Date: Sun, 20 Oct 2002 21:53:10 -0500 (CDT) From: "Jason A. Young" <jyoung@power.doogles.com> To: current@freebsd.org Subject: Reproduceable partial hang on -current as of a few days ago Message-ID: <Pine.LNX.4.44.0210202134260.5316-100000@power.doogles.com>
next in thread | raw e-mail | index | archive | help
While running 'time nice +10 make -j4 buildkernel KERNCONF=JYOUNGLP', the
compile process eventually grinds to a halt. The compiler wedges in state
'ufs' according to top (I keep it running while reproducing the issue) and
ctrl-T. After the compile process freezes, anytime I try to start
something new, the csh process I try to run it from hangs in 'thrd_sleep'.
I'm doing this compile from within X, and it seems fine, with the obvious
caveat that I can't start anything new). I can switch VTs and break into
the kernel debugger.
This problem or something very much like it was mentioned on either
-current or -hackers a week or two ago, but now I can't find the message
in the archives to know who brought it up. I'd seen it once or twice
before, but now I've reproduced it 3 times in a row with this kernel
compile. It even hangs during the same set of files each time in one of
the SCSI drivers.
I don't know if this is useful, but a start might be:
Debugger("manual escape to debugger")
Stopped at Debugger+0x54: xchgl %ebx,in_Debugger.0
db> show lockedvnods
Locked vnodes
0xc8c6e818: tag ufs, type VDIR, usecount 1, writecount 0, refcount 2, flags (VV_OBJBUF), lock type ufs: EXCL (count 1) by pid 2250
ino 178176, on dev ad0s2g (4, 9)
0xc903e000: tag ufs, type VREG, usecount 18, writecount 0, refcount 1, flags (VV_TEXT|VV_OBJBUF), lock type ufs: EXCL (count 1) by pid 2245 with 3 pending
ino 311836, on dev ad0s2g (4, 9)
0xc903dde0: tag ufs, type VREG, usecount 18, writecount 0, refcount 1, flags (VV_TEXT|VV_OBJBUF), lock type ufs: EXCL (count 1) by pid 2248 with 3 pending
ino 178212, on dev ad0s2g (4, 9)
db> show locks
exclusive sleep mutex Giant r = 0 (0xc02d82a0) locked @ /usr/src/sys/kern/kern_intr.c:533
Trace output is uninteresting; it only reflects the ctrl-alt-escape code
path into DDB.
I'll hold the system in this state where I can reproduce the problem and
get other debug/trace output if somebody here thinks it would be useful to
do so. Please let me know.
--
Jason Young, CCIE #8607, MCSE
Sr. Network Technician, WAN Technologies
(314)817-0131
http://www.wantec.com
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.LNX.4.44.0210202134260.5316-100000>