Search Results

Search found 7175 results on 287 pages for 'job hunting'.

Page 226/287 | < Previous Page | 222 223 224 225 226 227 228 229 230 231 232 233  | Next Page >

• Nginx, HAproxy, Unicorn, Rails and Node settings

  - by Julien Genestoux

  Our application is currently only a "regular" web app, with no fancy things like streaming HTTP or websockets. It's mostly a Rails app, served by a few (20 on 2 machines) Unicorn workers, proxied by a venerable nginx server which deals with load balancing. This has been working quite well for the past year and the app now serves between 400 and 800 requests per second at any point during the day. We're soon releasing 2 new APIs, which are both served by a Node application : a websocket one, as well as a long polling HTTP one. (the fancy thing like the Twitter streaming API where HTTP connections never end). They both use the same port on node and since the node app is stateless, we can certainly deploy a few of them to handle the traffic. The app (node) is now deployed in 5 instances and are now listening on 5 different 'private' ports on the same host. We need to put something in front of them to load balance, but also something that is able to deal with sockets (either websocket or HTTP streaming) which are intended to stay 'up' for days. The question is then : what? I read somewhere that HAProxy does a better job than Nginx at this. What do you recommend?

  Read the article

• Cannot start Postgres daemon after installing with Yum

  - by Sean the Bean

  I was trying to install Postgres 9.1.4 on Fedora 17 using Yum. If I do: sudo yum install postgres-libs sudo yum install postgres sudo yum install postgis All the installs appear to complete successfully (i.e., no errors), but I cannot start the Postgres daemon using: service postgresql initdb Like the official Postgres download guide says to do (http://www.postgresql.org/download/linux/redhat/). The error says Unknown operation initdb. RPM tells me that it installed psql to /usr/bin/, which I confirmed. It turns out that only a few components installed correctly (psql, pg_dump, pg_configure, and a few others), but most are missing (e.g., pg_ctl and postgres). I've tried several different configurations and had several of my coworkers (with more linux experience than me) look at it, but so far nothing has worked. Two of them have also run into similar issues installing Postgres using apt-get on Ubuntu, which makes me think the rpm isn't doing its job. It seems the only solution to build it from source, which is more robust anyway, but of course it takes longer. I'm wondering, though, if anyone else has run into this issue and/or has successfully installed Postgres on either Fedora or Ubuntu using a package manager like yum or apt-get? Is the rpm broken?

  Read the article

• Setting MSN or Yahoo! Messenger status to Invisible or Offline when idle for an hour

  - by Jian Lin

  Where, or how, do I set it up in MSN Messenger or Yahoo! Messenger to automatically switch my status to either "Invisible" or "Offline" when idle for a half hour, or an hour? I know how to set my status as "Away" or "Busy" after 10 minutes, but can't seem to find a way to set the offline status options without manual intervention. Back story As a software developer, I am very used to turning the computer on for the whole day and not turning it back off. (For example, checking email for urgent fixes, fix issue and push to web server). It's not even turned off when heading to sleep in case I might find it hard to fall asleep and come back to check on the computer. Or to have it there ready in the morning to check that everything is okay. If I'm seen as being online for 24 hours of a day, some people see me as weird. Their perception of my value decreases as I'm always there (hard to get = high value; always there = low value). Leaving it on makes everyone in my contacts list think I have nothing better to do all day than sit in front of the computer. Even though it's my job and I do admittedly spend more time online than other people. That's why I'd like to find a way to set my status as Invisible or Offline.

