HP StorageWorks Scalable File Share - Manual (Page 285)

Recovering degraded MDS services on systems using EVA4000 storage 9–61
For more information on reviewing SFS20 array information, see Section 6.1.4.2.
If, after further investigation, you decide to replace a disk that is logging URE errors, see Section 8.1.10 for
more information.
TIP: You can use the
diskinfo
wrapper script for the
hpls_cciss_info
command to report the drive
status on all SFS20 arrays attached to a server. To run the
diskinfo
script, enter the following command
on the server:
# /usr/opt/hpls/diags/bin/diskinfo -v
The command reports disk and controller information for all arrays attached to the server. The
-v
argument
is optional; if you do not use it, the command provides the report in a shorter summary format.
9.35 Recovering degraded MDS services on systems using EVA4000
storage
If an MDS service using an EVA4000 storage array that temporarily failed was mirrored, the RAID status of
the service may be running in
degraded
mode. The
show filesystem
command displays the state of
the service, as shown in the following example:
sfs> show filesystem
Name State Services
------ -------------- ----------------------------------
data started mds8: running(raid: degraded), ost[29-30]: running
In this example, the
mds8
service is shown to be running in
degraded
mode. This means that one mirror
device is operating normally and the other has failed.
Note the following points:
• If the disk group itself has failed, rebuild the disk group, create a new virtual disk and present it to the
servers before starting the recovery process. Use the
set lun
command to set the
role
and
used
by
values appropriately.
• In most cases where services are degraded, the disk group has not failed. Instead, either a controller
has failed or the Fibre Channel fabric between the controller and the server has failed. You can start
the recovery process described below as soon as both component LUNs are visible (as shown by the
show lun
command), even if the failed controller module has not yet been repaired (the LUN should
be visible through the other controller).
You can recover the service and restore it to correct operation as follows:
1. If the file system is stopped, start the file system.
2. Identify the LUNs that the MDS or OST service is based on, by using the
show filesystem
filesystem_name
command as follows. In this example, the
mds8
service is using the mirrored
LUN number 41:
sfs> show filesystem data
.
.
.
MDS Information:
Name LUN Array Controller Files Used Service State Running on
----- --- ----- ---------- -------- ---- ------------- ----------
mds8 41 1 ab 2.3M 20% running south2
OST Information:
Name LUN Array Controller Size(GB) Used Service State Running on
----- --- ----- ---------- -------- ---- ------------- ----------
.
.
.
ost29 34 3 a 290 30% running south4
.
.
.