OpenVZ Forum: Support » Not oops but crash with latest stable kernel.

Home » General » Support » Not oops but crash with latest stable kernel.

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Not oops but crash with latest stable kernel. [message #34758]

Sat, 31 January 2009 18:57

TokyoBeer
Messages: 1
Registered: June 2007

Junior Member

# uname -a
Linux * 2.6.18-92.1.18.el5.028stab060.2ent #1 SMP Tue Jan 13 12:43:59 MSK 2009 i686 i686 i386 GNU/Linux
# vzctl --version
vzctl version 3.0.23
# cat /etc/redhat-release
CentOS release 5.2 (Final)

netconsole output:

Jan 31 15:01:35 * Orphaned socket dropped (65,130 in CT2002)
Jan 31 15:03:53 * printk: 3 messages suppressed.
Jan 31 15:03:53 * Orphaned socket dropped (69,138 in CT2002)
Jan 31 15:04:08 * Orphaned socket dropped (68,136 in CT2002)
Jan 31 15:04:34 * Orphaned socket dropped (68,136 in CT2002)
Jan 31 15:04:35 * Orphaned socket dropped (67,134 in CT2002)
Jan 31 15:04:36 * Orphaned socket dropped (66,132 in CT2002)
Jan 31 15:04:39 * Orphaned socket dropped (65,130 in CT2002)
Jan 31 15:04:40 * Orphaned socket dropped (64,128 in CT2002)
Jan 31 15:04:40 * Orphaned socket dropped (64,128 in CT2002)
(system stopped responding at 15:29:19)
Jan 31 15:49:26 * TCP: time wait bucket table overflow (CT2005)

basic top output looks like:

top - 15:29:19 up 13:24, 2 users, load average: 60.62, 21.14, 11.00
Tasks: 759 total, 26 running, 731 sleeping, 0 stopped, 2 zombie
Cpu0 : 0.0%us,100.0%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu1 : 0.0%us,100.0%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu2 : 0.0%us,100.0%sy, 0.0%ni, 0.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu3 : 33.3%us, 11.9%sy, 0.0%ni, 0.0%id, 54.8%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 6223648k total, 6146504k used, 77144k free, 66772k buffers
Swap: 8193140k total, 1920k used, 8191220k free, 4078240k cached

System respond to ping but all web-sites are down.
logs and dmesg do not contain anything useful.

What should be my next troubleshooting steps?

Report message to a moderator

Re: Not oops but crash with latest stable kernel. [message #34772 is a reply to message #34758]

Mon, 02 February 2009 09:53

maratrus
Messages: 1495
Registered: August 2007
Location: Moscow

Senior Member

Hello,

unfortunately we cannot affirm for now that oops or crash happened since the are no messages point to it.
Probably the user settings caused such behavior.

But if you still sure that kernel is a suspected and we lost some messages because of the netconsole you could install serial console instead
http://wiki.openvz.org/Remote_console_setup#Serial_console

Quote:

System respond to ping but all web-sites are down.
logs and dmesg do not contain anything useful.

Where do web-sites locate? On the HN or inside CT?
Do CTs respond to ping or they lost network connectivity?
Please make sure that failcounters don't increase in /proc/user_beancounters.
May be it worth using tcpdump to obtain packets behavior when somebody try to connect to your web-server?

Report message to a moderator

Previous Topic:	VE connectivity problems. OUT to IN not working IN to OUT working perfectly
Next Topic:	User Seperation Issue

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Sep 27 19:24:38 GMT 2024

Total time taken to generate the page: 0.04218 seconds