Checkpoint deletion

hi,

Deleting Ckpt schedules will not delete the ckpts for the file system. you need to delete them manually if you would like to remove them.

any errors reported from server_log for the failed replication session   

Sameer Kulkarni

The only alert I'm getting is "Automatic extesion of the file system failed because it has no space to extend."

Event code : 0x13601b006c

Could you point me in the direction of some instructions on how to manually remove the unwanted checkpoints please.

Thanks

Nigel

nas_fs -delete -o umount=yes

yes, from putty you can list all checkpoints currently mounted on server_2

server_mount server_2 | grep ckpt

I assume this is from a putty session.

Do I just use that command at the from the interface when i login?

The thing is, I can't see the checkpoints in the GUI on the Replicas/Checkpoints page. I only know they're there when I look at the properties of the storage pools.

Nigel

if deleting a checkpoint requires root command, whats the name of the checkpoint?

Sure if this is not an IP replication checkpoint? If yes, is the replication session still running? ...

Then you would have to stop and delete it and the root_repl checkpoints are gone.

Almost there!

I login as nasadmin - can list the checkpoints. Attempt to run the delete command  - requires root command.

Type "su" to change to super user, re-run command - requires root command

open new session, login as root, re-run command - command not found

?????

Nigel

i am assuming you are not trying to delete replication checkpoints but regular checkpoints to free up space (which will be available to replication checkpoints)

In a word, yes.

Some of the replications have ceased working as the unit has run out of disk space. I have found that the checkpoints I originally set up, but decided were not required, are holding disk dspace that I now need. I therefore want to remove all checkpoints in one go to free up all available space for future needs.

Nigel

what are you trying to fix here ? Are you trying fix replication because it's does not have enough space ?

ok, so we need to delete regular checkpoints. If you run server_mount server_2 | grep ckpt, it will list all checkpoints, based on your naming convention you should be able to identify which checkpoints are regular and which ones are one for replication (there will be two replication checkpoints for each file system)

I wish to remove all unnecessary checkpoints from the system but the list does not corespond to any of our file system names.

e.g. root_rep_ckpt_110_86471_1 on /root_rep_ckpt_110_86471_1 ckpt,perm,ro,accesspolicy=NATIVE,nolock

------"-------"-----"-------------_2 -------"-----"---------"-----"---------_2 -----"-------"-------"----

and others without the accesspolicy.

Am I not ok to delete all of these?

those look like replication checkpoints, don't touch those.

fs_ckpt -list

this will list all checkpoints, excluding the two being used for replication

I had a similar issue. I stopped and deleted replication sessions and made sure its gone from both source and target but the root replication ckpt still exists on both source and target. I tried to delete them manually over CLI as root and it says 'NAS_DB environment not defined'

sudo nas_fs -delete root_rep_ckpt_99_1064545_2 -o umount=yes

NAS_DB environment not defined

Currently NAS_DB is set to /nas