OpenVZ Forum: Support => TCP: time wait bucket table overflow (CT230)

Members

Help

Home

Home » General » Support » TCP: time wait bucket table overflow (CT230)

Show: Today's Messages :: Show Polls :: Message Navigator

E-mail to friend

TCP: time wait bucket table overflow (CT230) [message #30499]

Tue, 27 May 2008 10:12

ehab
Messages: 9
Registered: February 2007

Junior Member

From: *static.link.net

I virtualized a hardware node according to the wiki. I used the latest RHEL based openvz kernel (2.6.18-53.1.19.el5.028stab053.14) and (2.6.18-53.1.13.el5.028stab053.10). When I load the virtualized vps i get the following in dmesg and a very hight server load.

printk: 42274 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 39860 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 38309 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 38956 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 37246 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 37389 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 36846 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 18077 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 36528 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 12413 messages suppressed.
TCP: time wait bucket table overflow (CT230)
printk: 5 messages suppressed.
TCP: time wait bucket table overflow (CT230)

could i have forgotten something to do in the physical to virtual migration? Or is this a kernel problem? Can it be hardware or driver related? I tried the fixes in the forum and did
echo 512 > /proc/sys/net/ipv4/tcp_max_tw_kmem_fraction
and that did not fix anything.

Report message to a moderator

Re: TCP: time wait bucket table overflow (CT230) [message #30522 is a reply to message #30499 ]

Wed, 28 May 2008 00:43

ehab
Messages: 9
Registered: February 2007

Junior Member

From: *217.54.224.184.nile-online.net

it only worked well when i adjusted
echo 50000 > /proc/sys/net/ipv4/tcp_max_tw_buckets_ub
is this normal or acceptable is 50000 too much?
i am on 64bit amd.

Report message to a moderator

Previous Topic:	fuse/sshfs problem
Next Topic:	Mounting audio devices.

Goto Forum:

-=] Back to Top [=-

[ Generate printable PDF ] [ Syndicate this forum (XML) ] [

]

Current Time: Thu Sep 2 11:11:27 EDT 2010

.:: Contact :: Home ::.