Dave Kleikamp wrote: > Ideally, it would be best to avoid this. There may be cases where we're > part of the way into a transaction and without allocating memory, we > can't complete it, leaving the file system in an inconsistent state. It > isn't normal to have 0-order allocations fail.
I didn't think so - the box is a dual Xeon 500MHz with 1280Mb memory. The application does use quite a bit of memory, but would normally be less 1Gb. > Is there anything in the syslog to indicate why jfs mounted read-only? > There should be. Not much - and nothing that actually says 'jfs': May 13 07:47:51 gold Last message repeated 2 times May 13 07:47:51 gold kernel: __alloc_pages: 0-order allocation failed (gfp=0x1d0/0) May 13 07:47:51 gold kernel: ERROR: (device sd(8,1)): txAbort May 13 07:47:52 gold kernel: __alloc_pages: 0-order allocation failed (gfp=0x70/0) May 13 07:47:52 gold kernel: __alloc_pages: 0-order allocation failed (gfp=0x1f0/0) May 13 07:47:52 gold kernel: __alloc_pages: 0-order allocation failed (gfp=0x1d0/0) May 13 07:47:52 gold kernel: __alloc_pages: 0-order allocation failed (gfp=0x1f0/0) an hour later I see another couple of "ERROR: (device sd(8,xx)): txAbort". At this point the memory situation looked ok, and I could also do a normal shutdown. I'm wondering if the txAbort indicates a hardware problem, except I've seen no other indications. -- /Per Jessen, Z�rich ------------------------------------------------------- This SF.Net email is sponsored by Oracle Space Sweepstakes Want to be the first software developer in space? Enter now for the Oracle Space Sweepstakes! http://ads.osdn.com/?ad_ids93&alloc_id281&op=click _______________________________________________ Jfs-discussion mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/jfs-discussion