  Read the article

• Hudson deploy specific git revision

  - by brad

  I'm using hudson to auto-deploy my Rails app to heroku. In my main build job I pull from a Git repo (hosted using gitosis on the same machine), master branch with the following: URL of repository: /home/git/repositories/my_app.git Name of repository: origin Refspec: +refs/heads/master:refs/remotes/origin/master Branches to build: master Then, assuming all tests pass, I want to kick off a new build that is the deploy to Heroku. I can't however figure out how to get that deploy build to checkout the particular revision that this build was using. I understand there's a parameterized trigger plugin that would allow me to pass this revision number, but I don't know how I can tell hudson to checkout this particular revision on the deploy build. I'm pretty sure this just has to do with my limited knowledge of git, but where in the hudson git config's is there an option to checkout a particular revision? Otherwise, I could have many commits happen whilst a build is happening, and when it kicks off a deploy build, that deploy build would just check out the HEAD of the branch, which may not be the same as the code that was pushed that triggered this build. I don't fully understand why I have a refspec in Hudson, then also specify a branch to build, I thought this was the same thing. Can refspec somehow specify the revision number? How would this be referenced if it was passed through with the parameterized trigger plugin? (I've never used that plugin, but someone else recommended it as a way to pass in vars to a new build, if there's another way I'm all ears)

  Read the article

• How can I make Internet Explorer 6 render Web pages like Internet Explorer 11?

  - by gparyani

  Now, I know that this may seem like a bad question in that I can just upgrade to Internet Explorer 8, but I am sticking with IE6 in that IE8 removes valuable features, like the ability to save favorites offline and the fact that a file path turns into a Windows Explorer window and typing a Web address into Windows Explorer changes it into an IE window. I know that Internet Explorer 6 does a really bad job at rendering some pages. I know of the Google Chrome Frame extension that brings Chrome-style rendering into IE, but that will soon be discontinued. So, I tried another thing: I know that C:\Windows\System32\mshtml.dll contains the Trident rendering engine that is used by IE, so I tried something: I first backed up the original file by renaming it on Windows XP to mshtml-old.dll, then I tried to copy in the DLL from a computer running Windows 7 with Internet Explorer 10. I noticed that, after copying, the system had replaced the new DLL with the old one, but left the one I backed up intact. Is there any way I can get the system to not replace the DLL like that so that I can transfer in IE11's mshtml.dll into Windows XP and make IE6 render like IE11? I'm looking for an answer that describes how to tweak my system to make IE6 render like IE11 (or IE10), not one that tells me to upgrade IE or install another browser. I don't care how tedious the method is, just as long as it works.

  Read the article

• Search inside of text files

  - by Matt

  So here is the situation. I currently run a mail server for my small non-profit company. My mail server (Merak Mail Server) keeps logs in .log files and mail as .tmp files. Essentially these are just text files that are kept on the server. Problem is that when I put text into the "Containing text" field on Windows Explorer, it always misses the files and tells me no results returned. Then when I search the files one by one (painful at best), I find the files I need. Do I not understand the search feature well enough, or maybe I have it indexing wrong. I really don't care what I need to use to search the files, even a third-party app is fine with me, I just want to type an email address into a box and search all of my log files or email files and find out which one I am looking for. It can be Windows Search or something else, as long as I can find a way to get the job done I will be happy. Pay solutions are fine as well. Thanks everyone in advance.

  Read the article

• Nginx and automatic updates

  - by Desmond Hume

  I'm on Ubuntu 12.04.1 with unattended-upgrades configured for automatic security updates, and I installed Nginx by first adding deb http://nginx.org/packages/ubuntu/ lucid nginx deb-src http://nginx.org/packages/ubuntu/ lucid nginx to /etc/apt/sources.list file, just as was suggested by the official wiki, and then by sudo apt-get update sudo apt-get install nginx which installed Nginx with all the standard modules. But now I think I could make good use of one or two of the Nginx optional modules, like the gzip precompression module or some security-related one. So far, I see two ways of adding an optional module to Nginx, one is compiling and installing from the source code and the other is described in this article. So, which of the ways should I choose so that automatic updates still run for and apply to Nginx and its optional modules? Or should I create a cron job with a command/script specific for Nginx instead of using unattended-upgrades utility? Can I choose between volume updates and security-only updates to be automatically applied to the standard and optional modules? And finally, is there a possibility to automatically update Nginx's modules on the fly (without any connections having been dropped), like the documentation suggests it's possible with sudo kill -USR2 $( cat /run/nginx.pid ) P.S. Actually I'm not certain if unattended-upgrades utility would automatically update the standard modules in the first place, not enough time has passed since Nginx was installed to say for sure.

  Read the article

• Affordable combined Ruby/Rails/Redmine + Subversion hosting?

