On 09/11/17 09:38 AM, Michael Loftis wrote:
On Sun, Sep 10, 2017 at 12:39 Ian Collins <[email protected] <mailto:[email protected]>> wrote:On 09/11/17 05:26 AM, Michael Loftis wrote: > Yeah I reported similar or same issue with the H330 here > https://www.illumos.org/issues/8391 > > I haven't had much time to pursue this personally. > I see the same messages, but my system is all SAS and the symptoms can be triggered by simply importing the pool, rendering the system FUBAR. After a weekend of cycling through reboots, the system is now back to life, which is well odd considering I didn't do anything. So far I have been able to start the key systems (Windows and Linux KVMs) and clean their disks without issue.My guess is probably it was making some progress each time at syncing txg or similar housekeeping tasks. Once it caught up the I/O load under import dropped and it made it...
Yes, that sounds about right - two days of import/hang/reboot! At least it was able to reboot its self.
My machine will run indefinitely as long as there's little to no I/O load. Once there's a fair amount of I/O the driver starts spitting out errors eventually resulting in it stopping all I/O and eventually the kernel declares it dead and resets (or drops to kdb)
I don't think it is the volume of I/O, but the location. I'm currently working though the client KMVs and zero filling their drives (dd in Linux, sdelete in Windows) which maxes out the pool I/O. Last time I tried this, one VM would always trigger the hang.
Cheers, -- Ian. ------------------------------------------- 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
