[prev in list] [next in list] [prev in thread] [next in thread] 

List:       xen-users
Subject:    Re: [Xen-users] domU freeze on Debian stretch
From:       Valentin Vidic <Valentin.Vidic () CARNet ! hr>
Date:       2017-08-04 8:05:39
Message-ID: 20170804080539.cke6qqwtabbzxeot () gavran ! carpriv ! carnet ! hr
[Download RAW message or body]

On Wed, Jul 26, 2017 at 09:06:08AM +0200, Valentin Vidic wrote:
> Having a strange problem with a database server after upgrading
> to Debian stretch.  Shortly after boot all disk IO on the system
> hangs and the messages bellow appear although the load on the
> machine is almost 0.  The only way to restart the system is
> using xl destroy.  When this happens the system is booted with
> all 56 VCPUs and reducing this to a lower number like 24 seems
> to help (no hangs after that).  Could this be a problem with the
> Xen scheduler since all threads listed hang in __schedule?

Setting gnttab_max_frames=256 seems to help with this issue,
as suggested by the xen developers. Just found a SUSE page with
the same info:

  I/O to LUNs hang / stall under high load when using xen-blkfront
  https://www.novell.com/support/kb/doc.php?id=7018590

-- 
Valentin

_______________________________________________
Xen-users mailing list
Xen-users@lists.xen.org
https://lists.xen.org/xen-users
[prev in list] [next in list] [prev in thread] [next in thread]