Page 1 of 2 12 LastLast
Results 1 to 25 of 26

Hybrid View

  1. #1
    Join Date
    Nov 2007
    Location
    England
    Posts
    239

    IP Addresses on one of our Nodes

    Hey,

    We have one server that all of the IPs continue to fail. We have fixed it a few times and it just happened again! I have no idea why it is happening. I just had to reboot the node and now everything is working again.

    Can anyone explain why this is happening because it is killing my company :'(

    Thanks.

  2. #2
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    We have 3 servers and it is only happening on one server.

  3. #3
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    It's just happened again! Does anyone have any idea on this?

  4. #4
    Join Date
    Mar 2007
    Location
    Washington State
    Posts
    1,088
    Define "fail"? What exactly is going wrong?
    FazeWire Web Services.
    || We have provided great prices and better support since 2006. Located in Seattle, WA!
    || -----------------
    || Shared Hosting - VPS - Dedicated Servers - Colocation - Software Licenses

  5. #5
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    The IPs die. They can't be pinged, none of the sites on all the VPS's will load.

    It happened 30 minutes or so ago as well...

  6. #6
    Join Date
    Apr 2005
    Location
    Cochin
    Posts
    2,452
    This could be an abuse in one of the VPS.

    Whats the specs of your VPS node ?

  7. #7
    Join Date
    Mar 2007
    Location
    Washington State
    Posts
    1,088
    When they die, are they still on the machine, viewable with "ifconfig"?

    OpenVZ or Xen?
    FazeWire Web Services.
    || We have provided great prices and better support since 2006. Located in Seattle, WA!
    || -----------------
    || Shared Hosting - VPS - Dedicated Servers - Colocation - Software Licenses

  8. #8
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    2x 1TB HDDs, 8GB RAM, Q9550

    OpenVZ - They aren't bound to the Host.

    This is what i posted on another useful forum;

    Hey,

    We have 4 servers all connected to the same VLAN for VPS hosting. All servers are running CentOS 64bit with OpenVZ installed.

    One of our servers, all of the IPs fail. We figured out the first time it happened it was NEIGHBOR_DEVS in the vz.conf file which it was saying "all" and we changed it to "detect" and it all worked again. Recently it all went down again and we got told to change it back to "all" so we did, rebooted VZ and everything worked again. Now today, all IPs failed again, I rebooted VZ, nothing, I rebooted the server and everything came back up again.....

    Can anyone tell me why this is happening? It's killing my business daily..

    Thanks.

  9. #9
    Quote Originally Posted by VPSRight View Post
    2x 1TB HDDs, 8GB RAM, Q9550

    OpenVZ - They aren't bound to the Host.

    This is what i posted on another useful forum;
    http://kb.parallels.com/en/5232

    This article tells you how to update something.

    Try updating your virtuozzo to see if it helps.

    Something is not being written probably.

    Does virtuozzo have a repair utility for the node itself?

    5. Obtaining:

    This update can be downloaded via ftp from downloads.swsoft.com.
    If you do not have an ftp account, please contact pavel@parallels.com.

    You can also download and install this tools update using the
    'vzup2date' utility included in the Virtuozzo Containers 4.0 distribution.

  10. #10
    Join Date
    Mar 2007
    Location
    Washington State
    Posts
    1,088
    Check for errors on the NIC.
    FazeWire Web Services.
    || We have provided great prices and better support since 2006. Located in Seattle, WA!
    || -----------------
    || Shared Hosting - VPS - Dedicated Servers - Colocation - Software Licenses

  11. #11
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    and again..

    It's dropping every 30 minutes

  12. #12
    Join Date
    Apr 2004
    Location
    Pacific Palisades, CA
    Posts
    3,641
    It should be spelled NEIGHBOUR_DEVS, and keep it =ALL

    Check for IP conflicts
    Collabora Hosting - Unlimited Windows and Linux Hosting
    Web Security - VPS - Dedicated Servers
    Cloud and Managed WordPress Hosting
    Read how we do Unlimited Hosting at the Unlimited FAQ

  13. #13
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    It is that and set to =all - There are no IP conflicts.

    The main IP to the server is accessible just all the IPs assigned to VPS servers become non responsive.

  14. #14
    Join Date
    Aug 2000
    Posts
    2,754
    Checked IPTables?

    Do the VPS's themselves stop running? can you enter them and can you see an output of ifconfig?

  15. #15
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    It just happened AGAIN! It's failing every 30 minutes to 1 hour ???

    All VPS servers are still running when this happens.

  16. #16
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    Aug 28 19:22:07 node2 kernel: EXT3-fs error (device dm-4): ext3_add_entry: bad entry in directory #53348367: rec_len % 4 != $
    Aug 28 19:22:41 node2 last message repeated 39 times
    Aug 28 19:23:42 node2 last message repeated 114 times
    Aug 28 19:24:11 node2 last message repeated 119 times
    This error is displaying in /var/log/messages and then after I see my reboot request.

