Search Results

Search found 13748 results on 550 pages for 'split testing'.

Page 352/550 | < Previous Page | 348 349 350 351 352 353 354 355 356 357 358 359  | Next Page >

  • DNS-Based Environment Determination

    - by zvolkov
    Found the following here. The questions is: where can I find more details on how exactly implement this on Windows? Any guide or how-to anybody? Or maybe you can provide your invaluable suggestions? Specifically, how do I make so that "all QA servers would first resolve entries in qa.example.com first and then if that lookup failed they would try example.com" (I'm a dev, not a DNS specialist, but our IT Support has refused to help on this:() Use DNS Based Environment Determination for your servers. Do this by initially splitting your top level domain into a number of sub domains depending on their function, and then creating DNS Service Names in each of the sub domains pointing to the relevant server for that service. Based on the list above we would then have: * clientdb.prod.example.com for Production * clientdb.perf.example.com for Performance Testing * clientdb.qa.example.com for QA * clientdb.dev.example.com for Development Servers then resolve entries in their relevant sub domain by function. That is, all QA servers would first resolve entries in qa.example.com first and then if that lookup failed they would try example.com. This allows you to have a single configuration entry for your client database hostname (clientdb) that would resolve correctly in all environments. This technique has the added advantage of still having global services defined in a common top level domain. This seems to be related to Providing "split horizon" DNS service. Reading that, I see that I will probably need separate DNS Server for each environment. Is this true or does Windows support some form of "tagging" the records to be visible depending on the requestor's IP?

    Read the article

  • Can I get ethernet out of my Verizon FIOS set-top box?

    - by Tom Hughes
    Setup: my home network is long & skinny, and the FIOS-connected router is all the way at one of the apartment. At the other end, far away (and a floor higher) is my HD TV, which gets a cable-TV signal from a Verizon set-top box that is coax-connected back to the FIOS on-premises equipment. Wi-Fi won't work, the apartment is too stretched out, with old, thick walls and floors. Goal: I think there are three ways to get ethernet back to where the HD TV is: 1) run a cable! this isn't crazy but isn't cheap either (my building won't let me do it, it involves hiring an electrician because the cable would run partly through the public hallway ceiling) 2) split the coax near the TV and put in... a MoCA device? 3) somehow tease the set-top box, which has an RJ-11 (ethernet) port on the back, to give me network access. Question: any other choices? and, is one choice better than the others? #3 is by far the most desirable because it would involve the least wiring -- but I can't find any resources to help make it happen. #2 is a bit scary, I don't want to degrade service to the TV or anywhere else for that matter.

    Read the article

  • Architecture for highly available MySQL with automatic failover in physically diverse locations

    - by Warner
    I have been researching high availability (HA) solutions for MySQL between data centers. For servers located in the same physical environment, I have preferred dual master with heartbeat (floating VIP) using an active passive approach. The heartbeat is over both a serial connection as well as an ethernet connection. Ultimately, my goal is to maintain this same level of availability but between data centers. I want to dynamically failover between both data centers without manual intervention and still maintain data integrity. There would be BGP on top. Web clusters in both locations, which would have the potential to route to the databases between both sides. If the Internet connection went down on site 1, clients would route through site 2, to the Web cluster, and then to the database in site 1 if the link between both sites is still up. With this scenario, due to the lack of physical link (serial) there is a more likely chance of split brain. If the WAN went down between both sites, the VIP would end up on both sites, where a variety of unpleasant scenarios could introduce desync. Another potential issue I see is difficulty scaling this infrastructure to a third data center in the future. The network layer is not a focus. The architecture is flexible at this stage. Again, my focus is a solution for maintaining data integrity as well as automatic failover with the MySQL databases. I would likely design the rest around this. Can you recommend a proven solution for MySQL HA between two physically diverse sites? Thank you for taking the time to read this. I look forward to reading your recommendations.

