Hung ASM Disk

From: alek <alexandru.tica_at_gmail.com>
Date: 10 May 2006 01:59:52 -0700
Message-ID: <1147251592.516513.244790@y43g2000cwc.googlegroups.com>

Hi,

We have RAC database comprised with 2 nodes and a 2 SAN shared devices. The OS si Linux AS4 and the Oracle version is 10gR2. The ASM instance is configured like this:

a DATA DISKGROUP with (DATA1 disk (from the SAN1) and DATA2 dis (from the SAN2))
a FLASH RECOVERY AREA DISKGROUP with (FRA1 disk from ths SAN1 and FRA2 from the SAN2)).

The diskgroups use normal redundancy. We've simulated a crash of a SAN1 device and we can see now that the DATA1 and FRA1 disk are in the HUNG state (as the Grid Controler shows). After we enabled once again the SAN1 device the status of the DATA1 and FRA1 disks is not changing. We forced a rebalance operation but it was useless. The workaround we founded was to add a third disk and to delete the HUNG one but this means that we always must have a third disk available for such cases.

Have anyone encounter such problems with the ASM in a RAC configuration?

TIA.
alec. Received on Wed May 10 2006 - 03:59:52 CDT