Re: [DRBD-user] DRBD crash with bad network

2010-04-01 Thread Lars Ellenberg
On Thu, Apr 01, 2010 at 11:49:33AM +, Maxence DUNNEWIND wrote: > > You should have some more interessting logs before that, > > which should help you in "guessing" what needs to be done. > Here are the previous lines : > > Mar 30 00:52:48 z2-6 kernel: [1685605.585338] block drbd0: Handshake >

Re: [DRBD-user] DRBD crash with bad network

2010-04-01 Thread Maxence DUNNEWIND
> You should have some more interessting logs before that, > which should help you in "guessing" what needs to be done. Here are the previous lines : Mar 30 00:52:48 z2-6 kernel: [1685605.585338] block drbd0: Handshake successful: Agreed network protocol version 91 Mar 30 00:52:48 z2-6 kernel: [1

Re: [DRBD-user] DRBD crash with bad network

2010-04-01 Thread Lars Ellenberg
On Thu, Apr 01, 2010 at 08:07:00AM +, Maxence DUNNEWIND wrote: > > The most interessting line is before that. > > > > > Mar 30 00:52:48 z2-6 kernel: [1685605.588315] CPU 2 > > > > > Mar 30 00:52:48 z2-6 kernel: [1685605.589086] Pid: 21781, comm: > > > drbd0_worker Tainted: GW 2.6.3

Re: [DRBD-user] DRBD crash with bad network

2010-04-01 Thread Maxence DUNNEWIND
> > I have about 40 drbd devices per node (primary and secondaries). Our > > provider > > has lot of network issues, which sometimes cause drbd to > > disconnect/reconnect > > very often : about 500 NetworkFailure in 1 hour before the last crash : > > # grep "Connected -> NetworkFailure" /var/log

Re: [DRBD-user] DRBD crash with bad network

2010-03-31 Thread Lars Ellenberg
On Tue, Mar 30, 2010 at 10:34:06AM +0200, Maxence DUNNEWIND wrote: > Hi, > > I have a cluster of 10 servers with many drbd devices. The drbd version is > 8.3.7, module loaded with : > drbd minor_count=128 usermode_helper=/bin/true > (because I use it with ganeti). > > I have about 40 drbd device

[DRBD-user] DRBD crash with bad network

2010-03-30 Thread Maxence DUNNEWIND
Hi, I have a cluster of 10 servers with many drbd devices. The drbd version is 8.3.7, module loaded with : drbd minor_count=128 usermode_helper=/bin/true (because I use it with ganeti). I have about 40 drbd devices per node (primary and secondaries). Our provider has lot of network issues, which