Search Results

Search found 1506 results on 61 pages for 'ben rokah'.

Page 15/61 | < Previous Page | 11 12 13 14 15 16 17 18 19 20 21 22  | Next Page >

• Sending USR2 to mongrel_rails sometimes results in an “Address already in use” on the restart

  - by Ben

  We have a rolling-restart mode for our mongrel cluster that sends a USR2 signal to each running process. This works great, most of the time. But very occasionally the mongrel process will shutdown, and then fail to restart, with the following error: /usr/local/lib/ruby/gems/1.8/gems/mongrel-1.1.5/bin/../lib/mongrel/tcphack.rb:12:in `initialize_without_backlog': Address already in use - bind(2) (Errno::EADDRINUSE) from /usr/local/lib/ruby/gems/1.8/gems/mongrel-1.1.5/bin/../lib/mongrel/tcphack.rb:12:in `initialize' from /usr/local/lib/ruby/gems/1.8/gems/mongrel-1.1.5/bin/../lib/mongrel.rb:93:in `new' from /usr/local/lib/ruby/gems/1.8/gems/mongrel-1.1.5/bin/../lib/mongrel.rb:93:in `initialize' from /usr/local/lib/ruby/gems/1.8/gems/mongrel-1.1.5/bin/../lib/mongrel/configurator.rb:139:in `new' from /usr/local/lib/ruby/gems/1.8/gems/mongrel-1.1.5/bin/../lib/mongrel/configurator.rb:139:in `listener' Looking though the mongrel source, the USR2 handler calls a synchronous stop on the running server, so it ought to block until the socket has been released. Has anyone seen this error? Does anyone have any ideas what might cause it? (I asked this question over on StackOverflow initially, but thought it might be more appropriate here)

  Read the article

• What does %st mean in top?

  - by Ben

  Here is an example from my top: Cpu(s): 6.0%us, 3.0%sy, 0.0%ni, 78.7%id, 0.0%wa, 0.0%hi, 0.3%si, 12.0%st I am trying to figure out the significance of the %st field. I read that it means steal cpu and it represents time spent by the hypervisor, but I want to know what that actually means to me. Does it mean I may be on a busy physical server and someone else is using too much CPU on the server and they are taking from my VM? If I am using EBS could it be related to handling EBS I/O at the hypervisor level? Is it related to things running on my VM or is it completely unaffected by me?

  Read the article

• Setting CATALINA_OPTS for tomcat6 on windows doesn’t work.

  - by Ben

  (I've copied this from Stack Overflow here, after someone suggested I post the question here) Hi, I'm trying to setup Tomcat6 to work with JMX on Windows Vista 64. To do that I need to pass the parameters below to Tomcat6. What I do in command prompt. (that doesn't work) set CATALINA_OPTS="-Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.port=9898 -Dcom.sun.management.jmxremote.ssl=false -Dcom.sun.management.jmxremote.authenticate=false" tomcat6.exe What I do that does work (but causes other problems) java -Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.port=9898 -Dcom.sun.management.jmxremote.ssl=false -Dcom.sun.management.jmxremote.authenticate=false -jar bootstrap.jar It seems as if tomcat is just ignoring the environment variable CATALINA_OPTS. Am I doing something wrong? I've also tried to edit catalina.bat and define the variable CATALINA_OPTS there. No success. (tried adding the parameters to JAVA_OPTS too, no success either) Thanks in Advance!!

  Read the article

• Does likewise-open > version 5.4 contain CIFS support?

