended for monitoring activities on the node. I have managed to install it, but I can't find much information about proper set-up to secure my node. How should I set-up auditd to make my node more secure? What should I monitor? Why? I am looking for set-up examples and recommendation from experienced administrators. Thanks!

    Read the article

  • 1and1: Unable to host an external domain

    - by Django Reinhardt
    I'm sorry if this isn't the right place for this question, but I'm presently having difficulties with my hosting provider (1and1). Two weeks ago, two of my clients bought hosting from them on my recommendation, but as it turned out, 1and1 are having severe technical difficulties. Right now non of their hosting packages are able to accept ANY external domains. So either you pay the costs of transferring the registrar of your domain, or you use the ugly 1and1 domain name. Not any good for a hosting company of 1and1's reputation! They have been promising me for two weeks that they're going to fix the problem, but as you have probably guessed by now, that hasn't been the case. I would like to know if a) Anyone else is in the same boat as me, and b) If there are other comparably reputable hosting providers that I should consider moving to instead? Very disappointing! :( Note: This is for 1and1 in the UK. I imagine it isn't affecting users in other countries(?) Clarification: 1and1 are unable to accept ANY external domains. That means that even if you update your DNS details on your domain, their system cannot be updated to add your external domain to your account.

    Read the article

  • A minimal Linux distribution for my ASUS EEE PC

    - by Andrioid
    I recently bought myself a ASUS EEE 1000HE and I intend to use it for note-taking and light browsing at the University. The machine has a 10" screen so the interface needs to be very compact. I've already tried: EEEbuntu: Very nice driver support and out of the box experience. But I feel that it is too slow booting and the general experience is too heavy in my opinion. Moblin 2: Looks very cool, boots just fine but is way too unstable to use. Also find it annoying that I can't find hotkey documentation anywhere. Any Netbook OS recommendation welcome (although those specific to my model would be great). There is an entire jungle of distributions out there, so if you've been on a safari, please share your experience.

    Read the article

  • What are the advantages of Maven when it comes to single man, educational projects

    - by Leron
    I've spend a few hours playing around with Maven + reading some stuff on the apache official site and also a few random googled articles. By this I mean that I really tried to find the answers myself - both by reading and by doing things on my own. Also maybe worth to mention that I installed the m2e plugin so most of the time I've tried things out from Eclipse and not using the command line too much. However aside from the generated project that for example prevent me from using the default package I didn't see that much of a difference with the standard way I've created my projects before try Maven. In fact I've almost decided to skip Maven for now and move on to the other technology I wanted to learn more in-depth - Hibernate, but when I start with opening the official page the first thing I've read was the recommendation to use Hibernate with Maven. That get me confused and made me taking a step back and trying once more to find what I'm obviously missing right now. As it's said in the maven.apache.. site, the true strength of Maven is shown when you work on large projects with other people, but I lack the option to see how Maven is really used in this scenario, still i think that there are maybe advantages even when it comes to working with small projects alone, but I really have difficulties to point them out. So what do you think are the advantages of Maven when it's used for small projects writing from a single person. What are the things that I should be aware of and try to exploit (I mean features offered by Maven) that can come in handy in this situations?

    Read the article

  • Which algorithms/data structures should I "recognize" and know by name?

    - by Earlz
    I'd like to consider myself a fairly experienced programmer. I've been programming for over 5 years now. My weak point though is terminology. I'm self-taught, so while I know how to program, I don't know some of the more formal aspects of computer science. So, what are practical algorithms/data structures that I could recognize and know by name? Note, I'm not asking for a book recommendation about implementing algorithms. I don't care about implementing them, I just want to be able to recognize when an algorithm/data structure would be a good solution to a problem. I'm asking more for a list of algorithms/data structures that I should "recognize". For instance, I know the solution to a problem like this: You manage a set of lockers labeled 0-999. People come to you to rent the locker and then come back to return the locker key. How would you build a piece of software to manage knowing which lockers are free and which are in used? The solution, would be a queue or stack. What I'm looking for are things like "in what situation should a B-Tree be used -- What search algorithm should be used here" etc. And maybe a quick introduction of how the more complex(but commonly used) data structures/algorithms work. I tried looking at Wikipedia's list of data structures and algorithms but I think that's a bit overkill. So I'm looking more for what are the essential things I should recognize?

