On Wed, Nov 18 at 14:20, Karl Katzke wrote:
The devices we're talking about are c1t15d0 and c1t14d0 -- corresponding to these entries in iostat:---- errors --- device s/w h/w trn tot ... sd27 2 15 31 48 sd28 1 4 8 13 With the caveat that sd27 has been replaced as of this morning without the iostat being reset; however, errors have occurred since the replacement as below.
Interestingly, the blog post you referenced said that an LSI driver change eventually resolved their issue, and disabling FM was only a temporary measure. Not sure about your configuration, but if you exported the pools, moved the SSDs to different ports, and imported the pool, it would be interesting to see if the faults followed the ports or the drives. If it follows the drives, there's something fundamental in the interaction of the controller firmware, driver stack and device firmware that's bad, since it's unlikely a brand new replacement would have the same failure mode. If it follows the ports, then you're down to swapping the HBA or cabling/backplane. When people report their issues are solved, it'd be neat to see what the actual "root cause" was from that referenced post. --eric -- Eric D. Mudama edmud...@mail.bounceswoosh.org _______________________________________________ storage-discuss mailing list storage-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/storage-discuss