  - by Ben Andken

  I'm trying to get the CIFS server working in likewise-open. I've found this set of instructions and everything seems to work until I try to connect ([url]http://www.likewise.com/resources/documentation_library/manuals/cifs/likewise-cifs-smb-file-server-guide.html#id2765992):[/url] 1.6. Build and Configure a Standalone Likewise-CIFS Server This section demonstrates how to build and configure a standalone instance of Likewise-CIFS from the command line. The following procedure assumes that you want to set up Likewise-CIFS on a Linux server to share files with Windows computers in a network without Active Directory. This procedure also assumes you know how to build Linux applications from their source code and then install them. Download Likewise-CIFS from its open source git location: $ git clone git://git.likewiseopen.org/ Download, build, and install the following tools. The tools listed are known to work, but earlier or later versions might work as well. Also, instead of downloading the tools, you might be able to install them on your platform with apt-get or some other means. http://ftp.gnu.org/gnu/autoconf/autoconf-2.65.tar.gz http://ftp.gnu.org/gnu/automake/automake-1.9.6.tar.gz http://ftp.gnu.org/gnu/libtool/libtool-2.2.6a.tar.gz http://pkgconfig.freedesktop.org/releases/pkg-config-0.23.tar.gz gcc --version 3.x or greater Build Likewise-CIFS: $ cd likewise-open $ build/mkcomp --debug all Install Likewise-CIFS: $ sudo su $ cd staging/install-root $ tar cf - . | (cd / && tar xvf -) Make sure Samba is not running: $ /etc/init.d/smb stop Make sure SELinux is either disabled or set to permissive. Make sure the ports required by Likewise are open. For a list of ports that Likewise uses, see the Likewise Open Installation and Administration Guide. Configure Likewise Open: $ /etc/init.d/lwsmd start $ for i in /etc/likewise/*.reg; do /opt/likewise/bin/lwregshell upgrade $i; done $ /etc/init.d/lwsmd stop $ /etc/init.d/lwsmd start $ /opt/likewise/bin/lwsm start srvsvc $ /opt/likewise/bin/domainjoin-cli configure --enable nsswitch Add a user account to the local Likewise provider database. In the following example, substitute the account name that you want for newuser. $ /opt/likewise/bin/lw-add-user --home /home/newuser --shell /bin/bash newuser Successfully added user newuser Enable the user and set the password: $ /opt/likewise/bin/lw-mod-user --enable-user --set-password newuser New Password: ********** Successfully modified user newuser Look up new user's identity as follows. Substitute the value from the command hostname -s for the hostname. Keep in mind that Likewise truncates a hostname longer than 15 characters to the first 15 characters of the string. % id hostname\\newuser uid=2000(HOSTNAME\newuser) gid=1800(HOSTNAME\Likewise Users) groups=1800(HOSTNAME\Likewise Users) context=system_u:system_r:unconfined_t:s0 Make a CIFS directory for the user: mkdir /lwcifs/newuser chown 2000:1800 /lwcifs/newuser From a Windows computer, map the Likewise-CIFS drive share: Computer->Map Network Drive... Folder: \\IP_hostname\c$ Click "Finish" Username: hostname\newuser Password: user_password The last step fails when I try to connect. I've tried with Windows XP Pro and Windows 7 Pro. The rest of the directions only appear to work for version 5.4 (the one that shipped with 10.04). For 12.04, version 6.1 is the only one available and it doesn't appear to have the srvsvc module mentioned in these instructions. Is CIFS support dropped in the 6.1 version of likewise-open?

  Read the article

• SharePoint 2010 Search Error 0x800703fa

  - by Ben

  We have migrated from SharePoint 2007 to 2010. Everything appears to be working correctly except for an intermitent error with search. Occastionally search results will crash for all of our sites and when we look up the coorliation id we get the following error: Exception when fetching results: System.ServiceModel.FaultException`1[System.ServiceModel.ExceptionDetail]: Illegal operation attempted on a registry key that has been marked for deletion. (Exception from HRESULT: 0x800703FA) (Fault Detail is equal to An ExceptionDetail, likely created by IncludeExceptionDetailInFaults=true, whose value is: System.Runtime.InteropServices.COMException: Illegal operation attempted on a registry key that has been marked for deletion. (Exception from HRESULT: 0x800703FA) at System.Runtime.InteropServices.Marshal.ThrowExceptionForHRInternal(Int32 errorCode, IntPtr errorInfo) at Microsoft.Office.Server.Search.Query.KeywordQueryInternal.Execute() at Microsoft.Office.Server.Search.Query.QueryInternal.Execute(QueryProperties properties) at Microsoft.Office.Server.Search.Administration.SearchServiceApplication.Execute(QueryProperties properties) at SyncInvokeExecute(Object , Object[] , Object[] ) at System.ServiceModel.Dispatcher.SyncMethodInvoker.Invoke(Object instance, Object[] inputs, Object[]& outputs) at System.ServiceModel.Dispatcher.DispatchOperationRuntime.InvokeBegin(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage5(MessageRpc& rpc) We reset IIS and the problem resolves itself for a while. Has anyone come across a perminant fix for this?

