Date: Wed, 12 Feb 2003 01:57:52 -0700 From: Scott Long <scott_long@btc.adaptec.com> To: Josh Brooks <user@mail.econolodgetulsa.com> Cc: freebsd-scsi@freebsd.org Subject: Re: confusing aaccli output Message-ID: <3E4A0C90.5000600@btc.adaptec.com> In-Reply-To: <20030212001508.U39971-100000@mail.econolodgetulsa.com> References: <20030212001508.U39971-100000@mail.econolodgetulsa.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Josh Brooks wrote: > Hello, > > I am seeing this in aaccli: > > AAC0> container list > Executing: container list > Num Total Oth Chunk Scsi Partition > Label Type Size Ctr Size Usage B:ID:L Offset:Size > ----- ------ ------ --- ------ ------- ------ ------------- > 0 Mirror 34.1GB Open 0:01:0 64.0KB:34.1GB > /dev/aacd0 array01 0:00:0 64.0KB!34.1GB > > 1 Mirror 34.1GB Open 0:02:0 64.0KB:34.1GB > /dev/aacd1 array02 0:03:0 64.0KB:34.1GB > > > At first glance it looks fine. Two healthy containers. But what is that > exclamation point: > > 0:00:0 64.0KB!34.1GB > > Interesting. Now I look at /var/log/messages: > > > > Feb 9 17:26:36 system /kernel: aac0: **Monitor** Drive 0:0:0 returning > error > Feb 9 17:26:36 system /kernel: aac0: **Monitor** ID(0:00:0) - drive > failure (retries exhausted) > Feb 9 17:26:36 system /kernel: aac0: **Monitor** Mirror Container 0 Drive > 0:0:0 Failure > Feb 9 17:26:46 system /kernel: aac0: **Monitor** Mirror Failover > Container > 0 no failover assigned > > > > So what is going on here ? messages tell me I have a broken mirror, but > aaccli shows me a healthy one. As noted above, the card took scsi driver 0:0:0 offline due to errors, so the mirror is running in degraded mode. I admit that the output from aaccli isn't terribly obvious here... > > Iam afraid to run a: > > container set failover > > because aaccli shows it as already belonging to the container... Eh? 'container set failover' allows you to assign one or more drives as either global or dedicated spares. The firmware won't really let you add the failed drive as a spare (not without some tickery, at least), so you'll need to shove a new drive into the mix and assign it as the spare. You'll also need to enable automatic rebuilds, and then everything will be good to go. If you're really in a bind because you didn't budget for a spare drive (shame on you!), pull out the drive that was marked as failed, put it on a regular scsi controller, do a low-level format of it, then put it back on the aac controller and add it as the spare. There's a good chance, though, that if the drive failed once then it's going to fail again. Scott To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3E4A0C90.5000600>