    Read the article

  • CLR 4.0: Corrupted State Exceptions

    - by Scott Dorman
    Corrupted state exceptions are designed to help you have fewer bugs in your code by making it harder to make common mistakes around exception handling. A very common pattern is code like this: public void FileSave(String name) { try { FileStream fs = new FileStream(name, FileMode.Create); } catch (Exception e) { MessageBox.Show("File Open Error"); throw new Exception(IOException); } The standard recommendation is not to catch System.Exception but rather catch the more specific exceptions (in this case, IOException). While this is a somewhat contrived example, what would happen if Exception were really an AccessViolationException or some other exception indicating that the process state has been corrupted? What you really want to do is get out fast before persistent data is corrupted or more work is lost. To help solve this problem and minimize the chance that you will catch exceptions like this, CLR 4.0 introduces Corrupted State Exceptions, which cannot be caught by normal catch statements. There are still places where you do want to catch these types of exceptions, particularly in your application’s “main” function or when you are loading add-ins.  There are also rare circumstances when you know code that throws an exception isn’t dangerous, such as when calling native code. In order to support these scenarios, a new HandleProcessCorruptedStateExceptions attribute has been added. This attribute is added to the function that catches these exceptions. There is also a process wide compatibility switch named legacyCorruptedStateExceptionsPolicy which when set to true will cause the code to operate under the older exception handling behavior. Technorati Tags: CLR 4.0, .NET 4.0, Exception Handling, Corrupted State Exceptions

    Read the article

  • I'm tasked with leading the documentation effort for an existing, entirely undocumented, software product - what resources are there to help me?

    - by Ben Rose
    I'm a software developer at a technology company. I have been tasked with leading the documentation effort for the product I work on. The goal is to produce documentation internal to developer, and the project spills over into the business side, where it covers requirements documentation. This project is challenging. Specifically, I'm dealing with a product which: - has been around for a long time, at least 6 years. - has no form of documentation other than some small, outdated pieces here and there. - has comments in the code, but they are technical and do not convey any over-arching behavior (even on technical side). - as a consequence of having little to no documentation, is often unnecessarily complex under the covers In addition, we have not been given a lot of time to work on this project. I do not have any formal documentation or writing background, training, or experience. I have displayed some ability in writing/communication around the office, which may be why I was assigned to this project. Please share your advice or recommendation for resources to help me prepare and deal with this project. I'm looking for references to books/website/forums/whatever, to help me come up with the design of a plan with milestones, learn about best practices, task delegation, templates, buy-in, etc. I'm hoping specifically for resources targeting or giving special mention of introducing good documentation to existing, undocumented, projects. I would be very grateful for your responses. Ben

    Read the article

  • Understanding branching strategy/workflow correctly

    - by burnersk
    I'm using svn without branches (trunk-only) for a very long time at my workplace. I had discovered most or all of the issues related to projects which do not have any branching strategy. Unlikely this is not going to change at my workplace but for my private projects. For my private projects which most includes coworkers and working together at the same time on different features I like to have an robust branching strategy with supports long-term releases powered by git. I find out that the Atlassian Toolchain (JIRA, Stash and Bamboo) helped me most and it also recommending me an branching strategy which I like to verify for the team needs. The branching strategy was taken directly from Atlassian Stash recommendation with a small modification to the hotfix branch tree. All hotfixes should also merged into mainline. The branching strategy in words mainline (also known as master with git or trunk with svn) contains the "state of the art" developing release. Everything here was successfully checked with various automated tests (through Bamboo) and looks like everything is working. It is not proven as working because of possible missing tests. It is ready to use but not recommended for production. feature covers all new features which are not completely finished. Once a feature is finished it will be merged into mainline. Sample branch: feature/ISSUE-2-A-nice-Feature bugfix fixes non-critical bugs which can wait for the next normal release. Sample branch: bugfix/ISSUE-1-Some-typos production owns the latest release. hotfix fixes critical bugs which have to be release urgent to mainline, production and all affected long-term *release*es. Sample branch: hotfix/ISSUE-3-Check-your-math release is for long-term maintenance. Sample branches: release/1.0, release/1.1 release/1.0-rc1 I am not an expert so please provide me feedback. Which problems might appear? Which parts are missing or slowing down the productivity?

