EVA8000: overcommit ==> lost snapshots

[Van Kerckhove Piet]

All

we've got this event on our SAN:
"The virtual disk has
failed. Data was lost because of overcommit. Space efficient snapshot ran out
of space."

4 snapshots were dropped due to this error. There were no space problemes because there was still 15TB free in this diskgroup.

Does anybody know what's happened or what means 'overcommit' in this situation?

[Cass Witkowski]

I think what this means is that when you do a snapshot it does not make a copy of the data you are using. It only makes a copy of the pointers to the data. Only when you write to the original or snapshot disks does more space actually gets allocated. If you write a lot of data to the snapshot or original disk then you will chw up a lot of disk space.

You may want to think about using the snapclone instead of the snapshot since that will allocate the space upfront.

Regards,

Cass