  Read the article

• Determine formatting of SD card in Ubuntu

  - by ben

  How do I, in Ubuntu, determine what the formatting (e.g FAT etc.) of an SD card is? Thanks.

  Read the article

• Combine multiple unix commands into one output

  - by Ben McCormack

  I need to search our mail logs for a specific e-mail address. We keep a current file named maillog as well as a week's worth of .bz2 files in the same folder. Currently, I'm running the following commands to search for the file: grep [email protected] maillog bzgrep [email protected] *.bz2 Is there a way combine the grep and bzgrep commands into a single output? That way, I could pipe the combined results to a single e-mail or a single file.

  Read the article

• Endian Destination NAT

  - by Ben Swinburne

  I have installed Endian Community Firewall 2.3 and am clearly misunderstanding/doing something wrong with it. I'm trying to create some destination NAT rules to allow incoming connections to various services within the network. Router - RED I/F - x.x.x.x Router - GREEN I/F - 192.168.11.253 ECF - RED I/F - 192.168.11.254/24 ECF - GREEN I/F - 192.168.12.254/24 Target server - 192.168.12.1 Please ignore the haphazard choice of subnets and addresses- I'm trying to quickly plop Endian into an existing network before a complete rework in 6-12 months so for now. Everything works except destination NAT, so outgoing connections are fine, the routes between the two subnets are OK etc. I want to create various incoming NATs but let's take for the sake of argument, SMTP port 25 from the Internet to Target server 192.168.12.1. I've tried almost every combination of options in the Destination NAT section to achieve this and clearly am doing something wrong. I suspect my confusion must be somewhere in the Access From and/or Target section. The rest seems OK Filter Policy = Allow Service = SMTP Protocol = TCP Port = 25 Translate to type = IP DNAT Policy = NAT Insert IP = 192.168.12.1 Port Range = 25 Enabled = Checked Position = First I can't work out what I'm doing wrong, or am I doing it right and it's just not working!? Any help would be greatly appreciated.

  Read the article

• New 3TB HDD, can see full 2.7TB in Linux and Windows, but shows up as 801.6GB in BIOS

  - by Ben Lee

  I recently purchased a Seagate Barracuda 3TB drive (ST3000DM001). After installing it, my BIOS recognized it but reported the size as 801.6gb. I went ahead and booted into Linux anyway (Ubuntu 11.10 64-bit). Linux saw it as a 2.7TB. Following some online instructions (don't have the link handy, unfortunately), it looks liked converting this drive to GPT was recommended. So I used gparted to do that, then formatted it to NTFS also using gparted. (I'm using NTFS because my machine is daul-boot and I want to have access to the drive in Windows too). I rebooted to Windows (Windows 7 64-bit), and Windows also sees the drive with 2.7TB free. Everything seems to be working fine. The only issue is that my BIOS is still reporting the drive as 801.6GB. My motherboard is an ASRock 770 Extreme3 and BIOS is the latest version. Since everything seems to be working with the new drive anyway, I'm hoping that the fact that the BIOS is reporting the wrong size is not an actual problem. But honestly, I don't really know. Anyone out there more familiar with this know if this could potentially cause any problems in the future? Any way to get the BIOS to report the correct size?

  Read the article

• Migrating Email to new hosting.

