Comment 3 for bug 863736

Revision history for this message
Arthur Peters (amp) wrote :

This bug has happened again. This is only the second time, and I'm really not sure how to replicate it. However there does appear to be a deadlock (or a crash with a lock held) happening.

This trace has a couple of threads that are a little past queue_log_writer:

One has a frame like "[<ffffffffa0384eec>] ? open_xa_root+0x8c/0xf0 [reiserfs]" after queue_log_writer.
Another has a frame "[<ffffffff81195486>] ? mark_buffer_dirty+0x66/0xa0".

The first program that froze (that I noticed) was emacs (3308).