  - by Pekka

  I'm a self employed web developer and after nine years of hard work, I'm looking to become a bit more "vagrant" starting next year, do some much-needed traveling and a bit and work off and on, making use of one of the greatest advantages of a programming job: The ability to work virtually from everywhere. For that, I am looking for a reliable hosting company I can entrust my code to in the form of a number of Subversion repositories, and an installation of the Redmine project management tool. As my financial situation may vary during traveling, I am looking for something I can pay up front for a year or two, and is obviously not too pricey. I don't care where the company is located, as long as it's trustworthy and solid, meaning it's not likely to go out of business next month. Does anybody know good recommendations? Preferably from own, personal, good experience. I have looked at CVSDude / Codesion and while they are certainly great, they don't offer Redmine of course, and seem to be aiming toward bigger organizations mainly. What I would need: 2-5 Gigs of space minimum, freely distributable between SVN, and Redmine attachments Unlimited number of Subversion projects Access control (team members / checkout-only accounts / etc.) I don't mind configuring the svn settings on file basis myself I need the possibility to map a custom domain to the package that is hosted elsewhere Frequent backups and access to those backups through FTP or other means I have been running my own virtual server for this until now, but I don't want the hassle, especially on the security side, while I may not always have the internet connection to fix problems that may come up.

  Read the article

• How do you do a keyword search the Services.msc (mmc) window in Windows 7?

  - by Warren P

  When you want to run a service, you have very limited capabilities, in all current Windows versions, as far as I can tell. I usually start Services by typing "services.msc" into the Start-Run box, on most versions of Windows, this works. I know how to click the "Name" column in the MMC view of Windows Services. If you know what the first few characters of a service name is, you can usually sort by the name, and type the prefix to scroll the list down (find Windows Search for example). This seems pretty weak to me, so I spent some time searching the interwebs for tools that do a better job of managing services. Usually I have a keyword that I know "fooWare" might be the keyword, and I need to find the (usually badly named) service and start it and stop it. This is often WAY too hard. The best I could do is "NET SERVICES" from the command line, and maybe add a grep in there, but that doesn't list every service, only a few of them. And the MMC snap-in in Win7 now has an Export List button, exporting to csv text file feature which I have used from time to time, to export and then search. I have thought of writing my own tool. I'm hoping a better "service manager" utility exists out there that sysadmins use. I'd like a search box at the top right corner, kind of the same way that the Add-Remove-Programs dialog in Win7 and Vista has a search facility. Does such a services utility exist out there?

  Read the article

• How do I ensure a process is running, even if it kills itself? (it needs to be restarted then)

  - by le_me

  I'm using linux. I want a process (an irc bot) to run every time I start the computer. But I've got a problem: The network is bad and it disconnects often, so I need to manually restart the bot a few times a day. How do I automate that? Additional information: The bot creates a pid file, called bot.pid The bot reconnects itself, but only a few times. The network is too bad, so the bot kills itself sometimes because it gets no response. What I do currently (aka my approach ;) ) I have a cron job executing startbot.rb every 5 minutes. (The script itself is in the same directory as the bot) The script: #!/usr/bin/ruby require 'fileutils' if File.exists?(File.expand_path('tmp/bot.pid')) @pid = File.read(File.expand_path('tmp/bot.pid')).chomp!.to_i begin raise "ouch" if Process.kill(0, @pid) != 1 rescue puts "Removing abandoned pid file" FileUtils.rm(File.expand_path('tmp/bot.pid')) puts "Starting the bot!" Kernel.exec(File.expand_path('./bot.rb')) else puts "Bot up and running!" end else puts "Starting the bot!" Kernel.exec(File.expand_path('./bot.rb')) end What this does: It checks if the pid file exists, if that's true it checks if kill -s 0 BOT_PID == 1 (if the bot's running) and starts the bot if one of the two checks fail/are not true. My approach seems to be quite dirty so how do I do it better?

  Read the article

• Can a pool of memcache daemons be used to share sessions more efficiently?

