Re: [ceph-users] mon down for 3 hours after clocksource glitch

2013-07-18 Thread Dan van der Ster
On Thu, Jul 18, 2013 at 9:29 PM, Sage Weil wrote: > this sounds exactly like the problem we just fixed in v0.61.5. Glad to hear that. Thanks for the quick help :) dan ___ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.

Re: [ceph-users] mon down for 3 hours after clocksource glitch

2013-07-18 Thread Sage Weil
On Thu, 18 Jul 2013, Dan van der Ster wrote: > On Thu, Jul 18, 2013 at 6:27 PM, Sage Weil wrote: > > Hi Dan, > > > > On Thu, 18 Jul 2013, Dan van der Ster wrote: > >> Hi, > >> Last night our cluster became unhealthy for 3 hours after one of the mons > >> (a > >> qemu-kvm VM) had this glitch: > >>

Re: [ceph-users] mon down for 3 hours after clocksource glitch

2013-07-18 Thread Dan van der Ster
On Thu, Jul 18, 2013 at 6:27 PM, Sage Weil wrote: > Hi Dan, > > On Thu, 18 Jul 2013, Dan van der Ster wrote: >> Hi, >> Last night our cluster became unhealthy for 3 hours after one of the mons (a >> qemu-kvm VM) had this glitch: >> >> Jul 18 00:12:43 andy03 kernel: Clocksource tsc unstable (delta

Re: [ceph-users] mon down for 3 hours after clocksource glitch

2013-07-18 Thread Sage Weil
Hi Dan, On Thu, 18 Jul 2013, Dan van der Ster wrote: > Hi, > Last night our cluster became unhealthy for 3 hours after one of the mons (a > qemu-kvm VM) had this glitch: > > Jul 18 00:12:43 andy03 kernel: Clocksource tsc unstable (delta = > -60129537028 ns).  Enable clocksource failover by adding

[ceph-users] mon down for 3 hours after clocksource glitch

2013-07-18 Thread Dan van der Ster
Hi, Last night our cluster became unhealthy for 3 hours after one of the mons (a qemu-kvm VM) had this glitch: Jul 18 00:12:43 andy03 kernel: Clocksource tsc unstable (delta = -60129537028 ns). Enable clocksource failover by adding clocksource_failover kernel parameter. shortly afterwords t