    Read the article

  • Need a PCIe desktop graphics card for dual-monitor

    - by Graham
    I have a mid-2008 workstation with two HD monitors supporting HDMI and DVI inputs. Since Ubuntu 11.10, I have experienced no end of trouble with my NVidia Quadro NVS 290 in TwinView dual-monitor output. Others have similar desktop TwinView woes. I want a new graphics card. Previously I asked for a graphics card recommendation and response was Nvidia Geforce GTS 450... but really I'm looking for someone who has actually got a working dual-monitor desktop to tell me what card they use so I can get something that is known to work. So please, people who have no-issues with their 64-bit Ubuntu 12.04 Unity 3D desktop spread across two HD-resolution external monitors (either DVI or HDMI connector), and who also run Google Chrome (which throws a spanner due to its own GPU compositing)... please let me know what graphics card you have so I can buy one. Gathering Options These seem to be the Nvidia cards featuring dual DVI. But they all seem to be gaming cards - what has dual-DVI, good support, but is not a massive gaming card? Nvidia GTS 450 (previously recommended) - 2x DVI Nvidia GTX 550 Ti (used by System76) - 2x DVI Nvidia GT 430 (used by System76) - 1xDVI, 1xHDMI Nvidia GT 640 (found on NVidia site) - 1xDVI, 1xHDMI (also GT 620, GT 630) Has anyone had a good desktop dual-monitor Unity 3D experience with ATI cards?

    Read the article

  • use of tcp_delack_min on redhat linux (kernel 2.6.18)

    - by user41466
    Hello, we're moving from Solaris to Redhat Linux, and trying to duplicate our low-latency setup, that, on solaris, includes the ndd settings related to TCP NO DELAY, and NAGLE ALGORITHM. I got the impression that those parameters are not all configurable system-wide, but still found some info. we have configured our applications to run with no nagle algorithm, but that is not sufficient. we have found an interesting RH article talking presenting the tcp_delack_min parameter, however, when browsing /proc/sys/net/ipv4/ , I can't find it there. would it be safe to assume that simply "adding" the parameter as it's said on the doc would be enough, or rather that the option is not supported by this version (would be strange, as RH specify that it "can be performed on a standard Red Hat Enterprise Linux installation") ? any other idea / recommendation to improve latency further ? thanks

    Read the article

  • What web technology could I use which would support a decision tree?

    - by Rami Alhamad
    I am a big game development fan but I haven't done any commercial work in the past. I have been asked by a non-profit to look at developing a game similar to the award-winning www.playspent.org They want the following features: support 5 scenarios mobile isn't important but compatibility with older browsers would be a big bonus they want it to be visual and audible bonus is to have it easily modifiable support 4 languages I don't have much knowledge of Flash and would rather avoid using it as a solution. I started breaking down the problem into segments that I will need to examine, they are as follows: ability to read the game flow from a file that they can produce (xml, etc.) db design to store decision tree language challenge browser compatibility I am leaning towards an Google app engine/GWT solution but I am not sure what technology is best for this. I am really hoping to get your opinion/recommendation on my approach and on what technology is best. A special thanks (and beer if you live in Toronto) will be awarded to anyone who can help give me a ballpark estimate on how much such a game should go for. I know it's tough to estimate but any rough figure will help (how much would you charge for building something like playspent.org?) Thanks in advance

    Read the article

  • Command-line HTTP crawler for Windows?

    - by Pekka
    Would somebody have a recommendation for a web site crawler that can be invoked and equipped with settings from the command line? This would need to run in a Windows environment. Saving the data, following stylesheet links etc. is not an issue. I only need the crawler to start with a page, parse it, and follow all the links on the same domain so that in the end, all pages on the site have been requested once. Background: I'm setting up a web site that gets frequently uploaded from an office location. Combining data from various sources, it has several levels of caching. I don't want the first user to visit the site after a fresh upload to have to wait until the page has been generated and saved in the cache.

    Read the article

  • Can't enable wireless lan on Fujitsu Siemens A1665G with Ubuntu 11.10 installed

    - by Theo
    I saw my old Notebook yesterday and wanted to make that work again. On Windows XP the wireless worked still fine. Then I installed new Ubuntu 11.10 32bit and I'm sadly not able to make the wireless enabled. [I replaced Win XP entirely] lspci lists following network: 08:0a.0 Network controller: Broadcom Corporation BCM4318 [AirForce One 54g] 802.11g Wireless LAN Controller (rev 02) So after recommendation from this link I installed the b43 firmware module. iwconfig prints the following: wlan0 IEEE 802.11bg ESSID:off/any Mode:Managed Access Point: Not-Associatd Tx-Power=off Retry long limit:7 RTS thr:off Fragment thr:off Power Management: off As you can see, my wireless lan adapter is not turned on. sudo iwconfig wlan0 txpower auto Doesn't change anything. Then I tried to make it work with rfkill. rfkill list 0: phy0: Wirless LAN Soft blocked: no Hard blocked: yes sudo rfkill unblock all rfkill list 0: phy0: Wireless LAN Soft blocked: no Hard blocked: yes remains the same. The question is now, how I could enable the hard blocked wireless LAN. There is no hardware switch for wlan integrated. However there is a button to change the state. I always thought this would be software sided, but it seems to make some hardware changes as well... The wireless LED is also not blinking (as it did on windows xp) I reset bios and searched for some settings in there, but it has only a few options and nothing to do with wireless settings, nothing works here.. At last I tried to install the acer hotkeys but I was not able to manage that. I installed the acerhkgui package, but in initializing progress, it was not able to compile acer hotkeys for my machine. There was a message that asm/linkage.h was not found while compiling. Do you have any ideas what I could do to make this hard blocked stuff disabled and my wireless card work? PS: I also tried sudo rm /dev/rfkill and a reboot to reinit that stuff... No success :(

    Read the article

  • How Often Do Windows Servers Need to be Restarted?

    - by Evan
    A little background: We have several Windows servers (2003, 2008) for our department. We're a division of IT so we manage our own servers. Of the four of us here I'm the only one with a slight amount of IT knowledge. (Note the "slight amount".) My boss says the servers need to be restarted at least weekly. I disagree. Our IT Department says that because she restarts them constantly that's the reason why our hard drives fail and power supplies go out on them. (That's happened to a few of our servers a couple times over the last four years, and very recently.) So the question is: How often does everyone restart their Windows servers? Is there an industry standard or recommendation? Is our IT department correct in saying that because we re-start that's why we're having hardware issues? (I need a reason if I'm going to change her mind!)

    Read the article

  • cannot mount root filesystem on 10.04

    - by howaryoo
    I tried to apply the recommendation of question: Kernel Panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) After running that command: sudo mount --bind /dev /mnt/dev I get this error message: mount: mount point /mnt/dev does not exist fdisk -l returns /dev/sda1 /dev/sda2 /dev/sda5 do I need to mount sda2 and sda5? Edited after @psusi's comment: /dev/sda1 is the boot file system It seems that I need to mount sda2 or sda5. Here is what I tried: (I tried this on a virtual machine so the sda(s) are now vda(s) ) ubuntu@ubuntu:~$ sudo fdisk -l Disk /dev/vda: 19.3 GB, 19327352832 bytes 16 heads, 63 sectors/track, 37449 cylinders Units = cylinders of 1008 * 512 = 516096 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x0008eece Device Boot Start End Blocks Id System /dev/vda1 * 3 496 248832 83 Linux Partition 1 does not end on cylinder boundary. /dev/vda2 498 37448 18622465 5 Extended Partition 2 does not end on cylinder boundary. /dev/vda5 498 37448 18622464 8e Linux LVM ubuntu@ubuntu:~$ ubuntu@ubuntu:~$ ubuntu@ubuntu:~$ sudo mount -t ext4 /dev/vda5 /mnt mount: wrong fs type, bad option, bad superblock on /dev/vda5, missing codepage or helper program, or other error In some cases useful info is found in syslog - try dmesg | tail or so ubuntu@ubuntu:~$ sudo mount -t ext2 /dev/vda5 /mnt mount: wrong fs type, bad option, bad superblock on /dev/vda5, missing codepage or helper program, or other error In some cases useful info is found in syslog - try dmesg | tail or so ubuntu@ubuntu:~$ Any info that can help me rescue that server would be greatly appreciated!

    Read the article

  • Is there a way to determine which service does an outgoing connection?

    - by fluxtendu
    I'm redoing my firewall configuration with more restrictive policies and I would like to determine the provenance (and/or destination) of some outgoing connections. I have an issue because they come from svchost.exe and go to web content/application delivery providers - or similar: 5 IP in range: 82.96.58.0 - 82.96.58.255 --> Akamai Technologies akamaitechnologies.com 3 IP in range: 93.150.110.0 - 93.158.111.255 --> Akamai Technologies akamaitechnologies.com 2 IP in range: 87.248.194.0 - 87.248.223.255 --> LLNW Europe 2 llnw.net 205.234.175.175 --> CacheNetworks, Inc. cachefly.net 188.121.36.239 --> Go Daddy Netherlands B.V. secureserver.net So is it possible to know which service does a particular connection? Or what's your recommendation about the rules applied to these ones? (Comodo Firewall & Windows 7)

    Read the article

  • More Tables or More Databases?

    - by BuckWoody
    I got an e-mail from someone that has an interesting situation. He has 15,000 customers, and he asks if he should have a database for their data per customer. Without a LOT more data it’s impossible to say, of course, but there are some general concepts to keep in mind. Whenever you’re segmenting data, it’s all about boundary choices. You have not only boundaries around how big the data will get, but things like how many objects (tables, stored procedures and so on) that will be involved, if there are any cross-sections of data (do they share location or product information) and – very important – what are the security requirements? From the answer to these types of questions, you now have the choice of making multiple tables in a single database, or using multiple databases. A database carries some overhead – it needs a certain amount of memory for locking and so on. But it has a very clean boundary – everything from objects to security can be kept apart. Having multiple users in the same database is possible as well, using things like a Schema. But keeping 15,000 schemas can be challenging as well. My recommendation in complex situations like this is similar to a post on decisions that I did earlier – I lay out the choices on a spreadsheet in rows, and then my requirements at the top in the columns. I  give each choice a number based on how well it meets each requirement. At the end, the highest number wins. And many times it’s a mix – perhaps this person could segment customers into larger regions or districts or products, in a database. Within that database might be multiple schemas for the customers. Of course, he needs to query across all customers, that becomes another requirement. Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • IP assignment in a /28 block

    - by mks
    Need help on setting up firewall router. My config is as below: Public static network ID: x.x.x.48/28 gateway: x.x.x.49 available IP for the hosts: x.x.x.52 to 62 gw_eth0 <-- fw_eth0 - fw_eth1 <-- dmz_switch Four servers are connected on dmz_switch (say s1, s2, s3, s4) all have to use public static IP address from the above block. Any recommendation on IP assignment and route setup? Do I need to subnet the above block further or simply use /32 netmask and point-to-point static routes in the above setup?

