I've been fighting this same problem today in a nearly identical setup (supermicro x7dca3 mb, 936 chassis). I tried it with a couple drives attached to the onboard ich9 chipset, and at least it doesn't hang the entire system, and you get error messages, but anything that tries to access the volume hangs. They are interruptable if you wait for the 30 second timeout.
I moved the disk into another slot to simulate failure/replacement, but devfsadm does not find the "new" disk, nor does format see anything (and takes forever to start up, reporting the missing disk as "drive type unknown"). I tried offlining the disk, and got "no valid replicas"; I'm not sure if that freed up something (though the error implied that it *hadn't* succeeded in offlining the drive), or an internal timeout gave up, but the zpool status command that had been hung returned about then, and hasn't hung since, listing the drive as FAULTED, and the volume is accessible again (though degraded, of course). /var/adm/messages is still reporting early timeout errors as "Informational" errors every 30 seconds. Clearly this setup is not useable as is for the fileserver it was intended to be, and am hoping to find something that is... This message posted from opensolaris.org _______________________________________________ storage-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/storage-discuss
