Using a different Storage Pool for checkpoints?

Hi,

the SavVol is a volume that holds the checkpoints, and is automatically created on the first ckpt creation.
By default, it uses the same pool as the PFS ( production filesystem ).
Although, you can specify the pool you want to create the ckpts on both CLI ( fs_ckpt )and GUI.
All checkpoints will be created on this volume afterwards.

So, to answer your question :

if you have existent checkpoints for this PFS, unfortunately you need to delete them, and create the first one specifying the pool clarata_r6.

If you don't have existent checkpoints, just create the first one manually, then create your ckpt schedule.

Please, let me know if this answer your question.

Gustavo Barreto.

I'm ok with recreating the checkpoints, I just dont recall ever seeing where you can put the checkpoint on a different storage pool.

I created a new FS and when I create a checkpoint I see where you can create it on the desired storage pool.

On the FS's that I want to recreate the checkpoints though, I do not have those same options.

We are doing Replications, would those checkpoints need to be deleted in order see the same options on the newly created FS?

Hey Sandip. We are answering on the same minute !...
At least, we are consistent....

As Gustavo mentioned earlier, all the checkpoints of a given FS use the same storage pool as their SavVol. This includes checkpoints for Replication sessions as well. So, you need to ensure that the replication checkpoints are not used as well.

As you have already seen, the option to choose the "Storage Pool" is only available for creating the very first check point of a File system.

Hope this helps,
Sandip

Yes...the replication creates internal checkpoints, and by default they are created on the same pool as the PFS, as a normal checkpoint.

In this case, you would need to delete then recreate also the replications.

Sorry Gustavo - I saw this thread earlier but since you were replying to it, I was hesitant to reply - but after waiting for few mins and not seeing your reply further, I thought you may be busy and thus posted my comment.

It's good to see that both of us have some time today to participate in forums at the same time

But more amazing is the timing of the post - on both the recent threads our timing is exactly the same ..

Cheers,
Sandip

before upgrading IP Replication v1 to v2 i had teh configuratione done like described now. after the upgrade to ip replication version 2 i'm not sure if this still is the case.

basically, i would like to have the checkpoints for replication done on the same PFS. I also have a pool_VAULT which is almost full and i wondered why. does anyone know a command which shows me if there are checkpoints on this pool_VAULT (i guess that in a earlier version of the celerra firmware the pool_VAILT was used for the checkpoints)

To see what filesystems are built on any particular pool, run the following command:

nas_pool -i {name_of_pool}

So, for example:

[nasadmin@Butterfinger ~]$ nas_pool -l
id      inuse   acl     name
3       y       0       clar_r5_performance
36      n       0       clar_r10
 
[nasadmin@Butterfinger ~]$ nas_pool -i id=3
id                   = 3
name                 = clar_r5_performance
description          = CLARiiON RAID5 4plus1
acl                  = 0
in_use               = True
clients              = root_fs_vdm_vdmNICK,emc_test
members              = v93,v97
default_slice_flag   = True
is_user_defined      = False
virtually_provisioned= False
disk_type            = CLSTD
server_visibility    = server_2,server_3
volume_profile       = clar_r5_performance_vp
is_dynamic           = True
is_greedy            = True
[nasadmin@Butterfinger ~]$

You can see in the list above the following line which indicates the filesystems built on that pool:

clients              = root_fs_vdm_vdmNICK,emc_test

hth

Also, replication will use the same SavVol as any other checkpoints built for that filesystem. So if you already have checkpoints on a separate pool from that of the filesystem, replication checkpoints will be built on that same pool.

In order to get replication checkpoints on the pool you want them, your filesystem has to have at least one checkpoint on the desired pool, and your replication checkpoints will then use that pool as well.