    Read the article

  • mod_rewrite REQUEST_FILENAME doesn't contain absolute path

    - by Paul Dixon
    I have a problem with a file test operation in a mod_rewrite RewriteCond entry which is testing whether %{REQUEST_FILENAME} exists. It seems that rather than %{REQUEST_FILENAME} being an absolute path, I'm getting a path which is rooted at the DocumentRoot instead. Configuration I have this inside a <VirtualHost> block in my apache 2.2.9 configuration: RewriteEngine on RewriteLog /tmp/rewrite.log RewriteLogLevel 5 #push virtually everything through our dispatcher script RewriteCond %{REQUEST_FILENAME} !-f RewriteCond %{REQUEST_FILENAME} !-d RewriteRule ^/([^/]*)/?([^/]*) /dispatch.php?_c=$1&_m=$2 [qsa,L] Diagnostics attempted That rule is a common enough idiom for routing requests for non-existent files or directories through a script. Trouble is, it's firing even if a file does exist. If I remove the rule, I can request normal files just fine. But with the rule in place, these requests get directed to dispatch.php Rewrite log trace Here's what I see in the rewrite.log init rewrite engine with requested uri /test.txt applying pattern '^/([^/]*)/?([^/]*)' to uri '/test.txt' RewriteCond: input='/test.txt' pattern='!-f' => matched RewriteCond: input='/test.txt' pattern='!-d' => matched rewrite '/test.txt' -> '/dispatch.php?_c=test.txt&_m=' split uri=/dispatch.php?_c=test.txt&_m= -> uri=/dispatch.php, args=_c=test.txt&_m= local path result: /dispatch.php prefixed with document_root to /path/to/my/public_html/dispatch.php go-ahead with /path/to/my/public_html/dispatch.php [OK] So, it looks to me like the REQUEST_FILENAME is being presented as a path from the document root, rather than the file system root, which is presumably why the file test operator fails. Any pointers for resolving this gratefully received...

    Read the article

  • Campus VLAN Segmentation - By OS?

    - by Moduspwnens
    We've been thinking through re-arranging our network and VLAN configuration. Here's the situation. We already have our servers, VoIP phones, and printers on their own VLANs, but our problem lies with end user devices. There are just too many to lump on the same VLAN without being hammered with broadcasts! Our current segmentation strategy has them split into VLANs like this: Student iPads Staff iPads Student Macbooks Staff Macbooks Gaming devices Staff (Other) Student (Other) *Note that our network has many more iPads and MacBooks than most. Since the primary reason we're splitting them is just to put them in smaller groups, this has been working for us (for the most part). However, this required our staff to maintain access control lists (MAC addresses) of all devices belonging in these groups. It also has the unfortunate side effect of illogically grouping broadcast traffic. For example, using this setup, students on opposite ends of campus using iPads will share broadcasts, but two devices belonging to the same user (in the same room) will likely be on completely separate VLANs. I feel like there must be a better way of doing this. I've done a lot of research and I'm having trouble finding instances of this kind of segmentation being recommended. The feedback on the most relevant SO question seems to point toward VLAN segmentation by building/physical location. I feel like that makes sense because logically, at least among miscellaneous end users, broadcasts will typically be intended for nearby devices. Are there other campuses/large-scale networks out there segmenting VLANs based on end-system OS? Is this a typical configuration? Would VLAN segmentation based on physical location (or some other criteria) be more effective? EDIT: I've been told that we will soon be able to dynamically determine device OS without maintaining access lists, although I'm not sure how much that affects the answers to the questions.

    Read the article

  • Performance: Nginx SSL slowness or just SSL slowness in general?

    - by Mauvis Ledford
    I have an Amazon Web Services setup with an Apache instance behind Nginx with Nginx handling SSL and serving everything but the .php pages. In my ApacheBench tests I'm seeing this for my most expensive API call (which cache via Memcached): 100 concurrent calls to API call (http): 115ms (median) 260ms (max) 100 concurrent calls to API call (https): 6.1s (median) 11.9s (max) I've done a bit of research, disabled the most expensive SSL ciphers and enabled SSL caching (I know it doesn't help in this particular test.) Can you tell me why my SSL is taking so long? I've set up a massive EC2 server with 8CPUs and even applying consistent load to it only brings it up to 50% total CPU. I have 8 Nginx workers set and a bunch of Apache. Currently this whole setup is on one EC2 box but I plan to split it up and load balance it. There have been a few questions on this topic but none of those answers (disable expensive ciphers, cache ssl, seem to do anything.) Sample results below: $ ab -k -n 100 -c 100 https://URL This is ApacheBench, Version 2.3 <$Revision: 655654 $> Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/ Licensed to The Apache Software Foundation, http://www.apache.org/ Benchmarking URL.com (be patient).....done Server Software: nginx/1.0.15 Server Hostname: URL.com Server Port: 443 SSL/TLS Protocol: TLSv1/SSLv3,AES256-SHA,2048,256 Document Path: /PATH Document Length: 73142 bytes Concurrency Level: 100 Time taken for tests: 12.204 seconds Complete requests: 100 Failed requests: 0 Write errors: 0 Keep-Alive requests: 0 Total transferred: 7351097 bytes HTML transferred: 7314200 bytes Requests per second: 8.19 [#/sec] (mean) Time per request: 12203.589 [ms] (mean) Time per request: 122.036 [ms] (mean, across all concurrent requests) Transfer rate: 588.25 [Kbytes/sec] received Connection Times (ms) min mean[+/-sd] median max Connect: 65 168 64.1 162 268 Processing: 385 6096 3438.6 6199 11928 Waiting: 379 6091 3438.5 6194 11923 Total: 449 6264 3476.4 6323 12196 Percentage of the requests served within a certain time (ms) 50% 6323 66% 8244 75% 9321 80% 9919 90% 11119 95% 11720 98% 12076 99% 12196 100% 12196 (longest request)

    Read the article

  • Tri-head linux system with Xmonad: is it possible to have HW acceleration

    - by progo
    What means there exists to have three monitors, all controlled by Xmonad and have hardware 3D acceleration as well? I had the pleasure of using three monitors earlier this year, and while Xmonad and Xinerama handle three monitors easily, I had to throw in an extra display driver, and also let go of Nvidia's own TwinView (which is a hack on Xinerama). This left me with no HW acceleration and some flickering as double buffering wouldn't work with certain applications. However, the three monitors handle so beautifully that I had hard time coming back to two. I understand the easiest way to achieve HW-accelerated tri-head combo is to split into two Xorgs. I wouldn't be able to switch windows between the Xorgs, so I'm not really into this solution. What's more, having a cheap and old PCI card along with even slightly better PCIe seemed to slow things down. Even if I occasionally disabled the third monitor from Xorg configure, I couldn't get HW acceleration to work. Only after I physically disconnected the old PCI card, I could get the games back in business. Would a Matrox Dual/Tri-head2go and a powerful Nvidia GPU do the trick? I understand Xmonad can be configured to "believe" that a "single" (as Dualhead2Go will merge) 3360x1050 display is actually two different ones? So that Xmonad's Mod-w and Mod-e would work properly there.

    Read the article

  • Can Subject Alternative Name accommodate multiple virtual mail domains?

    - by Lawrence
    I am currently running a postfix server with self signed certificates serving one mail domain, mycompany.com, the mail server is mail.mycompany.com and so is the CN of the certificate. Now, I need to add a new domain to it. The new domain name is mycompany.net to the same server. Since the users already have the root of the old certificate, I'd like to reuse that. However, I'd like to issue a new certificate so users using the SMTP from Outlook/Thunderbird of mail.mycompany.net do not get warnings. If I understand correctly, if I issue a new certificate with CN=mail.mycompany.com and a subjectAltName=DNS:mail.mydomain.net and have postfix serve this, the client will not complain either way about the cn not matching the target host name. Am I correct in this assumption or am I misunderstanding the concept of Subject Alternative Name? Just to avoid conversation, I do not want to have users on mycompany.net addresses use the mycompany.com server because I might (not a technical issue) have to split up into two different locations, and I want to produce an easily migrateable setup.

    Read the article

  • How to set up the jdbc driver to connect to hsqldb from libreoffice?

    - by rumtscho
    I am trying to "split" a LibreOffice .odb file into a HSQL database and an OpenOffice document containing forms and macros. I am trying to follow the instructions from this thread: Within a few minutes you can convert your embedded HSQLDB to a stand-alone HSQLDB which is just a very fine database engine. 1) Download and extract the current version from http://hsqldb.org/ and point the Java class path in ToolsOptionsJava to the new hsqldb.jar 2) Extract the database folder from your embedded database and rename the files data, properties, script to name.data name.properties, name.script where "name." is an arbitrary name prefix. 3) Connect a Base document to an existing JDBC database such as jdbc:hsqldb:file:/home/chenier/hsqldb/name;default_schema=true;shutdown=true;hsqldb.default_table_type=cached;get_column_name=false (again, "name" refers to your own file name prefix). This local single-user connection gives you much more than the embedded HSQLDB. 4) Copy queries, forms and reports from the old database over to the new one. The wizard presents me with a window expecting two inputs: a "Datasource URL" and a "JDBC driver class". As far as I can tell, the tutorial above only tells me what to put into the Datasource URL. As for the JDBC driver class, I have no idea what to write into this field. I tried the fully-qualified name of the Java class, org.hsqldb.jdbc.JDBCDriver as given in the HSQLDB documentation. When that failed, I tried the physical path /var/lib/hsqldb/lib/hsqldb.jar (although that should have been unnecessary, because first I pointed to this path as described under 1 and then restarted LibreOffice). In both cases, "Test class" failed with the message "The JDBC driver could not be loaded". OpenOffice's documentation doesn't say anything sensible about the field, it was something like "enter the JDBC driver in this box". Any ideas what I should enter there to get the connection working?

    Read the article

  • Backup plan for linux webserver in small business?

    - by radman
    Hi, I am currently in the process of writing a backup plan for the webserver in use by my business. I am very new to this area and have a few ideas about how things should work but am unsure of what tools to use and what sort of restore process is appropriate. I'm looking for something relatively simplistic and it doesn't have to be 100% paranoid just enough to give me a reliable backup. Speed is not of the essence and there is not going to be a live fallback in place. The backup will be onto a single hdd that will be stored onsite (no option for offsite as yet). Backups will be taking place weekly. I am constrained by both time and money which is why I'm aiming for a good enough solution. Is taking an image of the webserver system drive periodically and using that as the backup appropriate? Should I be testing that the backups restore correctly every time that I perform one? This is a bit broad but what setup would you use if you were in my place, given the services I am running? Should I add additonal machines and split the services? Any advice is much appreciated! See below for server details Webserver Platform Linux Ubuntu server Running mail-server svn-server mediawiki wordpress apache-webserver Hardware single 500gb sata drive Architecture Single machine behind router (with firewall) accessible to the internet.

    Read the article

  • Move 53,800+ files into 54 separate folders with ~1000 files each?

    - by ane
    Trying to import 53,800+ individual files (messages) using Gmail's POP fetcher. Gmail understandably refuses, giving the error: "Too many messages to download. There are too many messages on the other server." The folder in question looks like similar to: /usr/home/customer/Maildir/cur/1203672790.V57I586f04M867101.mail.net:2,S /usr/home/customer/Maildir/cur/1203676329.V57I586f22M520117.mail.net:2,S /usr/home/customer/Maildir/cur/1203677194.V57I586f26M688004.mail.net:2,S /usr/home/customer/Maildir/cur/1203679158.V57I586f2bM182864.mail.net:2,S /usr/home/customer/Maildir/cur/1203680493.V57I586f33M740378.mail.net:2,S /usr/home/customer/Maildir/cur/1203685837.V57I586f0bM835200.mail.net:2,S /usr/home/customer/Maildir/cur/1203687920.V57I586f65M995884.mail.net:2,S ... Using the shell (tcsh, sh, etc. on FreeBSD), what one-line command can I type to split this directory full of files into separate folders so Gmail only sees 1000 messages at a time? Something with find or ls | xargs mv maybe. Whatever is fastest. The desired output directory would now look something like: /usr/home/customer/Maildir/cur/1203672790.V57I586f04M867101.mail.net:2,S /usr/home/customer/Maildir/cur/1203676329.V57I586f22M520117.mail.net:2,S ... /usr/home/customer/set1/ (contains messages 1-1000) /usr/home/customer/set2/ (contains messages 1001-2000) /usr/home/customer/set3/ (etc.) Ideally, cron could run another command to automatically reverse the process in 1000 message increments every hour. So Gmail only sees & downloads 1000 at a time.

    Read the article

  • Migrating away from LVM

    - by Kye
    I have an Ubuntu home media server setup with 4.5TB split across a few hard-drives (1x3TB, 2x1TB) and I'm using LVM2 to manage the volumes. I have recently added a 60GB SSD to my server, and I wish to use it to house the 'root' partition of my server (which is currently under the LVM group). I don't want to simply add it to the LVM volume group, because (afaik) there's no way to ensure that the SSD will be used for the root filesystem. If I just throw it at the VG, it may be used to house my media, which would defeat the purpose of having the SSD in the first place. I feel that my only solution is to somehow remove my root partition from the LVM setup and copy it across to the SSD. My boot partition is, of course, not part of the LVM group. My disk setup is as follows: 60GB SSD: EMPTY. 1TB HDD: /boot, LVM space. 1TB HDD: LVM space. 3TB HHD: LVM space. I have a few logical volumes. my root (/), a 'media' volume for my media collection, a backup one for my network backups.etc. Does anyone have any advice as to how to go about this? My end goal is to have the 60GB SSD used for my boot and root partitions, with everything else on the 3TB/1TB/1TB hard-drives.