  - by Tom

  We are moving from a 1 webserver setup to a two webserver setup and I need to start sharing PHP sessions between the two load balanced machines. We already have memcached installed (and started) and so I was pleasantly surprized that I could accomplish sharing sessions between the new servers by changing only 3 lines in the php.ini file (the session.save_handler and session.save_path): I replaced: session.save_handler = files with: session.save_handler = memcache Then on the master webserver I set the session.save_path to point to localhost: session.save_path="tcp://localhost:11211" and on the slave webserver I set the session.save_path to point to the master: session.save_path="tcp://192.168.0.1:11211" Job done, I tested it and it works. But... Obviously using memcache means the sessions are in RAM and will be lost if a machine is rebooted or the memcache daemon crashes - I'm a little concerned by this but I am a bit more worried about the network traffic between the two webservers (especially as we scale up) because whenever someone is load balanced to the slave webserver their sessions will be fetched across the network from the master webserver. I was wondering if I could define two save_paths so the machines look in their own session storage before using the network. For example: Master: session.save_path="tcp://localhost:11211, tcp://192.168.0.2:11211" Slave: session.save_path="tcp://localhost:11211, tcp://192.168.0.1:11211" Would this successfully share sessions across the servers AND help performance? i.e save network traffic 50% of the time. Or is this technique only for failovers (e.g. when one memcache daemon is unreachable)? Note: I'm not really asking specifically about memcache replication - more about whether the PHP memcache client can peak inside each memcache daemon in a pool, return a session if it finds one and only create a new session if it doesn't find one in all the stores. As I'm writing this I'm thinking I'm asking a bit much from PHP, lol... Assume: no sticky-sessions, round-robin load balancing, LAMP servers.

  Read the article

• Bandwidth Suggestion

  - by Campo

  I have been asked to analyze the bandwidth usage of a company and make a recommendation for upgrading their Internet connection(s). Here is the layout 3 DLS lines so it is 3x(6 Down, 1 Up Each) into a load balancer out to the office's network. 30 VOIP phones run on a T1 (1.5 Down, 1.5 Up) The users at the company are heavily uploading. It is my suspicion that the issue in slowdown is being cause by multiple people uploading and others not being able to get requests out for even simple http requests. My initial idea is to get them a fiber line with a 10 down and 10 up. What do others think on this plan? Will that be enough to host their network traffic? What do I do about the VOIP line afterward? The fiber is expensive and I know the T1 does a great job for their VOIP so I do not want to suggest a DSL line because I know it may not be sufficient. I would also like to save them some money if I can. Maybe even get a faster fiber line and forgo the T1. Though I know their load balance/switch can only handle 20MB/S throughput. Looking for some confirmation/suggestions on my plan. I am planning on going in to get some real diagnostic numbers. Any suggestions on software to use for that? Preferably Windows software.

  Read the article

• MySQL password changed every other day in windows?

  - by PHP

  Every morning when I check server status,I will find MySQL's password is changed: mysql -uuser -ppassword will report ERROR 1045 (28000): Access denied for user 'user'@'localhost' (using password: YES) And then I restart server,and when it's up,MySQL will be back to normal. It has now become a routinely job. What can be the cause for this? How can I know what's exactly happening to MySQL? Here is the error log: 100122 10:11:16 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: Normal shutdown 100122 10:11:16 InnoDB: Starting shutdown... 100122 10:11:18 InnoDB: Shutdown completed; log sequence number 0 22939338 100122 10:11:18 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: Shutdown complete 100122 10:12:40 InnoDB: Started; log sequence number 0 22939338 100122 10:12:42 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: ready for connections. Version: '5.0.24-community-nt' socket: '' port: 3306 MySQL Community Edition (GPL) 100123 16:20:44 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: Normal shutdown 100123 16:20:44 InnoDB: Starting shutdown... 100123 16:20:46 InnoDB: Shutdown completed; log sequence number 0 22939832 100123 16:20:46 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: Shutdown complete 100123 16:22:09 InnoDB: Started; log sequence number 0 22939832 100123 16:22:11 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: ready for connections. Version: '5.0.24-community-nt' socket: '' port: 3306 MySQL Community Edition (GPL) 100125 9:18:59 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: Normal shutdown 100125 9:18:59 InnoDB: Starting shutdown... 100125 9:19:00 InnoDB: Shutdown completed; log sequence number 0 22941001 100125 9:19:00 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: Shutdown complete 100125 9:20:22 InnoDB: Started; log sequence number 0 22941001 100125 9:20:25 [Note] D:\MySQL\MySQL Server 5.0\bin\mysqld-nt: ready for connections. Version: '5.0.24-community-nt' socket: '' port: 3306 MySQL Community Edition (GPL)

