I've been doing some more failure testing to see how zfs handles problems; setup and background:
snv_86 supermicro x7dca-3 mb supermicro AOC-SAT2-MV8 disk controller (2, 4 500G sata drives on each in mirrored pairs across controllers) 2 500G sata on onboard ich9 controller in mirrored pair In trying to bound the system lockups on disk removal, I started a scrub to generate disk activity, then offlined one of a mirrored pair on the mv8 controllers and removed it, no problem. put it back in, ok, then told it to bring the drive back online. the ssh session echoes, but the online command did not return and a new ssh session timed out. disk activity continued for a bit, then stopped and the console was then unresponsive. had to reboot. drive was still offline, so told it to be back online and it thought all was well, so started a scrub to check and clean things up. after a bit, it decided there were too many checksum errors and marked it as degraded. meanwhile, the pool on the onboard controller failed to return. I did a "zpool import test" and it came back, however... This message posted from opensolaris.org _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