    Read the article

  • Website Use Monitoring for 3 People

    - by linkedlinked
    I work in an IT startup with 2 partners, and I'm the programmer/IT guy -- in other words, the work horse. To make a long story short, I'm doing most of the work right now, while they spend all day on Facebook. That's OK, because they're paying my salary, but if the project fails, I'm sure they'll blame me for it (I'm doing my best to make sure that doesn't happen!), and I want some sort of recourse. I already have an app that blocks time-wasters on my local PC, and keeps logs of when the app is enabled (so I can say "I had Facebook blocked from 9am-5pm today.") Is there any way I can get a brief summary of the most heavily visited sites, split up by client PC? At the end of the month, I want to be able to say "You both load Facebook, on average, every 10 minutes. You spend hours a day on Youtube, and haven't opened up our bugtracker in weeks" and maybe have a nifty chart or graph to match it. We have a crappy D-Link router, and no IT budget. They are both on Windows Vista, I run Ubuntu Linux. I don't want to install any monitoring software on their PC, but I'm totally fine with, say, routing all the network traffic through my machine. I guess I can think of lots of ways to accomplish this (telnet into JSSH and list open tabs? log all the DNS requests, per-domain? even thinking of setting up a webcam on my desk and just keeping 5-minute snapshots...), I just don't really know where to start. Any advice is appreciated, thanks!

    Read the article

  • Loading dependencies for custom puppet functions

    - by Ben Smith
    I have written a custom puppet function, which is working fine, that depends on the cloudservers gem (a Rackspace client library). This is fine if I have pre-installed the gem on a server before running puppet but totally breaks if I have not installed the gem as the function seems to be run during the 'compilation' sweep, well before my package definition is realised. Here's what my .pp looks like, with get_hosts the function that requires the cloudservers gem. package { "rubygems": ensure => installed, provider => "gem"; } package { "cloudservers": ensure => installed, provider => "gem", require => Package["rubygems"]; } class hosts::us { $hosts = get_hosts("us") hostentry { $hosts: } } define hostentry() { $parts = split($name, ',') $address = $parts[0] $ip = $parts[1] $aliases = $parts[2] host{ $address: ip => $ip, host_aliases => $aliases } } Is there a way to stop the function getting run so early, or at least having it's run depend up the library being installed. Alternatively, is there a way that I can add dependencies somewhere in the functions folder that will be available to the function?

    Read the article

  • Ram question in VMware Server 2

    - by ToreTrygg
    Hi, I understand from the VMware Server 2 documentation that VMware Server 2 is capable of running a 64-bit guest OS underneath a 32-bit host OS, as long as the hardware running the box is 64-bit capable. Here's my situation. We currently have an underutilized XEON X3220 Quad Core 64bit Server, running Server 2003, 32-bit and 2gb of RAM (the motherboard is capable of 8gb ram). The server is currently used mainly for file and print services. It is also running Active Directory, Novell eDirectory and Groupwise 6.5. We are planning a micration to Microsoft Exchange, so the Novell eDirectory and Groupwise services will eventually be purged from this box, leaving only Active Directory, File and Print services. Being that this server is underutilized we are hoping to save hardware costs and virtualize our new Exchange investment. My question is this. Will VMware allow access to the "invisible" extra memory that Windows 32-bit won't see. Meaning, if we increase the full amount of system ram to 8gb (yes, I know the 32-bit host OS will only see a maximum of 4gb), will I be able to assign maybe 5gb to the new Server 2008 64-bit OS running Exchange and leave 3gb for the Guest OS (or maybe even a 6, 2 split). The second part of that would be, would it be better to just convert the main OS currently running to an image, convert the machine itself to ESXi and run both OSes as images under ESXi. Downtime for this box is critical, so my preference is most definitly with the first option because it presents very minimal downtime. Doing the second would make downtime quite a few hours to image the machine and then convert the image to a VMware Image.