  Read the article

• Printing on Windows 8 with PDF viewer (Adobe Reader) from network

  - by Bongo

  i have a problem with the Adobe Reader 8, but the problem seems to be equally bad with other pdf viewers. Here is the configuration: My PDF viewer is located on network drive "Z:" which is the network adress \dgs-main\progs. I tried to start the adobe reader from here - \\dgs-main\progs\Adobe\Reader 8.0\Reader\AcroRd32.exe and open the PDF from here - C:\Users\ServiceDesk\AppData\Local\Temp\GeneratedPDF.pdf The problem is as follows, if i open the PDF with a local PDF viewer everything works fine and i can print the document. If i open the PDF with the Network PDF viewer then it opens, but printing is impossible. The error message states: "Unable to start print job. Is printer available?" As mentioned above, it works with a local pdf viewer. In both cases i use the same printer. The Printer is a network printer but even with a local printer it fails. The error occurs only on Windows 8 machines. On windows 7 it works fine. I Hope somebody can tell me what the problem is. Thanks in advance and have a fine day.

  Read the article

• Private staff network within public network

  - by pianohacker

  I'm the sysadmin at a small public library. Since I got here a few years ago, I've been trying to set up the network in a secure and simple way. Security is a little tricky; the staff and patron networks need to be separated, for security reasons. Even if I further isolated the public wireless, I'd still rather not trust the security of our public computers. However, the two networks also need to communicate; even if I set up enough VMs so they didn't share any servers, they need to use the same two printers at the very least. Currently, I'm solving this with some jerry-rigged commodity equipment. The patron network, linked together by switches, has a Windows server connected to it for DNS and DHCP and a DSL modem for a gateway. Also on the patron network is the WAN side of a Linksys router. This router is the "top" of the staff network, and has the same Windows server connected on a different port, providing DNS and DHCP, and another, faster DSL modem (separate connections are very useful, especially as we heavily depend on some cloud-hosted software). tl;dr: We have a public network, and a NATed staff network within it. My question is; is this really the best way to do this? The right equipment would likely make my job easier, but anything with more than four ports and even rudimentary management quickly becomes a heavy hit on our budget. (My original question was about an ungodly frustrating DHCP routing issue, but I thought I'd ask whether my network was broken rather than asking about the DHCP problem and being told my network was broken.)

  Read the article

• Setting up a Network Bridge on Linux VM (Windows 7 Host)

  - by GrandAdmiral

  I would like to use NetEm to simulate a low bandwidth environment while testing an Internet-connected device. My plan is to setup a bridge in a Linux VM (Linux Mint 13) on a Windows 7 host. Unfortunately I'm having trouble setting up the bridge. Then I can use NetEm in the Linux VM to limit the bandwidth to an external device. I went with the following script: ifconfig eth0 0.0.0.0 promisc up ifconfig eth1 0.0.0.0 promisc up Then create the bridge and bring it up: brctl addbr br0 brctl setfd br0 0 brctl addif br0 eth0 brctl addif br0 eth1 dhclient br0 ifconfig br0 up When I run that script, I see the following warning: Rather than invoking init scripts through /etc/init.d, use the service(8) utility, e.g. service smbd reload Since the script you are attempting to invoke has been converted to an Upstart job, you may also use the reload(8) utility, e.g. reload smbd The device connecting to the bridge is able to obtain an IP Address, but it can only ping the IP Address of the bridge (both are 10.2.32.xx). Then after a few minutes, other parts of our network go down. I'm not sure why, but once I kill the bridge the network is fine. Is it possible to setup a network bridge in a Linux VM? Do I need to do something else with the dhclient br0 part of the script? By the way, I'm using VirtualBox. The wired connection is eth0 and the wireless connection is eth1. The wired connection is connecting to the device and the wireless connection is going to the network. Both adapters are set up as bridged adapters with promiscuous mode set to "allow all".

