Date: Sun, 17 Nov 2002 14:39:29 -0500 (EST) From: Robert Watson <rwatson@FreeBSD.org> To: Kris Kennaway <kris@obsecurity.org> Cc: current@FreeBSD.org Subject: Re: Processes hanging in thrd_sleep Message-ID: <Pine.NEB.3.96L.1021117143738.93303E-100000@fledge.watson.org> In-Reply-To: <20021117105919.GA677@rot13.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
I ran into that during heavy builds on one of my boxes a few months ago -- I never really got around to properly debugging it because the UFS file systems promptly ate themselves. Oddly, I had two boxes in particular that this happened on, and none of my others, and it wasn't clear to me if there was a useful pattern. I will try reproducing it sometime this weekend. Basically, the system seemed fairly live, but any attempt to execve() would hang the process in that sleep state. It looked a bit like a VM lock leak followed by piling up on locks into a deadlock staet. Robert N M Watson FreeBSD Core Team, TrustedBSD Projects robert@fledge.watson.org Network Associates Laboratories On Sun, 17 Nov 2002, Kris Kennaway wrote: > Since upgrading my kernel to today's current (from a couple of weeks > ago) I have had a number of hangs where processes block in the kernel, > usually in the thrd_sleep state (but once one hung in the ufs state). > > e.g: > > > load: 0.01 cmd: cc 708 [ufs] 0.00u 0.00s 0% 56k > > > load: 0.01 cmd: tcsh 709 [thrd_sleep] 0.00u 0.00s 0% 1220k > > I've seen this on my sparc64 box as well as an i386 box. > > Any bright ideas? Anyone feeling guilty? :) > > Kris > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.96L.1021117143738.93303E-100000>