    Aug 28 19:22:07 node2 kernel: EXT3-fs error (device dm-4): ext3_add_entry: bad entry in directory #53348367: rec_len % 4 != $
    Aug 28 19:22:41 node2 last message repeated 39 times
    Aug 28 19:23:42 node2 last message repeated 114 times
    Aug 28 19:24:11 node2 last message repeated 119 times
    Aug 28 20:05:21 node2 shutdown[2220]: shutting down for system reboot
    I'd rather not give access as this is a live VPS node.

    We are having to reboot the server at the exact same time every hour.

  17. #17
    Join Date
    Aug 2000
    Posts
    2,754
    Tried running e2fsck ?

  18. #18
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    Tried that when we found a corrupted file the other day via CentOS rescue(disk) and it failed saying a device is busy. After searching the system for the mentioned inode, I found this;

    [root@node2 ~]# find / -inum 53348367
    find: /proc/15324/task/15324/fd/4: No such file or directory
    find: /proc/15324/fd/4: No such file or directory
    find: /vz/vztmp/vzctl-rm-me.XQ6Uwax/home/scbuzz/public_html/blag/wp-includes/js/jquery: Not a directory
    find: /vz/vztmp/vzctl-rm-me.XQ6Uwax/home/scbuzz/public_html/blag/wp-includes/js/scriptaculous: Not a directory
    The VPS with the scbuzz account was terminated today, so that shouldn't exist on the system. I also have another thread on here about "Strange File Permissions" which concerns the above quoted files acting as devices.

    Is this relating to a HW failure or something?

  19. #19
    Join Date
    May 2006
    Location
    NJ, USA
    Posts
    6,645
    Looks to me like some kind of hardware issue.

  20. #20
    Join Date
    Aug 2000
    Posts
    2,754
    Tried rebuilding your raid array? Look's like you could have a failing disk.

  21. #21
    Join Date
    Dec 2001
    Location
    MO
    Posts
    629
    Edit: Totally missed the fact that there was a page 2 to this thread, lol. You have data corruption. This isn't the exact same error, but the advice is the same I would give: http://linux.derkeiler.com/Mailing-L.../msg10740.html
    Last edited by subigo; 08-29-2009 at 01:55 AM.

  22. #22
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    We have ran FSCK on the node but the IPs are still failing 5 minutes past every hour!

    Does anyone know how to stop this from happening?!?

  23. #23
    Join Date
    Feb 2008
    Location
    Houston, Texas, USA
    Posts
    3,262
    Can you run this script to see if it causes your VEs to die: /usr/share/vzctl/scripts/vpsnetclean? The script might be located elsewhere (locate vpsnetclean).

    Also check to see if you have a firewall running on the hw node like csf. It might have some funny rules.

    Regards
    Last edited by UNIXy; 08-29-2009 at 07:37 AM.
    UNIXy - Fully Managed Servers and Clusters - Established in 2006
    Server Management - Unlimited Servers. Unlimited Requests. One Plan!
    cPanel Varnish Plugin -- Seamless SSL Caching (Let's Encrypt, AutoSSL, etc)
    Slow Site or Server? Unable to handle traffic? Same day performance fix: joe@unixy

  24. #24
    Join Date
    Mar 2007
    Location
    UK
    Posts
    852
    Is this hapening at the exact same time every hour?
    Have you checked to make sure you have no Cron Jobs set to run anyting which could be causing the problem.
    Have you tried stopping iptables when you cant connect to the VPS and see if that fixes it?

  25. #25
    Join Date
    Nov 2007
    Location
    England
    Posts
    239
    @Ashley Merrick;

    1. Every hour, 5 minutes past (This hour, it hasn't failed YET! Famous last words though eh.

    2. [root@node2 ~]# crontab -l
    no crontab for root

    3. IPtables is always stopped and off in the config.

    @UNIXy;

    Will try that script in a while... Moving a VPS using vzmigrate to another node of ours.

Page 1 of 2 12 LastLast

Similar Threads

  1. Nodes
    By Shikhir A in forum Dedicated Server
    Replies: 31
    Last Post: 04-29-2009, 10:29 PM
  2. New UK vps nodes
    By wise in forum VPS Hosting Offers
    Replies: 0
    Last Post: 02-12-2008, 06:27 PM
  3. Fast VPS Nodes
    By Andrew.S in forum VPS Hosting
    Replies: 9
    Last Post: 07-25-2007, 11:44 PM
  4. nodes...what's it?
    By danieln in forum VPS Hosting
    Replies: 24
    Last Post: 05-20-2006, 01:46 PM
  5. Replies: 18
    Last Post: 04-19-2004, 03:59 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •