Acopia Adaptive Resource Switch, Cli Maintenance Manual

Page: 121 / 324

Metadata Utilities: nsck and sync
De-Staging a Namespace’s Managed Volumes
CLI Maintenance Guide 5-41
The CLI prompts for confirmation before releasing the volume: enter yes to continue.
Enter no to cancel the destage operation.
You can run one nsck job at a time on a given namespace. Use
show nsck
to monitor
the progress of nsck jobs; see “Showing the Progress of nsck and sync Jobs” on
page 5-2 .
For example, the following command sequence exits to priv-exec mode and releases a
volume from the “wwmed” namespace:
bstnA6k(gbl)# end
bstnA6k# nsck wwmed destage volume /it
This operation will remove entries for all shares in the volume
from the namespace metadata. To reimport a destaged share,
enable the share in namespace-volume-share mode.
Destage all shares from volume /it in namespace wwmed? [yes/no] yes
bstnA6k# ...
Forcing the De-Stage
As mentioned in the section about rebuilding a namespace, it is possible for filer error
or network glitch to freeze an nsck job in an unfinished state. Once the error is
resolved, you may want to de-stage the volume or namespace; however, nsck only
allows a single job to run on a given namespace. To override the unfinished nsck job
and start a new one, add the optional
force
flag to the
nsck destage
command.
nsck name destage [volume volname ] force
For example, the following command sequence exits to priv-exec mode and forces a
de-stage for the ‘wwmed~/it’ volume:
For a short interval (roughly 10 seconds) all front-end NFS or CIFS services that host the
volume are inaccessible. This prevents access to
all volumes hosted by these front-end
services; that is, other volumes in the namespace could lose service for a short time. To
avoid this, you can remove the current volume from its services (see
“Stopping an NFS
Export” on page 11-4 or “Stopping a CIFS Share” on page 11-21 of the CLI
Storage-Management Guide ) before de-staging it, then add it back after the repairs are
finished.