  - by Ben C

  I've made a site for a charity, and now have to move hosting for them. They have 5 or so email addresses on their current hosting account, which will of course need to move too. What's the best way to migrate their email addresses to the new server without too much hassle for them? They use POP3, so should I just create the account on the new server and then get them to update their settings? That won't remove their old emails from Outlook Express, will it?

  Read the article

• Diagnosing laptop charging issues

  - by Ben

  My HP dv6000 laptop will not run on battery power. It will only run while plugged into the wall. I have already replaced the battery, but it still will not operate on battery power. The problem began slowly. The original battery's life got shorter and shorter, until eventually it would only last ~45 seconds before going into hibernation. Finally, the battery would not work at all, so I threw it out and just used it without a battery. I bought a replacement recently.... but the problem persists. When I connect the new battery to the system, the "charging" light turns on for about 30 seconds, then goes dark. I'm assuming there's something wrong with the internal charging circuitry? If this is the case, how would I go about fixing this, short of buying a new motherboard? If anyone could give me some assistance in confirming/rejecting my diagnosis I'd be very appreciative! :)

  Read the article

• TicTac Photo and Windows 7

  - by Ben

  Hello, My wife has been creating a tictac photo album. I had to upgrade to windows 7 as i had enough of Vista so i backed up the tic tac photo file and the photos to an external hard disk and performed a fresh install of win7. Now here is the problem. TicTacPhoto says it can find the photos in the album. The locations were as follows: Vista: C:\Users\Kelly\Pictures Win 7 C:\Users\Kelly\My Pictures When i try to create a Pictures folder under Kelly it popups a message about merging the two folders and simply moves the pictures to the My Pictures folder. Does anyone know a way to make a foler called pictures so i can eliminate the file path problem and then try again with tic tac photo support to get them to fix my file. My wife is going to kill me as its our wedding album and she has spent upwards of 30hrs designing it and me upgrading to win 7 means its all my fault. She does not understand file paths etc. Im going to try and open the album file in a text editor and see if i can see anything but thought i would ask here as well. Any help appreciated.

  Read the article

• Task Sequence boots to logon screen instead of task sequence mode

  - by Ben M.

  I'm running a task sequence, and so that users don't accidentally interfere, I have the task sequence reboot to currently installed operating system, which as I understand, is supposed to boot to a sort of single user mode and all that shows is task sequence progress. However, this does not happen, it boots up like normal and comes to the logon screen and the task sequence runs in background. How can I adjust this behavior to the desired result?

  Read the article

• MacBookPro running Windows 7: accidental trackpad input is driving me mad

  - by Ben Hammond

  I am running Windows7 Professional 64bit on a 2010 MacBookPro using BootCamp 3.1. I am using an external trackball. When I am typing, I accidently brushing the trackpad and accidently overtyping randomly selected pieces of text. Which is driving me mad. I have tried to install TrackPad++, but I could not get the Trackpad++ control panel to recognise that the driver software was installed. I tried TrackPad Magic, but although it gives me a system tray icon telling me it is working, but it does not appear to disable the track pad. A quick Google implies that there should be an option in the BootCamp Control Panel 'Ignore accidental Input while typing'. But I can't see one of those in my BootCamp Control Panel. Am I looking in the wrong place? Is this feature 32bit only? Is there anything else I should try?

  Read the article

• High load average, low CPU and IO (Centos 5.7)

