Results 1 to 9 of 9
  1. #1
    Join Date
    Feb 2004
    Location
    Sofia
    Posts
    1,349

    Kernel Problem or cPanel Issue?

    This is what the SM tech recommended after having investigated the causes for our 6-hours downtime:

    "I consoled into the server and found it had been locked up. I could not get a response from the keyboard. The display had the last bit of log entry created by APF, but no errors. I rebooted the server and it came up without a problem. I checked the system log to see if there were any errors loged before the buffer on the screen and didn't see anything that would have caused a lockup. Your server is now back online, though.

    If this is a recurring problem, you can try updating your kernel. People have found that the latest release through Fedora, a Redhat project, has pretty well solved the problem. Also, you can try going to the 2.6 kernel, which is the latest, but not supported by Redhat."

    But I see also that 1 hour after the server had allegedly gone down, the cPanel has increased the MaxClients limit to 260 (10 higher). I have manually configured it to 150, 2 days ago. On the other hand, the Apache log shows that the server was up for the last 5 hours following the exact time of the increased MaxClients, while the external reporting via HyperSpin.com says the server was down for the last 6 hours.

    What do you think, is that really a kernel problem or a cPanel issue?

    Thank you.
    :: :: :: :: :: ::
    :: VDSP.Net :: Directory of virtual and dedi serv providers by location and price

  2. #2
    Join Date
    Mar 2003
    Location
    California USA
    Posts
    13,294
    Steven Ciaburri | Industry's Best Server Management - Rack911.com
    Software Auditing - 400+ Vulnerabilities Found - Quote @ https://www.RACK911Labs.com
    Fully Managed Dedicated Servers (Las Vegas, New York City, & Amsterdam) (AS62710)
    FreeBSD & Linux Server Management, Security Auditing, Server Optimization, PCI Compliance

  3. #3
    Join Date
    Feb 2004
    Location
    Sofia
    Posts
    1,349
    SuperCeleron 2.4 (standard)
    RHE 3
    Linux 2.4.21-4.0.1.EL
    cPanel 9.1.0-R64

    Thank you for the link. I've read most of it, but considered this issue a little bit different that's why dared to start a new thread.
    Last edited by Imago; 03-23-2004 at 03:36 AM.
    :: :: :: :: :: ::
    :: VDSP.Net :: Directory of virtual and dedi serv providers by location and price

  4. #4
    Same box here. It has been running alot time without any problems untill the last kernel update

    Going down due to a crash every 1-3 days. Someone from SM checked the box, and saw some hda errors when it freezed, so there is definately something wrong with the kernel or harddrive. (Probably the controller drivers inthe kernel)

    And SM? They don't have time to do something.. Asking for a traceroute and then close the ticket.. Managed? Yeah, up2date in shell is all you need for server management..

    The reboot this morning was done is 22 minutes, not bad at all!
    I wish I was a carpenter

  5. #5
    Join Date
    Feb 2004
    Location
    Sofia
    Posts
    1,349
    I wish I was a carpenter too! :-))))))))
    :: :: :: :: :: ::
    :: VDSP.Net :: Directory of virtual and dedi serv providers by location and price

  6. #6
    Originally posted by Imago
    I wish I was a carpenter too! :-))))))))
    lol, I have been thinking about it for a few years
    I wish I was a carpenter

  7. #7
    Join Date
    Feb 2004
    Location
    Your Screen
    Posts
    3,998
    SM's tickets auto-close when the customer doesn't update them. If you find a ticket closed, open a new one, or pick up the phone.

    Bailey
    Let's Connect on Twitter! @thatsmsgeek2u || Fighting mediocrity one thread at a time.

  8. #8
    Join Date
    Feb 2004
    Location
    Sofia
    Posts
    1,349
    If it is an outage ticket, they close it immediately with their final response. No time to update it or ask for a SLA refund. For this purpose, you have to open a sales ticket and provide as much statistics as you can.

    As for my case, a working hypothesis is that the machine is running out of her memory. I am monitoring it closely for the last 3 hours, so it is around the yellow edge (set to 80%). If it goes up to 95%, the server will be locked up, IMHO.
    :: :: :: :: :: ::
    :: VDSP.Net :: Directory of virtual and dedi serv providers by location and price

  9. #9
    Join Date
    Feb 2004
    Location
    Sofia
    Posts
    1,349

    Follow Up

    Are there registered crashes with no logged kernel errors on machines with 1024 MB RAM and above?
    :: :: :: :: :: ::
    :: VDSP.Net :: Directory of virtual and dedi serv providers by location and price

Posting Permissions

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