    Read the article

  • Excel 2007 - Adding line breaks in a cell and no line over 50 characters

    - by Richard Drew
    I have notes stored in an excel cell. I add line breaks and dates every time I add a new note. I need to copy this to another program, but it has a line limit of 50 characters. I want a line break for each new date and for when each date's comment goes over 50 characters. I'm able to do one or the other, but I can't figure out how to do both. I'd prefer words not to be split up, but at this point I don't care. Below is some sample input. If needed for a SUBSTITUTE or REPLACE function, I could add a ~ before each date in my input as a delimiter. Sample Input: 07/03 - FU on query. Copies and history included. CC to Jane Doe and John Public 06/29 - Cust claiming not to have these and wrong PO on query form. Responded with inv sent dates and locations, correct PO values, and copies. 06/27 - New ticket opened using query form 06/12 - Opened ticket with helpdesk asking status 05/21 - Copy submitted to [email protected] 05/14 - Copy sent to John Public and [email protected] Ideal Output: 07/03 - FU on query. Copies and history included. CC to Jane Doe and John Public 06/29 - Cust claiming not to have these and wrong PO on query form. Responded with inv sent dates an d locations, correct PO values, and copies. 06/27 - New ticket opened using query form 06/12 - Opened ticket with helpdesk asking status 05/21 - Copy submitted to [email protected] om 05/14 - Copy sent to John Public and email@custome r.com

    Read the article

  • Getting Dell E6320 with I7 to work with 3 monitors at 1920x1080p x 3

    - by MadBoy
    I want to buy Dell E6320 which comes with Intel Core I7-2620M (2.70GHz, 4MB cache, Dual Core) with Intel HD Graphics 3000. Laptop will come with docking station. I want to connect 3 monitors to that docking station so that working at home would give me some additional boost. Docking station will allow me to connect only 2 monitors so I'm looking at following other options: Matrox TRIPLEHEAD2GO DIGITAL Edition or TRIPLEHEAD2GO DP Edition. But reading Matrox Support Page intel GPU can't run the highest resolution with 3 monitors connected, it even gets worse since it seems monitors would have to be able to work at 50hz. Also I'm not sure but it seems that Matrox doesn't split the monitors as 3 separate monitors but simply as one big space (which is a bit opposite to what I need) Buy 2 or maybe just 1 USB based monitor but it would also mean having 1 or 2 different monitors then the main one, unless I buy 3 USB based monitors which would mean more money to spend. Also I found only couple of models and most of them require USB 3.0 and no other cables to plug in (nice but costly - couldn't find decent monitor with only USB for sending signal and having power connected normally) . But docking station has only one USB 3.0 port. Can I use hub and still get it to work? Find some converters from Digital to USB (I think DisplayLink does some?) Buy different laptop but what kind? I need it to be I7, small (13"), fast and lightweight. At same time it requires docking station that I can use at home to connect 3 external monitors. Some other suggested solution... Edit: I need 3 monitors for work in terms of coding in Visual Studio or having word/excel/outlook open. Nothing fancy. Maybe some movie once in a while.

    Read the article

  • Wiring my internet

    - by u8sand
    I have Verizon internet service and am currently using wifi. My router is in the basement and my desktop computer is 2 floors and on the other side of the house above it... Worst possible positioning but that's just how things worked out. My wireless currently is extremely unstable so I've decide to correct the problem by wiring my computer directly. The problem lies here: when redoing the room next to it (when the wall was open) we went ahead and wired some coaxial cable from our attic to our basement (with plenty of slack on both ends, don't ask me why we didn't go ahead and wire a CAT6 cable). The question is: Can I use the coaxial cable to bring me internet connection? Naturally the router (which needs to stay where it is) takes a coaxial cable input and has Ethernet outputs. So maybe I would have to take a ethernet cable, convert to coaxial-coaxial to my computer, convert back to ethernet. Is this even possible to convert from coaxial to ethernet? Or do I have to attempt to go ahead and fish a cat6 cable through my house. I cannot just split the signal because that would require two routers and two networks (which I don't believe would work with one cable-one ISP correct me if I'm wrong). Thanks

    Read the article

  • SBD killing both cluster nodes when there are even small SAN network problems

    - by Wieslaw Herr
    I am having problems with stonith SBD in a openais-based cluster. Some background: The active/passive cluster has two nodes, node1 and node2. They are configured to provide an NFS service to users. To avoid problems with split-brain, they are both configured to use SBD. SBD is using two 1MB disks available to the hosts via an multipath fibre-channel network. The problems start if something happens with the SAN network. For example, today one of the brocade switches got rebooted and both nodes lost 2 out of 4 paths to each disks, which resulted in both nodes committing suicide and rebooting. This, of course, was highly undesirable because a) there were paths left b) even if the switch would be out for 10-20 seconds a reboot cycle of both nodes would take 5-10 minutes and all NFS-locks would be lost. I tried increasing the SBD timeout values (to 10sec+ values, dump attached at the end), however a "WARN: Latency: No liveness for 4 s exceeds threshold of 3 s" hints that something isn't working as I would it expect to. Here is what I would like to know: a) Is SBD working as it should killing nodes when 2 paths are available? b) If not, is the multipath.conf file attached correct? The storage controller we use is an IBM SVC (IBM 2145), should there be any specific configuration for it? (as in multipath.conf.defaults) c) How should I go about increasing the timeouts in SBD attachements: Multipath.conf and sbd dump (http://hpaste.org/69537)