  - by Ben

  A Drupal 7 site with CiviCRM, after running smoothly for a year on a 1&1 VPS suddenly became unresponsive. Now pages eventually load, but can take more than a minute. Looking at resource use in Virtuozzo, the load average carries a warning, and has remained above 1. While I understand this isn't particularly high, this is a change from when the site was working. Here is a typical snapshot of top: top - 03:10:32 up 3:21, 1 user, load average: 1.16, 1.22, 1.30 Tasks: 43 total, 1 running, 42 sleeping, 0 stopped, 0 zombie Cpu(s): 0.1%us, 0.1%sy, 0.1%ni, 99.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 2097152k total, 1015112k used, 1082040k free, 0k buffers Swap: 0k total, 0k used, 0k free, 0k cached CPU idle level never seems to go much below 70%. wa is virtually always at 0. There appears to be lots of free memory. And here is some vmstat output, again showign wa at 0, plenty of free memory, and an idle CPU: procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------ r b swpd free buff cache si so bi bo in cs us sy id wa st 2 0 0 1100872 0 0 0 0 2783 23672 0 538 1 0 99 0 0 1 0 0 1100872 0 0 0 0 0 16 0 101754 0 0 100 0 0 0 0 0 1100872 0 0 0 0 0 17 0 103133 0 0 100 0 0 0 0 0 1100872 0 0 0 0 0 1 0 102080 0 0 100 0 0 1 0 0 1100872 0 0 0 0 0 6 0 99881 0 0 100 0 0 0 0 0 1100872 0 0 0 0 0 1 0 105187 0 0 100 0 0 I've spoken to 1&1 but they don't have any ideas as to what could be causing the high load average. Instead they suggested an upgrade :) I've looked for processes that might be causing this, examined MySQL showprocesslist, and restarted the container with no result. Does anyone have more troubleshooting suggestions or insights?

  Read the article

• Any ideas out there as to how the data can be recovered from an SSD?

  - by ben

  A friend had some form of catastrophic failure on an HP mini 1000, unbootable. Of course there was data that wasn't backed up. I've removed the SSD and hooked it up to a ZIF 40 enclosure but can not seem to get the drive to be recognized in Windows 7. In Disk Management it displays as present, but uninitialized. Attempting to initialize it presents an error Virtual Disk Manager - "The device is not ready". There is scant information on MIE (the custom OS), so I'm not even sure what kind of file system I'm dealing with. In any case, if the filesystem is indeed some flavor other than FAT or NTFS, is this error consistent with that? Are there any creative ideas out there as to how the data can be recovered? Update: Thanks for all the suggestions! I hadn't even considered running a live cd. Unfortunately no luck with Ubuntu (live cd) or explore2fs. The zif connection seems ok (color coded green led for proper connect, orange for not). The drive can't be initialized and therefore can't be formatted, so I guess there may be some real damage. Probably needs to head to a specialist. Thanks again for the feedback, much appreciated.

  Read the article

• How can Bonjour be setup to function over a VPN connection using Mac OS X — Mountain Lion Server?

  - by Ben Coppock

  I purchased Mountain Lion Server for our office thinking that Bonjour would automatically enable any computers connected via VPN to see all computers and applications (such as Bento) running on the office network. The hope was that those of us working at home would feel just like we were in the office, with all network services working transparently over the VPN connection. However, I see that Bonjour (aka mDNS) is not enabled to work over the VPN by default. Can I configure Mountain Lion Server to automatically pass Bonjour traffic over the VPN? Is there any reason not to do this?

  Read the article

• Windows somehow got convinced that a file is a folder. How do I change it back?

  - by Ben Collins

  I recently moved an external drive from my usb-enabled router to my desktop machine, and ran into some permissions-related issues. A number of files were giving me errors when I tried to take ownership or set permissions, and in all my fiddling on a particular file, it somehow got switched to a folder. Anyone have any idea how this might have happened, and how to flip it back? Here's a screenshot:

  Read the article

• Do i need a dedicated server for load balancing?

  - by Ben

  I'm completely new to the concept of load balancing so i hope this question isn't a "stupid question" because i've been searching around and im having a hard time understanding this. So to my understanding, in order to load balance, i need a separate machine with an ip address i can direct all traffic to. I initially thought i needed to rent 3 dedicated servers, one for load balancing and the other two as backend servers. Would a dedicated server be too much for a load balancer or do hosting companies have special types of computers for that process? Then i read somewhere else that i can install a load balance software in both of the two servers and configure it in a way that doesn't require me to rent another machine/dedicated server for load balancing. So im a bit confuse on how to actually implement a load balancer and whether or not i need a dedicated server for the sole purpose of acting as a load balancing machine. Also, i was recommended to use HAproxy so i'll be heading that direction for load balancing.

  Read the article

• Can't Install Windows 2008 R2 on Lenovo Laptop With SSD

  - by Ben

  I am trying to install Windows Server 2008 R2 on a new Lenovo X201 or T410 laptop. Setup halts with the following pop-up: A required CD/DVD device driver is missing. If you have a driver floppy disk, CD, DVD, or USB flash drive, please insert it now. Note: If the windows installation media is in the CD/DVD drive you can safely remove it for this step. The CD drive is obviously working, as it's booting from CD to get to this point. The only thing I can think is that it is to do with the fact they have SSD disks in - but that's just a guess. (Edit - One extra thing that may or may not be relevant: it's the 64 bit version of Server 2008 R2)

  Read the article

• After Switching from Apache to Nginx, HTML Truncated

  - by Ben Buras

  Our site was running fine with Apache, but I am switching us over to Nginx (with php-fastcgi) to be more scalable. Everything is working fine except our HTML gets truncated sometimes (usually on large pages). Live server running Apache: http://bros4bros.com/index.php?action=browse_members&members_page=50 (WARNING: half naked guys!) New server running Nginx: http://50.56.238.208/index.php?action=browse_members&members_page=50 I have tried modifying these settings with no avail: max_execution_time fastcgi_send_timeout

  Read the article

• Persistent Issues on small business network using Cisco 871W and Catalyst Express 500

  - by Ben Campbell

  Being the most qualified (read: still not qualified) to solve our persistant network issues, I've turned to serverfault for guidance. I've done some searching, reading related documentation on cisco.com and tried a bit of troubleshooting. Here is the config: 100mb synchronous connection from a business internet provider (tested multiple times at 100meg at the source) Cisco 871W wireless point & router is where the WAN connection starts (this serves all our wireless). The only wired connection in the 871W is the Catalyst switch listed below. Cisco Catalyst Express 500 (24TT) is where all the wired connections terminate. About 20 Windows workstations and servers (AD/Webservers only). Some services in EC2 including mail and other web servers/apps. I've been TOLD cabling internally should be gigabit-ready. Here are the problems: generally slow download rates from the internet to the desktop/laptop frequent "page cannot be displayed" errors in browsers-sometimes 3 or 4 reloads are necessary... often times CSS wont load or other content requiring the browser to connect to a different server. slow speed within the LAN from workstation to workstation copying files. I would expect extremely fast data transfer workstation to workstation / server to workstation in this simple network. Several things I need to admit: I'm not primarily a network guy. Funding is relatively low, I need to be the guy that finds the solution. I understand most of the terminology and most of the technology. Implementation is where I fail due to lack of experience. Getting to the point: I'm wondering whether experienced network admins think that our small network should be sufficiently served with our current hardware if configured properly... or if we should purchase new equipment and start fresh? If starting fresh is the plan, whatever that new equipment may be is a likely different question entirely. If I haven't provided enough information, I will happily do some troubleshooting and update with the results. I have experience using wireshark and some other tools. Please let me know what you think would be most helpful and thanks in advance. EDIT: I forgot to add that the Cisco applicance will not finish loading the SDM Express console. It hangs every time at the "populating modules... DHCP". It eventually crashes and closes. I've rebooted the hardware and this still happens.

