Subject: Re: Getting "TLB IPI rendezvous failed..."
To: Stephan Uphoff <firstname.lastname@example.org>
From: Manuel Bouyer <email@example.com>
Date: 01/18/2005 00:46:13
On Sat, Jan 15, 2005 at 11:36:59AM -0500, Stephan Uphoff wrote:
> > If you think that a current kernel has additionnal fixes that may be relevant,
> Mhhh .. is see a change in the i386 spl logic:
> > Updaing ci_ilevel and testing ci_ipending must be done with all
> > interrupts
> > off, or priority inversion can occur, which can lead to IPI deadlocks.
> > Leaves interrupts off for a bit longer, sadly, but with no noticeable
> > effects on the systems I tested on.
> > From YAMAMOTO Takashi.
> That did not make it to the 2-0-RELEASE.
> > I can try a current kernel.
> This would be helpful.
Well, a current kernel paniced too, when the amanda backup started.
Unfortunably I forgot to set ddb.onpanic to 1, so I don't know how it
paniced (I just know thw box rebooted). I will enable ddb for the
next amanda run.
BTW, memory isn't an issue on this box. It has 1Gb RAM, there is always
at last 200-300MB free. Even the buffer cache won't eat it all, and
the backup use dump on the raw device.
Manuel Bouyer <firstname.lastname@example.org>
NetBSD: 26 ans d'experience feront toujours la difference