In that case (if it's happening due to i/o), start with the zpool list, zpool iostat and iostat -o (look at msps, millisecond per transaction). Basically, look for any high i/o usage, latencies or odd behaviours.
On Sun, Aug 31, 2014 at 10:06 AM, David Finster < [email protected]> wrote: > I’ll see how I go, but it is difficult to know when the issue would have > otherwise occurred as it is fairly sporadic/random. > > If it were a Windows issue, how would running vmadm > info/pstack/lockstack cause the machine to begin responding again? > > I’ve just noticed that one of the machines has gone to sleep again and > since it isn’t in production I can keep it down for the time being. If > anyone has any debugging suggestions it would be appreciated. > > On 31 Aug 2014, at 1:39 pm, Micky <[email protected]> wrote: > > Does it happen if you keep an RDP open? Do a ping -t or something to > keep the network sockets open on a fresh install or something like that. > Seems like a Windows issue. > > > On Sat, Aug 30, 2014 at 12:28 PM, David Finster via smartos-discuss < > [email protected]> wrote: > >> As an update, the issue occurred again on two different hosts for KVM >> machines. >> >> Running pstack across the qemu process for the first KVM caused it to >> wake up, but the following was captured: >> http://pastebin.com/Js20eZPH >> >> On the second machine, running plockstat also woke the KVM up, but >> provided this: >> http://pastebin.com/pbEt2bnj >> >> >> On 29 Aug 2014, at 1:11 pm, David Finster via smartos-discuss < >> [email protected]> wrote: >> >> I should also be more specific in saying that it appears that the VM >> locks up during this period of sleep. Once woken up, there are no event >> logs for the period that the machine was unreachable. Network activity >> isn’t the only thing affected. >> >> Thanks, >> Dave >> >> >> On 29 Aug 2014, at 1:07 pm, David Finster via smartos-discuss < >> [email protected]> wrote: >> >> Hi Everyone >> >> I’m seeing a weird issue with 3 particular KVM VMs whereby the virtual >> machine appears to simply stop responding to network traffic (unpingable). >> The weird thing is that as soon as I jump into the hypervisor and run >> ‘vmadm info <uuid>’, the machine immediately starts responding to pings. >> >> I obtained the stacks of the qemu process, but accidentally ran vmadm >> info before I went further. >> >> Stacks can be found here: http://pastebin.com/8FBxrcJJ >> >> Does anyone have any suggestions/further debugging steps? Next time it >> happens I’ll see if I can obtain lock info and anything else I can think of. >> >> The KVM machines are running Server 2012 R2 64-bit (there are other >> machines also running this OS that aren’t having issues) and the issue is >> occurring on two separate hosts. >> >> Build version is 20140717T041004Z >> >> Thanks, >> Dave >> *smartos-discuss* | Archives >> <https://www.listbox.com/member/archive/184463/=now> >> <https://www.listbox.com/member/archive/rss/184463/25738179-216c4b5f> | >> Modify <https://www.listbox.com/member/?&;> Your Subscription >> <http://www.listbox.com/> >> >> >> *smartos-discuss* | Archives >> <https://www.listbox.com/member/archive/184463/=now> >> <https://www.listbox.com/member/archive/rss/184463/25738179-216c4b5f> | >> Modify <https://www.listbox.com/member/?&;> Your Subscription >> <http://www.listbox.com/> >> >> >> *smartos-discuss* | Archives >> <https://www.listbox.com/member/archive/184463/=now> >> <https://www.listbox.com/member/archive/rss/184463/25253051-d319687c> | >> Modify >> <https://www.listbox.com/member/?&;> >> Your Subscription <http://www.listbox.com/> >> > > > ------------------------------------------- smartos-discuss Archives: https://www.listbox.com/member/archive/184463/=now RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00 Modify Your Subscription: https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb Powered by Listbox: http://www.listbox.com