  Read the article

• Server 2003 Remote Desktop loses its virtual printer image of the local printer

  - by Charles Hart

  Server 2003 Remote Desktop provides service to stores served by several ISPs. The server loses its virtual printer image of the local printer (as seen from the remote store site) and a copy of the original local printer appears on the local computer with a different driver without notice. Specifically: A remote desktop session is opened on a local computer that has a Brother HL2140 USB printer connected and the associated software installed with a correct driver shown under the “advanced” button. The server has the same Brother software and driver. An application that is running on the server attempts to print on the local printer connected to the local computer running Vista Pro or XP Pro. Either it works correctly (Good) or it does not print (Bad) or it prints on another Local Printer connected to another local computer logged into the server (Bad and Odd). When it doesn’t print (or prints somewhere else) we ask the customer to look for the (virtual) printer using the Remote desktop view of the server and the printer is gone. Then we ask the customer to look at the printers folder in the local computer. There are several possibilities: The printer is there, but the driver is mysteriously changed in the drop down to MDX something; we have the customer select the other (proper) Brother driver, and all is well again, as now after the change, the virtual printer in the server (which now matches the local printer) appears again, and so printing can resume. A “copy” of the printer mysteriously appears in the local printer’s folder and after we delete it the virtual printer in the server appears again and so printing can resume. Note that in both case 1 and 2, the server sometimes sends the print job elsewhere, to some other local computer. Meanwhile in the log file, endless errors are reported and the server eventually crashes, sometimes twice a day. I’m puzzled what changes the local printer driver and I’m puzzled what loads the copy 2 or copy 3 of the printer in the local printer folder. This entire description randomly occurs on any of 40+ local computers in eight different locations in different ISPs, all sharing one Domain.

  Read the article

• Windows Domain Chaos - Any Solving Approach

  - by Chake

  we are running an old Window 2003 Server as Domain Controller (DC2003). To safely migrate to Windows 2008 R2 we added a 2008 R2 (DC2008R2) to the domain as domain controller (adprep etc.). After dcpromo on DC2008R2 everything seemed to be ok. The new DC appeared under the "Domain Controlelrs" node. It wasn't checked at this time, if DC2008R2 can REALLY act as domain controller. Later we tried to shutdown DC2003 and ran into a total mess with non functional Exchange and Team Foundation Services. After that I got the job to fix... First i thought it could be an Problem with DC2008R2. So I removed it as Domain Controller and installed a new Windows 2008 R8 Server DC2008R2-2. I ran into similar Problems. I tried a bunch of stuff, but nothign helped. I won't list it, maybe I made an mistake, so I'm willing to redo it with your suggestions. To have a starting point I tried the best practise analyser whicht ended up with 24 "Compatible" and 26 "Not Compatible" tests. From these 26 tests 19 read the same. (I'm translating from german, so that may to be the exact wording) Problem: Using the Best Practise Analyser for Active Directory Domain Services (Active Directory Domain Services Best Practices Analyzer, AD DS BPA) no data can be be gathered using the name of the forest and the domain controller DC2008R2-2. I appreciate any suggestions, this really bothers me.

  Read the article

• Notebook Operating System with extreme support cycles/security updates

  - by leto

  Hello there, after reading the announcements about Mac OS X "Lion" and Apples political decision, I've had enough. I'm a longtime Apple User since 1992, have always felt at home there, but am trying to switch to alternative Operating System since a year. I've also been working with Unix machines since 2001, so I'm looking in one of the free Unices or a Linux. Since I last looked at the desktop in 2002 choke much has changed, it seems. So I'm lost once more in the war between desktop environments and software. To be honest: I don't care what it's name is, I want to get my job done. Here's what I set me as landmark for an operating system/software to be considered: Has to be atleast four years old Has to supply security updates for current release for atleast a year Production quality stability for the whole desktop environment (!) No f****g commercial stuff that tends to supply me with privacy invading App Store or Cloud space So far I'm running a MacBook from 2007, 4 Gig memory, 250 Gig disk and I need: IMAPs for Mail since 1995 Webbrowser sic Shell Keeping current with Updates/Upgrades with no more than 5 Minutes spent in entering commands (makes it hard for OpenBSD ;-) ) A desktop filemanger would be nice, but is a bonus. What can you suggest as operating system? The one with the longest support cycles and best chance to survive the next 10 years will win a new user, even sending patches when needed :-) Greets

  Read the article

• How/when do you study?

