VNX: Unable to create checkpoint of a file system (Replication Destination)
Summary: This article explains the workaround when the VNX is unable to create a checkpoint of a file system.
Symptoms
Checkpoint creation is unavailable.
Example:
[nasadmin@VNX ~]$ fs_ckpt FS1 -C operation in progress (not interruptible)...
Unisphere Creation might show a creation task running (and is unavailable) for ever.
File system is a replication target (Read-Only File system with Replication Internal Checkpoints exist)
Cause
Replication destination file systems have 2 internal (root) checkpoints.
During replication data transfer (whether manually refreshed or automatically pushed with new incremental data), Replicator service locks the SavVol and it releases the SavVol once the transfer is completed.
Some time, this may is unavailable for the following reasons:
- SavVol hit maximum space and unable to auto extend the
savvol - Interconnect is paused or some network issue between source and destination, which causes the replicator unable to finish the transfer and release the
savvol - Disk failed and resulting in file system corruption/unmount
Until the above conditions are resolved, no checkpoint activities like create, delete is possible for all kinds of checkpoints.
Resolution
Fix the issues explained in the Cause Section.
If still unable resolve, quote this KB to EMC Support for advanced methods which might require (be prepared):
- Data Mover (destination only) Reboot
- Replication might require deletion, and new full replication may be required to re-create