OpenVZ Forum: Support » TCP: time wait bucket table overflow

Home » General » Support » TCP: time wait bucket table overflow - memory leak?

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

Re: TCP: time wait bucket table overflow - memory leak? [message #36785 is a reply to message #36772]

Mon, 20 July 2009 06:23

maratrus
Messages: 1495
Registered: August 2007
Location: Moscow

Senior Member

Hi,

Quote:

I am now nearly sure it's a kernel bug - i can't imagine any other reason.

You can file a new bugreport whenever you want. Please don't hesitate doing it. I just want to clarify the situation because I don't understand clearly what your problem consists of.

The last "top" output doesn't contain anything frightening from my point of view.
A "huge" load average is just a consequence of a "great" number of process. A good idea is to obtain the status of these processes. You can do it with help of "ps" utility. Please, examine their states.

Quote:

I can't restart the failed VE, can't raise the kmemsize or anything.

Why do you want to increase kmemsize? The only failcounters I can see on proivided vzstats output are those concerning with privvmpages not kememsize.
What process are running in a failed VE?

You have mentioned that ssh dies when this occurs. Don't you think that this problem relates to network? Do you have a direct access to that server to confirm that the node completely dies?

Do you have any specific settings, for example nfs inside VE?

Please the next time it occurs please gather alt-sysrq-* information from the problem node.
Alt-sysrq-
1) "m" - for memory info dump
2) "p" - for registers - several times, please, twice the number of CPUs
3) "a" - for scheduler stat - 3 times
4) "w" - another scheduler info - 3 times
5) "t" - for all processes calltraces. Warning - this is a resource consuming operation. At least - twice.

To gather it you probably have to install serial console
http://wiki.openvz.org/Remote_console_setup#Serial_console

Report message to a moderator

[Message index]

		TCP: time wait bucket table overflow - memory leak? By: nksupport on Tue, 14 July 2009 08:45
		Re: TCP: time wait bucket table overflow - memory leak? By: maratrus on Tue, 14 July 2009 15:57
		Re: TCP: time wait bucket table overflow - memory leak? By: nksupport on Wed, 15 July 2009 14:36
		Re: TCP: time wait bucket table overflow - memory leak? By: maratrus on Wed, 15 July 2009 16:27
		Re: TCP: time wait bucket table overflow - memory leak? By: nksupport on Fri, 17 July 2009 16:29
		Re: TCP: time wait bucket table overflow - memory leak? By: nksupport on Fri, 17 July 2009 20:31
		Re: TCP: time wait bucket table overflow - memory leak? By: maratrus on Mon, 20 July 2009 06:23
		Re: TCP: time wait bucket table overflow - memory leak? By: nksupport on Mon, 20 July 2009 06:56
		Re: TCP: time wait bucket table overflow - memory leak? By: maratrus on Mon, 20 July 2009 10:26
		Re: TCP: time wait bucket table overflow - memory leak? By: nksupport on Mon, 20 July 2009 10:29
		Re: TCP: time wait bucket table overflow - memory leak? By: maratrus on Mon, 20 July 2009 10:44
		Re: TCP: time wait bucket table overflow - memory leak? By: nksupport on Mon, 17 August 2009 13:56
		Re: TCP: time wait bucket table overflow - memory leak? By: atxadmin on Tue, 25 August 2009 11:28
		Re: TCP: time wait bucket table overflow - memory leak? By: seanfulton on Thu, 27 August 2009 13:46
		Re: TCP: time wait bucket table overflow - memory leak? By: nksupport on Thu, 27 August 2009 17:10

Previous Topic:	number of cpus on a x64 kernel
Next Topic:	dump broken in Debian 5 lenny?

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Mon Feb 16 22:44:18 GMT 2026

Total time taken to generate the page: 0.18384 seconds