Subscribe
Accepted Solution

There are insufficient spare disks

[ Edited ]

Simple question:

 

If a have one aggr in raid-dp without spare disks.

Which probable impact if one disk fail ? or other problems?

 

thanks!

Re: There are insufficient spare disks

How Data ONTAP handles a failed disk that has no available hot spare

When a failed disk has no appropriate hot spare available,  Data ONTAP puts the affected RAID group into degraded mode indefinitely  and the storage system automatically shuts down within a specified time  period.

If the maximum number of disks have failed in a RAID group (two for  RAID-DP, one for RAID4), the storage system automatically shuts down in  the period of time specified by the raid.timeout option. The default timeout value is 24 hours.

To ensure that you are aware of the situation, Data ONTAP sends an  AutoSupport message whenever a disk fails. In addition, it logs a  warning message in the /etc/message file once per hour after a disk fails.

Attention: If a disk fails and no hot spare disk is available, contact technical support.

Re: There are insufficient spare disks

Hi all,

Regarding this thread, we have the following question:

In the scenario of no spare disks available, when only one disk fails on a RAID-DP, does the storage system shutdown automatically after a given timeout?

If so, is this timeout configurable?

It is clear that, if two disk fail on a RAID-DP, the system shuts down after "options raid.timeout" has passed.

Re: There are insufficient spare disks

The man page for raid.timeout (in na_options) implies that it ONLY applies to single or double drive failures in raid4 or raid-dp raidgroups, but I'm unable to find anything that explicitly says that a raid-dp group will run indefinitely with a single drive failure....

Re: There are insufficient spare disks

Hi Bill

That is exactly our doubt.

Does ONTAP run forever with a Raid-DP with a single failed disk?

Re: There are insufficient spare disks

I ran this question by a couple SEs I know (collectively they have something like 10 years working for NetApp).   When a raid-DP raid group suffers from a single drive failure, it is NOT in degraded mode and the raid.timeout option does NOT apply.  It will run indefinitely with a single drive failed.

It'd be nice to have some backing documentation, but lacking that, I take their word as gospel....

Bill

Re: There are insufficient spare disks

Hi Bill

Thank you so much for the answer. We think the same as you.

The main problem at NetApp documentation is this paragragh:

    https://library.netapp.com/ecmdocs/ECMM1278264/html/smg/provisioning/concept/c_oc_prov_spares-fail-wo.html

When a failed disk has no appropriate hot spare available, Data ONTAP puts the affected RAID group into degraded mode indefinitely and the storage system automatically shuts down within a specified time period.

We think that this statement (1 Failed Disk with no spare Disk implies Raid Group degraded) is only correct for RAID-4 but not for RAID-DP.