    Read the article

  • Can enabling a RAID controller's writeback cache harm overall performance?

    - by Nathan O'Sullivan
    I have an 8 drive RAID 10 setup connected to an Adaptec 5805Z, running Centos 5.5 and deadline scheduler. A basic dd read test shows 400mb/sec, and a basic dd write test shows about the same. When I run the two simultaneously, I see the read speed drop to ~5mb/sec while the write speed stays at more or less the same 400mb/sec. The output of iostat -x as you would expect, shows that very few read transactions are being executed while the disk is bombarded with writes. If i turn the controller's writeback cache off, I dont see a 50:50 split but I do see a marked improvement, somewhere around 100mb/s reads and 300mb/s writes. I've also found if I lower the nr_requests setting on the drive's queue (somewhere around 8 seems optimal) I can end up with 150mb/sec reads and 150mb/sec writes; ie. a reduction in total throughput but certainly more suitable for my workload. Is this a real phenomenon? Or is my synthetic test too simplistic? The reason this could happen seems clear enough, when the scheduler switches from reads to writes, it can run heaps of write requests because they all just land in the controllers cache but must be carried out at some point. I would guess the actual disk writes are occuring when the scheduler starts trying to perform reads again, resulting in very few read requests being executed. This seems a reasonable explanation, but it also seems like a massive drawback to using writeback cache on an system with non-trivial write loads. I've been searching for discussions around this all afternoon and found nothing. What am I missing?

    Read the article

  • Rename Active Directory domain following Windows 2000 -> 2008 migration.

    - by ewwhite
    I'm working with a site that needs an internal DNS domain rename. It currently has a DNS name of domain.abc.com and NT name of ABC. I'm trying to get to a DNS name of abctrading.com and NT name of ABCTRADING. Split DNS would be used. The site originally ran from a single Windows 2000 domain controller hosting AD, file, print, DHCP and DNS services. There was no Exchange system in the environment. The 50 client PCs are all Windows XP with a handful of users using roaming profiles. All users are in a single OU and there are no group policy/GPOs. I'm a Linux engineer, but have been trying to guide another group of consultants to reach a more suitable setup. With the help of this group, we were able to move the single Windows 2000 system to a set of Windows 2008 R2 servers separated into domain controller and file/print systems (virtualized). We are also trying to add an Exchange 2010 system to this mix. The Windows 2000 server was demoted and is no longer in the picture. This is the tricky part, as client wants the domain renamed and the consultants aren't quite sure how to get through it without another 32-40 hours of testing/implementation. THey say that there's considerable risk to do the rename without a completely isolated test environment. However, this rename has to be done before installing Exchange. So we're stuck at this point. I'd like to know what's involved in renaming the domain at this point. We're on Windows Server 2008. The AD is healthy now. Coming from a Linux background, it seems as though there should be a reasonable path to this. Also, since the original domain appears to be a child/subdomain, would that be a problem here. I'd appreciate any guidance.

    Read the article

  • IPv6 Routing / Subnetting

    - by nappo
    Recently I have installed Citrix Xen Server 6.2 on a machine. My Provider (Hetzner) gave me the IPv6 Subnet 2a01:4f8:200:xxxx::/64. Followed an article in the providers wiki (1) i got it working and can assign IPs to my guests (CentOS). However i can't assign a second IP to a single guest - it will result in a timeout. I'm not very familiar with IPv6 routing / subnetting - any help or tips for further troubleshooting is welcome! My Setup: XenServer 6.2 IPv6: 2a01:4f8:200:xxxx::2/112 ip -6 route: 2a01:4f8:200:xxxx::/112 dev xenbr0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit 0 fe80::1 dev xenbr0 metric 1024 mtu 1500 advmss 1440 hoplimit 0 default via fe80::1 dev xenbr0 metric 1024 mtu 1500 advmss 1440 hoplimit 0 Guest 1 IPv6: 2a01:4f8:200:xxxx::3/64 IPv6: 2a01:4f8:200:xxxx::4/64 ip -6 route: 2a01:4f8:200:xxxx::/64 dev eth0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit 4294967295 fe80::/64 dev eth0 proto kernel metric 256 mtu 1500 advmss 1440 hoplimit 4294967295 default via fe80::1 dev eth0 metric 1 mtu 1500 advmss 1440 hoplimit 4294967295 Guest 2 IPv6: 2a01:4f8:200:xxxx::5/64 Guest 1 IPv6 is working fine, Guest 2 too. As suggested by the wiki article (1) i split my /64 network into a /112. Is it right to set the host /112 and the guests /64? Why is that?