    Read the article

  • Bad disks in ancient server

    - by Joel Coel
    I have a 1998-era Netware 3.12 server that runs everything on our campus: general ledger, purchasing, payroll, student information, grades, you name it. The server has an Adaptec RAID controller with two volumes: RAID 1, 2 17GB scsi disks, Seagate ST318417W RAID 5, 3 4GB scsi disks, 2 Seagate ST34573W and 1 ST34572W. We are currently in the early stages of a project to replace this system, but you don't just jump into a new system like that and so I need to keep this server running until at least November 2011. This week we had not one but two hard drives fail. Thankfully they are from different volumes and we're able to keep running for the moment, but given the close nature of these failures I have serious doubts that I'll be able to avoid catastrophic failure from this server through the November target as is without restoring the RAID redundancy — it'll only take one more drive failure anywhere and I'm completely hosed. We are fortunate enough to have exact match "spares" lying around for both drives, but the spares are in unknown condition. I tried swapping just them in, but the RAID controller isn't smart enough to handle this and it renders the system unbootable. As for the RAID controller itself, there is utility I can get into during POST via a Ctrl-A shortcut, but I can't do much useful from there. To actually manage volumes I must first boot in to Netware, at which point I can use CI/O Array Management Software Version 2.0 to actually look at volume information. I suspect that the normal way to manage things is to boot from a special floppy with the controller software on it, but that floppy is long gone. Going through the options in the RAID software, I think the only supported way to replace a disk in an existing RAID volume is to physically add the disk, boot up and configure it as a "spare" for a volume, force the volume to use the spare to replace an existing down disk (and at this point I'm only guessing) so that the down disk becomes the spare, repair the volume, remove the spare from the volume, and then shut down and remove the disk. Then start all over for the other failed disk. All this amounts to a lot of downtime, assuming I can even make it work and that my spares are any good. As for finding reliable spares, I have no clue where to even begin looking to find a new 4GB scsi drive, or even which exact scsi system I'm looking for, as it's gone through a few different iterations over time. Another option is to migrate this to a virtual machine (hyper-v), but all previous attempts we've made in this area have failed to get very far. When this machine was installed I was just graduating from high school, and so it requires lower level knowledge of netware and dos than I ever developed, or if I did have since forgotten (I'm not exactly a dos neophyte, either). Part of my problem is this is a high-use server, and taking it down for a few days to figure things out isn't gonna fly very well. As for the question, I'm looking for anything that might be helpful in this situation: a recommendation on a place to find good spares from this era, personal experience repairing RAID volumes using a similar controller or building a hyper-v vm from an old netware server, a line on a floppy with better software for the RAID controller, recommendation on a good Novell consultant in Nebraska that would be able to put things right, a whole other option I haven't considered yet, etc. Update: For backups, we have good (recently verified via restore) backups of the data only -- nothing for the software that actually runs things. Update 2: Just a progress report that I currently have a working Netware 3.12 install in VMWare Virtual Server 2.0, thanks largely to the guide I found here: http://cerbulescubogdan.blogspot.com/2010/11/novell-netware-312-on-vmware.html The next steps are preparing empty netware volumes to match the additional volumes on my existing server, taking a dump of everything on the C:\ drive and netware volumes on my existing server, and figuring out from that information what modules need added to netware, installing my licenses (we do still have that disk, if it's any good), and moving data over. I have approval to bring the server down for a week after the first of the year (sadly not before), so, aside from creating empty volumes, the rest of the work will have to wait until then. Final Update (Jan 5, 2011): I was able to get spares working in both raid arrays without data loss this week. Both are now listed by the controller as "FAULT TOLLERANT" (yay!). I was also able to build on the progress from my last update and now have a functional "spare" server in VMWare Server 2.0. The spare can run and use our erp software, but I can't put it into production because I can't (yet) print from that box (and I have no idea why). Even so, this VM will do in a pinch if I have no other choice, and between it and the repaired RAID arrays I'm comfortable pushing on until I can junk the machine in November.

    Read the article

< Previous Page | 23 24 25 26 27 28 29 30 31 32 33 34  | Next Page >