NetBSD Problem Report #55630

From yorickhardy@gmail.com  Mon Aug 31 08:45:14 2020
Return-Path: <yorickhardy@gmail.com>
Received: from mail.netbsd.org (mail.netbsd.org [199.233.217.200])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.NetBSD.org", Issuer "mail.NetBSD.org CA" (not verified))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 65FFF1A923A
	for <gnats-bugs@gnats.NetBSD.org>; Mon, 31 Aug 2020 08:45:14 +0000 (UTC)
Message-Id: <5f4cb896.1c69fb81.1ce0a.d755@mx.google.com>
Date: Mon, 31 Aug 2020 10:44:55 +0200
From: yorickhardy@gmail.com
Reply-To: yorickhardy@gmail.com
To: gnats-bugs@NetBSD.org
Subject: 9.99.72: panic: ufs_inactive: dirty filesystem?
X-Send-Pr-Version: 3.95

>Number:         55630
>Category:       kern
>Synopsis:       9.99.72: panic: ufs_inactive: dirty filesystem?
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Mon Aug 31 08:50:00 +0000 2020
>Originator:     Yorick Hardy
>Release:        NetBSD 9.99.42
>Organization:

>Environment:
System: NetBSD HOME 9.99.42 NetBSD 9.99.42 (YORICK.amd64) #15: Sat Jan 25 23:26:20 SAST 2020 root@HOME:/root/build.amd64.local/obj/sys/arch/amd64/compile/YORICK.amd64 amd64
Architecture: x86_64
Machine: amd64
>Description:
I can reliably crash kernel 9.99.72 (GENERIC with options DEBUG and
LOCKDEBUG) while booting (clearing /tmp):

dmesg buffer:

ufs_inactive: unlinked ino 8854425 on "/" has non zero size 0 or blocks fffffffffffffff8 with allerror 0
panic: ufs_inactive: dirty filesystem?
cpu0: Begin traceback...
vpanic() at netbsd:vpanic+0x152
snprintf() at netbsd:snprintf
ufs_inactive() at netbsd:ufs_inactive+0x219
VOP_INACTIVE() at netbsd:VOP_INACTIVE+0x3c
vrelel() at netbsd:vrelel+0x13d
ufs_remove() at netbsd:ufs_remove+0xd3
VOP_REMOVE() at netbsd:VOP_REMOVE+0x40
do_sys_unlinkat() at netbsd:do_sys_unlinkat+0x226
syscall() at netbsd:syscall+0x283
--- syscall (number 10) ---
netbsd:syscall+0x283:
cpu0: End traceback...

crash does not provide much more information (running on 9.99.42, the last stable kernel I have available):

Crash version 9.99.42, image version 9.99.72.
WARNING: versions differ, you may not be able to examine this image.
System panicked: ufs_inactive: dirty filesystem?
Backtrace from time of crash is available.
crash> bt
_KERNEL_OPT_NARCNET() at 0
?() at ffff8f971b708000
sys_reboot() at sys_reboot
vpanic() at vpanic+0x15b
snprintf() at snprintf
ufs_inactive() at ufs_inactive+0x219
VOP_INACTIVE() at VOP_INACTIVE+0x3c
vrelel() at vrelel+0x13d
ufs_remove() at ufs_remove+0xd3
VOP_REMOVE() at VOP_REMOVE+0x40
do_sys_unlinkat() at do_sys_unlinkat+0x226
syscall() at syscall+0x283
--- syscall (number 10) ---
7754f4d77baa:
crash>

I first noticed the problem on 9.99.71, but I have not updated my kernel frequently.
Of course, 9.99.42 is working fine!
>How-To-Repeat:
Boot 9.99.72 with many files in /tmp.
>Fix:
I am unfortunately not able to bisect the problem, since my last stable kernel is 9.99.42.

Home	PR Database Search