  - by Sergei

  Would be interesting to find out how other sysadmins are educating themselves. I find myself in the need of constantly learning new things. I prefer to spend more time on the subject and know it thoroughly knowing that not doing so will kick me back in the future. This can be frustrating sometimes as it feels that I move too slowly. Our company has account on safari.oreilly.com and I am reading a book or two at any given time. I also read sysadmin related blogs for ideas and tips and to keep myself in the tune with the trends. I cannot do any study at home as I would rather spend my out of work hours with my family plus I find it hard/impossible to study at home due to the inability to concentrate at home. So I mostly study while on the train, luckily my commute time takes up to 2 hours a day. I also read a lot at work and don't feel guilty about it. To fix/implement/plan, I need to have a solid knowledge and if it requires time then this is a part of my job being a sysadmin. There is a joke that says "sysadmin is a person that knows a lot about everytihng and as a result knows nothing" - I think ther is a grain of truth here...

  Read the article

• Cisco IOS BVI ACL: Only allow established UDP

  - by George Bailey

  Related: Cisco IOS ACL: Don't permit incoming connections just because they are from port 80 I know we can use the established keyword for TCP.. but what can we do for UDP (short of replacing a Bridge or BVI with a NAT)? Answer I found out what "UDP has no connection" means. DNS uses UDP for example.. named (DNS server) is lisenting on port 53 nslookup (DNS client) starts listening on some random port and sends a packet to port 53 of the server and notes the source port in that packet. nslookup will retry 3 times if necessary. Also the packets are so small that it does not have to worry about them coming in the wrong order. If nslookup receives a response on that port that comes from the servers IP and port then it stops listening. If the server tried to send two responses (for example a response and a response to the retry) then the server would not care if either of them made it because the client has the job to retry. In fact.. unless ICMP 3/3 packet gets through the server would not know about a failure. This is different from TCP where you get connection closed or timed out errors. DNS allows for an easy retry from the client as well as small packets.. so UDP is an excellent choice because it is more efficient. In UDP you would see nslookup sends request named sends answer In TCP you would see nslookup's machine sends SYN named's machine sends SYN-ACK nslookup's machine sends ACK and the request named's machine sends the response That is much more than is necessary for a tiny DNS packet

  Read the article

• Including email, IMs, configs, etc. in documentation or notes

  - by Jason Antman

  The shop I work in is pretty laid-back. We're on a documentation kick, only because historically we've been very bad with it. We do a lot of our brainstorming in face-to-face meetings, and also do a lot of communication via IM in addition to email. While I'm usually pretty good about documentation and keeping copious lab notes, I just finished a build of a host and spent hours searching through IMs, emails, files on my workstation, etc. to pull out anything I missed in my lab notes, which formed a large amount of the basis for the internal documentation. Does anyone have any thoughts on, aside from manually saving things to a project directory, managing various data sources (especially email and IM) and tracking them on project basis? Ideally, I'd like an easy way to put copies of emails, IM logs, etc. into a project-specific directory on my workstation and then just have a cron job that syncs that up with a shared folder. This isn't really a candidate for anything more advanced, as the bulk of the data will be copies of configs, code, etc. Here are the big restrictions: Email is via a centralized Zimbra install, so nothing can happen server-side. My workstation is Linux. Aside from writing Pidgin and Thunderbird plugins that let me tag chats and emails as belonging to a project, and then copy them to the appropriate place... any thoughts? Suggestions? Thanks, Jason

  Read the article

• How can I take a one time full backup of a Windows Server without the need for a restore program?

