On Sat, Jan 31, 2004 at 12:47:41AM -0800, Ben Rosenberg wrote:
* Alain Barthélemy (cassandre@bartydeux.be) [040131 00:13]:
During boot-up (SuSE-8.1) I always have the two next lines:
hda: dma_intr: status=0x051 { DriveReady SeekComplete Error } hda: dma_intr: error:0x841 { DriveStatus Error BadCRC }
I had a Maxtor that did this...it died a short time later. The only saving grace in what happened was it was the root drive and my home directory was on a seperate drive. I would do a backup of your data as soon as possible and then download any utils that the drive maker may have to test it.
Ben, I doubt it is the case here. I think this patch explains much what's going on: http://lwn.net/Articles/21923/ Look at this piece: + /* + * if a cache flush fails, disable ordered write support + */ + if (rq->cmd_flags & (RQ_WRITE_PREFLUSH | RQ_WRITE_POSTFLUSH)) { + struct request *real_rq = rq->special; + + /* + * best-effort currently, this ignores the fact that there + * may be other barriers currently queued that we can't + * honor any more + */ + if (err) { + printk("%s: cache flushing failed. disable write back cacheing for journalled file systems\n", drive->name); + blk_queue_ordered(&drive->queue, QUEUE_ORDERED_NONE); + } Drive is not going to fail, it just does not support cache flush Regards, -Kastus