Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On 9/06/10 5:53 PM, Shane Goulden wrote: > Thanks for the reply Lars. I've disabled TSO on the drbd NIC and so far haven't had a crash but the highest uptime in the past 3 months on this server has been 18 days. If it exceeds 18 days it would be safe to say this has resolved the problem. I'll keep the list updated. It's at 5 days uptime at the moment. > > Regards, > Shane > > ----- Original Message ----- > From: "Lars Ellenberg"<lars.ellenberg at linbit.com> > To: drbd-user at lists.linbit.com > Sent: Wednesday, 9 June, 2010 5:46:50 PM GMT +10:00 Canberra / Melbourne / Sydney > Subject: Re: [DRBD-user] Kernel BUG - Xen& DRBD setup > >> Any ideas? >> > http://www.gossamer-threads.com/lists/drbd/users/17207 > http://www.gossamer-threads.com/lists/drbd/users/16962 > > If workarounds suggested there don't help, the workaround setting drbd module > parameter disable_sendpage=1, should work in any case. > > That module parameter has been introduced in response to those threads. > > Can be set on module load, > or by echo 1> /sys/modules/drbd/parameters/disable_sendpage > after module load. > > One of those threads ends with the question > "Any chance this could be fixed in DRBD, for example by explicitly cleaning > up the connection before returning I/O complete to Xen? " > > Which unfortunately must be answered as > "We already do cleanup the connection. > If that does not help, we are sorry." > I can confirm that I have not had a crash in 21 days now. It appears this issue has been resolved by disabling tx checksumming on the DRBD interface (/sbin/ethtool -K eth1 tx off). Thank you for the assistance. Regards, Shane -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20100625/64338fcd/attachment.htm>