Shrinking old group

Question

Hi

Just a quickie.

I have 5 members in an old group, they only have 1 controller attached and are on FW 6.0 - Uptime is like 3 years or something crazy so nobody has touched them and they have had no issues for ages. Setup isn't good, nothing to best practice.

Total storage is 60TB over the 5 members, 2 members will leave me with 23TB and enough IOPs to keep the last few remaining VMs running.

I plan to create a tmp pool and move each member into this pool then delete and re-purpose 3 of them, this time setting them up properly.

So procedure is move, wait for data to vacate then delete.

I have a few unknowns ( only 1 controller plugged into a single switch, and also the iscsi timeout settings on the initiators in VMware are set to the default 15 seconds, no MEM , single path.

Will any of these bad configurations affect the background member move/deletes?

IO is low, averaging about 500 IOPS across all 5 members.

Thanks

l4ndo · Answer

6.0.0 which I believe doesn't have the HB issue - its been fine for 3 years.

Anyway, the whole reason for shrinking is to re-provision and configure it properly, as cant update firmwares easily when on a single controller without downtime.

l4ndo · Answer

Well I've kicked off the move, its at 80% now.

Yes, all under contract so if something bad does happen I'll ring support.

Only 30 VM's on this group, and I have good backups. if the worse should happen it'll be a few hours to restore them all to our good arrays.

l4ndo · Answer

Once finished, I could just use this single member to create a new group (setup properly with good FW) and vmotion everything onto it, no downtime.

Also, this setup that is single path, no mem - a basic connection might not have this issue, that's why its been fine for 3 years.

Anyway, 98% complete now! :)

EqualLogic

Shrinking old group

Was this post helpful?