  Read the article

• Postfix send remotely for X account

  - by Ben Scobie

  I setup Postfix a few days ago to deliver email bound for my host name remotely instead of locally as I want email sent to my remote [email protected] account without trying to be sent locally. However, this causes a problem as it tries to send local bound mail(root) to my remote server as well ([email protected]). I would like E-mail bound for [email protected] to be sent to the remote server(GMail Apps in this case) and everything else sent locally(using the alias file perhaps), is this possible? My current configuration: http://pastebin.com/htwSQB61

  Read the article

• Ubuntu only boots with USB plugged in

  - by Ben

  I'm new to the Linux world so please bear with me! :-) I installed Ubuntu via USB drive onto my hard drive. If I boot the PC without the usb drive I used, Ubuntu will not load. After booting I can unplug without any consequences. I looked on the hard drive and there is a boot folder. On the USB drive, this is the tree contents: /media/disk$ tree . |-- adtext.cfg |-- boot.cat |-- f10.txt |-- f1.txt |-- f2.txt |-- f3.txt |-- f4.txt |-- f5.txt |-- f6.txt |-- f7.txt |-- f8.txt |-- f9.txt |-- initrd.gz |-- isolinux.bin |-- isolinux.cfg |-- ldlinux.sys |-- linux |-- menu.c32 |-- menu.cfg |-- po4a.cfg |-- prompt.cfg |-- splash.png |-- stdmenu.cfg |-- syslinux.cfg |-- text.cfg |-- ubnfilel.txt |-- ubnpathl.txt `-- vesamenu.c32 Am I correct in my assumption that the boot aspect is associated to the USB drive? If so, how do I get it to boot without the USB? I'm guessing copying into some location and modifying grub?

  Read the article

• Need to increase nginx throughput to an upstream unix socket -- linux kernel tuning?

  - by Ben Lee

  I am running an nginx server that acts as a proxy to an upstream unix socket, like this: upstream app_server { server unix:/tmp/app.sock fail_timeout=0; } server { listen ###.###.###.###; server_name whatever.server; root /web/root; try_files $uri @app; location @app { proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header X-Forwarded-Proto $scheme; proxy_set_header Host $http_host; proxy_redirect off; proxy_pass http://app_server; } } Some app server processes, in turn, pull requests off /tmp/app.sock as they become available. The particular app server in use here is Unicorn, but I don't think that's relevant to this question. The issue is, it just seems that past a certain amount of load, nginx can't get requests through the socket at a fast enough rate. It doesn't matter how many app server processes I set up, it doesn't even matter what the app is (tried it with a dummy app with just a single endpoint that returned an empty page with status 404). The bottleneck seems to be the socket, not the app. I'm getting a flood of these messages in the nginx error log: connect() to unix:/tmp/app.sock failed (11: Resource temporarily unavailable) while connecting to upstream Many requests result in status code 502, and those that don't take a long time to complete. The nginx write queue stat hovers around 1000. Anyway, I feel like I'm missing something obvious here, because this particular configuration of nginx and app server is pretty common, especially with Unicorn (it's the recommended method in fact). Are there any linux kernel options that needs to be set, or something in nginx? Any ideas about how to increase the throughput to the upstream socket? Something that I'm clearly doing wrong? Additional information on the environment: $ uname -a Linux app1 3.2.0-24-generic #39-Ubuntu SMP Mon May 21 16:52:17 UTC 2012 x86_64 x86_64 x86_64 GNU/Linux $ ruby -v ruby 1.9.3p194 (2012-04-20 revision 35410) [x86_64-linux] $ unicorn -v unicorn v4.3.1 $ nginx -V nginx version: nginx/1.2.1 built by gcc 4.6.3 (Ubuntu/Linaro 4.6.3-1ubuntu5) TLS SNI support enabled Current kernel tweaks: net.core.rmem_default = 65536 net.core.wmem_default = 65536 net.core.rmem_max = 16777216 net.core.wmem_max = 16777216 net.ipv4.tcp_rmem = 4096 87380 16777216 net.ipv4.tcp_wmem = 4096 65536 16777216 net.ipv4.tcp_mem = 16777216 16777216 16777216 net.ipv4.tcp_window_scaling = 1 net.ipv4.route.flush = 1 net.ipv4.tcp_no_metrics_save = 1 net.ipv4.tcp_moderate_rcvbuf = 1 net.core.somaxconn = 8192 net.netfilter.nf_conntrack_max = 131072

  Read the article

< Previous Page | 11 12 13 14 15 16 17 18 19 20 21 22  | Next Page >