My company has an Open-E based file server (Celeros) that uses near-line SAS drives in RAID6 configuration managed by an LSI 9266-8i RAID card. We currently have one "dedicated hot spare" drive, no "global hot spare" drives, and one very strange "commissioned hot spare" drive that appears to be built into the array. The array is comprised of 15 drives, 16 if you include the global hot spare.

The MegaRAID Storage Manager interface shows those 15 drives as part of the array and the dedicated hot spare as a separate drive. The virtual drive comprised of 15 drives shows total capacity as 23.645TB, which looks to be 15 drives x 1.819TB each, minus 2 parity drives for RAID6. As far as I can tell, the drive labeled "commissioned hot spare" is counted as part of the total capacity of the array. On the information page for that drive, it says "Commisioned spare: yes" and "emergency spare: no". This is very confusing, and I've been looking for days for clarification. I'm hoping someone out there has run into this and figured out what it means...

My question is whether this drive is part of the array proper, or if it is still a hot spare and can be safely removed. We are looking to use the spare disk slots in the machine to build a second array with bigger drives and migrate our data to the new array. That extra slot would come in handy for the migration, and I need to know if it is safe to remove it from the array without triggering a rebuild or migration... The company that made the file server seems to not know, and there is virtually nothing for documentation on "commissioned" hot spares by LSI, only global and dedicated hot spares...


Based on looking through some datasheets here, here, and here, it would appear that LSI uses the term "commissioned" to describe a hot spare that has been put into active duty, for one reason or another.

In general, your analysis of the array capacity is probably the most reliable fact here - if the array capacity only makes sense when you include that drive's capacity, then that drive is probably an active member of that array.

Based on both of these pieces of information, I would confidently say that if you remove that drive, the array is going to fall into a degraded state. Since you're using RAID6, you wouldn't have too much to worry about, but you should definitely figure out how this happened in the first place. My best advice would be if you still have active support for this setup, get on the phone with the vendor until they give you an answer.

|improve this answer|||||
  • 1
    Thanks for the info. I was about 90% sure that was what was meant based on the analysis of the virtual drive's capacity. I have re-planned my data migration to work without the need for that drive because there was no way to be sure and I want to save resources for the migration instead of triggering a rebuild. – user167684 Nov 14 '15 at 19:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.