Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Mon, Feb 02, 2009 at 10:16:26AM +0100, Max Serafini wrote: > Thanks for your reply. > > I installed ubuntu server and configured drbd there using the same fibre > disks and surprise it works! > > I then built another fibre raid disk and on ubuntu is fine; on openfiler > 64 bit I get this: > > Jan 30 09:31:53 filer1 kernel: drbd1: disk( Diskless -> Attaching ) > Jan 30 09:31:53 filer1 kernel: drbd1: Starting worker thread (from cqueue [4561]) > Jan 30 09:31:53 filer1 kernel: drbd1: No usable activity log found. > Jan 30 09:31:53 filer1 kernel: drbd1: Method to ensure write ordering: barrier > Jan 30 09:31:53 filer1 kernel: drbd1: max_segment_size ( = BIO size ) =32768 > Jan 30 09:31:53 filer1 kernel: drbd1: drbd_bm_resize called with capacity == 975790304 > Jan 30 09:31:53 filer1 kernel: drbd1: resync bitmap: bits=121973788 words=1905841 > Jan 30 09:31:53 filer1 kernel: drbd1: size = 465 GB (487895152 KB) > Jan 30 09:31:53 filer1 kernel: drbd1: recounting of set bits took additional 4 jiffies > Jan 30 09:31:53 filer1 kernel: drbd1: 465 GB (121973788 bits) marked out-of-sync by on disk bit-map. > Jan 30 09:31:53 filer1 kernel: drbd1: disk( Attaching -> Inconsistent ) > Jan 30 09:31:53 filer1 kernel: end_request: I/O error, dev sda, sector 975820215 > Jan 30 09:31:53 filer1 kernel: drbd1: drbd_md_sync_page_io(,975820152s,WRITE) failed! > Jan 30 09:31:53 filer1 kernel: drbd1: meta data update failed! > Jan 30 09:31:53 filer1 kernel: drbd1: disk( Inconsistent -> Failed ) > Jan 30 09:31:53 filer1 kernel: drbd1: Local IO failed. Detaching... > Jan 30 09:31:53 filer1 kernel: drbd1: disk( Failed -> Diskless ) > Jan 30 09:31:53 filer1 kernel: drbd1: drbd_bm_resize called with capacity == 0 > Jan 30 09:31:53 filer1 kernel: drbd1: worker terminated > Jan 30 09:31:53 filer1 kernel: drbd1: Terminating worker thread > > Drbd version in OF is 8.2.7-2-1 on ubuntu is 2:8.0.11-0ubuntu3 > > Anything I can do on Openfiler drbd? Switch the question to Openfiler > forums? did you do what suggested? did you _verify_ those sectors? use dd, as indicated in my previous mail. verify the _storage_. it is your underlying _disk_ (sda, whatever that is) that reports back the io error, not DRBD. please verify those sectors. if that disk reports io errors to DRBD 8.0.11 or 8.2.7, it does not matter much, both will detach. if you want DRBD to use that storage of yours, you have to make sure that storage does actually work, and not error out. DRBD cannot magically heal whatever is reporting IO errors to it. find out why your sda gives back IO errors for those sectors. because it does. this is not DRBD imagining things. ok? ;-) -- : Lars Ellenberg : LINBIT | Your Way to High Availability : DRBD/HA support and consulting http://www.linbit.com DRBD® and LINBIT® are registered trademarks of LINBIT, Austria. __ please don't Cc me, but send to list -- I'm subscribed