Hi all,
We encountered a problem by chance today, a kernel bug which causes the
following error:
Assertion failure in journal_fl sh() at journal.c:1310:
"!journal->j_running_transaction"
kernel BUG at journal.c:1310!
This happened while remounting an ext3 filesystem read-only. We often do
this and haven't seen this bug before, so I suspect some kind of race.
It appears to be related to a message posted to this list in January 2002
by Carsted Otte
[https://listman.redhat.com/pipermail/ext3-users/2002-January/002572.html].
I noticed the repeat_lock patch given by Stephen Tweedie later in that
thread. I don't have the original source used to build this kernel
(2.4.19-FW25), but it appears that the patch has been applied in 2.4.20 at
least, so this may not be the same problem.
The full decoded oops is attached. Currently we have no way to reproduce
this problem.
Cheers, Chris.
--
___ __ _
/ __// / ,__(_)_ | Chris Wilson -- UNIX Firewall Lead Developer |
/ (_ / ,\/ _/ /_ \ | NetServers.co.uk http://www.netservers.co.uk |
\ _//_/_/_//_/___/ | 21 Signet Court, Cambridge, UK. 01223 576516 |
Hi all,> We encountered a problem by chance today, a kernel bug which causes the > following error: > > Assertion failure in journal_fl sh() at journal.c:1310: > "!journal->j_running_transaction" > kernel BUG at journal.c:1310!Looks like this was a hardware fault. Was able to duplicate it fairly reliably (about 1/2 the time) using the same disk, both on that system and another with the same hardware. Reformatting the disk made it unusable. Replaced the disk, with the same image on the same hardware, and now it works fine. Sorry for the noise. Cheers, Chris. -- ___ __ _ / __// / ,__(_)_ | Chris Wilson -- UNIX Firewall Lead Developer | / (_ / ,\/ _/ /_ \ | NetServers.co.uk http://www.netservers.co.uk | \ _//_/_/_//_/___/ | 21 Signet Court, Cambridge, UK. 01223 576516 |
Hi all,> We encountered a problem by chance today, a kernel bug which causes the > following error: > > Assertion failure in journal_fl sh() at journal.c:1310: > "!journal->j_running_transaction" > kernel BUG at journal.c:1310! > > This happened while remounting an ext3 filesystem read-only. We often do > this and haven't seen this bug before, so I suspect some kind of race.The first time we saw this problem, we were able to make it disappear by swapping the CompactFlash modules, so I put it down to a hardware problem. Now I've seen it happen again, so I thought I'd better report it as a possible kernel bug anyway. This time the oops trace is different, even though the result is the same; full ksymoops output is attached. Cheers, Chris. -- ___ __ _ / __// / ,__(_)_ | Chris Wilson -- UNIX Firewall Lead Developer | / (_ / ,\/ _/ /_ \ | NetServers.co.uk http://www.netservers.co.uk | \ _//_/_/_//_/___/ | 21 Signet Court, Cambridge, UK. 01223 576516 |