Serviceguard Extended Distance Cluster (EDC) with VxVM/CVM Mirroring on HP-UX, May 2008

• If the storage ISL remains unavailable, choose the site at which the last complete mirror of each
volume resides as the remaining site, and halt the nodes at the other site. See the steps in Restore
the Inter-Site Link and restart the detached site for procedures to make the LUNs visible to the
remaining nodes at the OS and DMP level.
• Each disabled volume needs to be recovered individually so the nodes from the remaining site will
work with the storage from the same site. The following steps were performed during one of the
tests and are only shown as an example. There might be different and/or additional steps needed
to complete a recovery depending on the exact condition the volumes are left in:
– Disassociate DCL volumes (containing DRL and FMR maps) from their parent volumes. This is
necessary, because their content will be invalid after the plexes of the parent volumes are
disassociated in a later step:
vxsnap -g <dg> -f unprepare <vol>
– If disks on the local site are marked as “FAILING”, it is necessary to reset their state:
vxedit -g <dg> set failing=off <disk>
– Disassociate all plexes from either site:
vxplex –g <dg> dis <plex>
– For each volume, attach the plex from the remaining site which contains the last complete mirror
back to the volume:
vxplex –g <dg> att <vol> <last-complete-mirror_plex>
– Start each volume:
vxvol –g <dg> start <vol>
• If the storage ISL is restored, the nodes and storage from the remote site can be added to the
cluster. Most of the steps are the same as the steps described in
Restoring a failed site
in a prior
chapter. There might be different or other steps required. Some additional steps which were
identified during tests are provided as an example:
– Before mirror resynchronization can be started, the disassociated plexes from the remote site must
be re-attached to their volumes:
vxplex –g <dg> att <vol> <plex-from-remote-site>
– Re-create a mirrored DCL volume for each volume to hold the DRL and FR maps:
vxsnap –g <dg> prepare <vol> ndcomirs=2 drl=on
and verify that each mirror is located at a different site.
VxVM/CVM 5.0 Tip:
With site consistent detach configured for the disk groups, CVM disables
all plexes on the site which encounters the first I/O error to any of the disk
group’s plexes. This will ensure that all plexes that contain the last complete
mirror for all volumes of the disk group will reside at the same site.
To reduce the probability of such a failure scenario, it is highly recommended to have a redundant
and common ISL for both heartbeat and storage traffic. This means the same redundant physical
media transports the heartbeat and the storage traffic.
22