  - by TheCleaner

  I have a Windows SBS server with about 500GB of data that I'm decommissioning but I'd like to take a final backup of the server and place it on an external USB drive. I already have multiple backups of the server on disk from the past but they are through Simpana Commvault. I'd like a backup that will simply copy the file structure, ACLs, timestamps, etc. as is to a NTFS volume on the external drive. This way if someone says "I need x file on the server you decommed" I can search the external drive real quick instead of firing up Commvault, cataloging, restore, etc. I know the built-in Windows backup is great, I just don't feel like running it for a restore job on this. I'd like an option where in the future it won't require a program to run a restore. Rather a simple mount of the drive will suffice. I believe I can use robocopy just fine, but I'm not sure if it will grab the Windows directory, system files, and full user profiles correctly even with the /ZB option. Options? Is Robocopy /E /ZB /COPYALL /DCOPY:DAT /MT:32 /R:5 /W:5 /LOG:copylog.log the way to go here?

  Read the article

• My internet drops and will only reconnect if I troubleshoot it or restart my computer.

  - by Paul

  My internet loses its connection and at the same time, and my speaker seems broken if that happens and my mouse will not be smooth as it was before it happens. It already happened before and I didnt mind it. But now, it's happening literally all the time. After startup, my connection seems okay but after some few minutes, even without me doing something, my laptop will just lose its connection. It says that it is not connected and connections are available, but if I check if there are connections where I can connect it shows nothing. Even if, there is actually a wi-fi connection where my phone and other laptops connect. They dont seem to have problem with it. Just my laptop. After that happens, I usually restart my laptop but I discovered that troubleshooting it would somehow repair the problem, but not all the time. It says that restarting my wireless adapter do the job. Are there any long term fix? I mean that could last long or totally erase the problem? I have Atheros AR5B97 Wireless Network Adapter and Microsoft Virtual WiFi Miniport Adapter, both enabled. On the services, I have Wired AutoConfig and WLAN AutoConfig on Automatic. Aspire 4750, Windows 7 (x86)

  Read the article

• How to test nginx proxy timeouts

  - by mkorszun

  Target: I would like to test all Nginx proxy timeout parameters in very simple scenario. My first approach was to create really simple HTTP server and put some timeouts: Between listen and accept to test proxy_connect_timeout Between accept and read to test proxy_send_timeout Between read and send to test proxy_read_timeout Test: 1) Server code (python): import socket import os import time import threading def http_resp(conn): conn.send("HTTP/1.1 200 OK\r\n") conn.send("Content-Length: 0\r\n") conn.send("Content-Type: text/xml\r\n\r\n\r\n") def do(conn, addr): print 'Connected by', addr print 'Sleeping before reading data...' time.sleep(0) # Set to test proxy_send_timeout data = conn.recv(1024) print 'Sleeping before sending data...' time.sleep(0) # Set to test proxy_read_timeout http_resp(conn) print 'End of data stream, closing connection' conn.close() def main(): s = socket.socket(socket.AF_INET, socket.SOCK_STREAM) s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1) s.bind(('', int(os.environ['PORT']))) s.listen(1) print 'Sleeping before accept...' time.sleep(130) # Set to test proxy_connect_timeout while 1: conn, addr = s.accept() t = threading.Thread(target=do, args=(conn, addr)) t.start() if __name__ == "__main__": main() 2) Nginx configuration: I have extended Nginx default configuration by setting explicitly proxy_connect_timeout and adding proxy_pass pointing to my local HTTP server: location / { proxy_pass http://localhost:8888; proxy_connect_timeout 200; } 3) Observation: proxy_connect_timeout - Even though setting it to 200s and sleeping only 130s between listen and accept Nginx returns 504 after ~60s which might be because of the default proxy_read_timeout value. I do not understand how proxy_read_timeout could affect connection at so early stage (before accept). I would expect 200 here. Please explain! proxy_send_timeout - I am not sure if my approach to test proxy_send_timeout is correct - i think i still do not understand this parameter correctly. After all, delay between accept and read does not force proxy_send_timeout. proxy_read_timeout - it seems to be pretty straightforward. Setting delay between read and write does the job. So I guess my assumptions are wrong and probably I do not understand proxy_connect and proxy_send timeouts properly. Can some explain them to me using above test if possible (or modifying if required).

  Read the article

< Previous Page | 222 223 224 225 226 227 228 229 230 231 232 233  | Next Page >