    Read the article

  • SQL Server replication and load balance

    - by Ahmed Galal
    I'm running a web service that serves a mobile app on IIS 8 and SQL Server 2014, my service has a massive load and i'm trying to improve performance, most of the load is happening on SQL. i don't think i have a bottleneck, my processor and ram is up to the max and i think my code is not that bad, am already using memcached and other stuff to avoid hitting SQL too much. i know i can always upgrade the server hardware but i already have a spare server that i would like to use, so i was thinking to split the SQL load on the 2 servers. What i was thinking of is to setup replication on the other server and do some load balancing, but am not sure how to do the load balance. I know i can adjust my code to hit the other server for some queries but i was hoping to find a solution that avoid changing my code. So my question is, What are the ways of doing load balancing between 2 SQL servers ? I would appreciate suggestions or best practices or some directions. Thanks.

    Read the article

  • Strange battery behavior on laptop

    - by EpsilonVector
    My laptop is behaving rather strangely lately, and I was hoping to get some idea as to what may be causing such symptoms. The problem: When charging, very minute or so it loses connectivity with the AC adapter for a split second, and regains it back immediately. When this happens the little light that indicates the computer is plugged in does flicker off and back on, but I checked the adapter by replacing the battery on my laptop, and this indeed solves the problem, so it is probably the battery which is at fault, not the adapter (I also tried to move the adapter's wire around just to make sure it had nothing to do with torn wires). I suppose that the obvious solution is to get a new battery, but as far as battery defects go- this is a rather strange one; it loses connection with the adapter, but still powers the computer, and changing the power setting to a balanced plan (was maximum performance) seem to have solved the problem too. Is there a chance this is not simply the battery, but some kind of other electronic defect? And if not, what can cause it to behave so strangely? PS I tried to recalibrate it- didn't help.

    Read the article

  • Which upgrade path for disk IO bound postgres server?

    - by user41679
    Hi all, We currently have a Sun x4270 with 2xquad core Xeon Nehalmen 2.93ghz cores (16 threads), 72 gig of ram and 16 x 10k SAS disks split between the os raid 1, a partition for the Write Ahead Logs which is raid 10 and a partition for the database tables and indexes which is also raid 10, all xfs. I'm currently evaluating which path to go down in terms of upgrades. We'll be sharding the DB at some point soon, but for now I need to focus on hardware upgrades specifically. The machine is not CPU or memory bound at all at the moment, just IOWait is become an issue. The machine is mostly write access as we have a heavy caching layer. We're seeing about 300 write IOPS average on both the database partitions. We don't have any additional storage infrastructure like a Fiber Channel or ISCSI network. Budget isn't too much of a concern, something inline with the size of this server (i.e no $1m IBM machines) Space is ok on the DB side of things, we're running out obviously but there's also some reduction we can do. Additional space would be good though. My current thoughts are either: * ISCSI SAN, possible with 10Gbit network that has solid state acceleration. * FusionIO card / Sun F20 card (will the FusionIO card work in the Sun box? * DAS shelf (something like this http://www.broadberry.co.uk/das-direct-attached-storage-servers/cyberstore-224s-das) which a combination of 15k sas disks and some Intel X25-E drives for DB indexes etc) what would I need to put in the x4270 to add a DAS shelf? I think it's a SAS HBA card, do I have to use Sun's own card or will any PCI Express card work? Anything else??? what would you guys do from your experience? I appreciate it's a lot of questions, but I haven't expanded a DB machine for a number of years and the landscape has changed dramatically since then! Any advice or feedback would be very much appreciated. Let me know if there's anything else I can clarify. Thanks in advance!

    Read the article

< Previous Page | 348 349 350 351 352 353 354 355 356 357 358 359  | Next Page >