From: Matthew Moncaster (mmoncaster@gmail.com)
Date: Fri Mar 03 2006 - 11:26:12 EST
Sorry I'm not an expert on any of this and we have no documentation:-)
so I may be missing very simple stuff...
We have an E3500 running solaris 8, just patched to latest recommended
cluster patch, with 2 JNI (sadly) 146x HBAs connecting them to a
Storagetek (a.k.a LSI) d178 san via 2 Brocade 3800 switches.
One of them is no longer working, it seems like a hardware issue, as
the switch says NO_LIGHT (even when I swapped in a known good cable)
on that port. In theory they are using rdac but its not helping i.e.
no failovers have taken place...
This is preventing me accessing 2 of the 6 "volumes" in Santricity
terms, or several concat/stripes in metastat terms and indeed whole
filesystems. Connection attempts appear in dmesg like:
Mar 3 14:30:30 orchard corrupt label - wrong magic number
Mar 3 14:30:30 orchard scsi: [ID 365881 kern.info] Vendor 'STK', product
'
PENstorage', 143292416 512 byte blocks
Mar 3 14:30:30 orchard scsi: [ID 107833 kern.warning] WARNING:
/sbus@2,0/fcr@2
400000/sd@0,5 (sd383):
Mar 3 14:30:30 orchard i/o to invalid geometry
Mar 3 14:30:30 orchard rdriver: [ID 486355 kern.notice]
ID[RAIDarray.rdriver.4
03] The Array driver is returning an Errored I/O, with errno 5, on Circle 33
D1
8, Lun 5, sector 16
Mar 3 14:30:30 orchard md_stripe: [ID 641072 kern.warning] WARNING: md: d93:
r
ad error on /dev/dsk/c5t0d5s4
Maybe the volumes are just toast but I doubt it I suspect the jni card
and/or drivers but don';t know how to troubleshoot really.
Is there a quick way to try and access these volumes via the other
path, or any troubleshooting steps I can take?
Thanks very much for any help.
-Matt
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:39:11 EDT