Change disk on EVA 3000 (8 Disks Disk Gorup)

[Goran Kubelka]

Dear All,

We have an alert on EVA 3000 that one FC -path on one FATA HDD is missing. The storage group is consist of 8 FATA Drives 250 GB. Protection level is 1. When we try to ungroup affected disk from disk group, to change the affected disk, the message appears like
"Disk cannot be ungrouped because disk group is under minimum disk drives..."
Which should be the steps for replacing disk drive without losing data or disk group?
Regards.

[Uwe Zessin]

If you have a free slot, you could add the new disk first and then remove the bad FATAl disk.

If you don't have a free slot, you might be able to get one free by ungrouping a disk drive from another disk group first - provided that there is enough free space or you temporarily lower the so-called "protection level".

If nothing helps, make a good backup, pull the disk. Wait until the recovery has finished (check the event log) - then put in the replacement disk and group it. You can put in / add the disk as soon as the recovery has finished - you don't need to wait for the data leveling to complete.

[Goran Kubelka]

Thank you Uve.
Actually I do not have any free disk slot. It is very strange behaviour. What will hapened if one disk really fails in a disk group consisted of 8 HDD?

[Dileep..]

Its not strange. thats the way it designed for. EVA need to have minimum 8 disk to function properly. If a disk fails in 8 Disk Disk Group, EVA firmware will manage to keep it runnning till you have replacement disk. So its suggested to replace the disk asap.

[Goran Kubelka]

I Agree.
But question is how to ungroup affected disk, and add a new one?

Command view reports an error:
"Operation Failed. Command rejected- Disks in the group would go bellow minimum required number"

[Patrick Terlisten]

Hello,

you can add the new disk, group it, and then ungroup the failed disk. This is the safe way. If you like it more riskier, just pull the failed disk, wait for the rebuild and add the new one. But, as Uwe already said: You should have a valid backup.

Best regards,
Patrick

[Uwe Zessin]

> What will hapened if one disk really fails in a disk group consisted of 8 HDD?

Well, the behaviour is the same as if the disk group had 9, 10, 56 or more (on a larger EVA) disks:

- if there is free space in the disk group, the EVA will use it to rebuild redundancy for VRAID-1 and VRAID-5 virtual disks [1]

- if there is no free space, but a "protection level" of single/1 or double/2, the EVA will use it to rebuild redundancy ...

- if there is no no free space at all to rebuild redundancy, the EVA is supposed [2] to continue and do normal "on-the-fly" RAID recovery for missing data. As soon as capacity is made available, the rebuild should start.


[1] there is NO redundancy for VRAID-0. If a disk fails, data is gone!!

[2] 'supposed', because I have never tried it myself ;-)

[laf_hc_chn_omc]

Hi All,

I am quite new to handling HP EVA Storage array. There is a disk failure in one of the disk group and protection Level is RAID 0 and the data is lost as per protection there will be no reducancy in RAID 0 Protection Level only performance.

Now we have add new disk and when I try to add the new disk group to the existing disk group in which the Disk failed.

I get below message:

The existing data on the disk will be erased. Can one suggest me please on this? 

Please provide your suggestion whether I can ignore and add the new disk to the disk group if not request you all please kindly share the steps to configure this new disk to the existing disk group.

ARRAY details

==========

1. EVA 3000 Storage

  HSV 100