Hello,
We had a disk failure, I attempted to replace the disk with servicemag start -pdid 6. It was successful at this point. I tried replacing the disk with a new one and it failed with the disk not normal. I tried to dismisspd and it had referenced chunklets. I manually moved off all the chunklets and was able to dismisspd the disk.
I added a new physical disk, did admitpd and it found a new disk, but it still has the same WWN? I try a servicemag resume 0 2 and it gets stuck checking for a valid disk. It eventually fails with disk not normal.
I tried the unmark and clearstatus for servicemag to reset it all. I re-ran the servicemag start -pdid 6, and it runs fine. I again try the new physical disk, and it continues to fail for checking for a valid disk, and reports the old WWN/disk #6 as not normal and servicemag resume fails.
Any ideas? We do not have support on the array any longer.
P10000/V400 can't replace disk?
Re: P10000/V400 can't replace disk?
If you vacated and dismissed the drive you shouldn't do servicemag. You should only do admitpd/admithw.
One question I have is, what type of drive is this and what did you do when you replaced the drive (considering you had 3 healthy drives in the mag you pulled to access the failed drive.
One question I have is, what type of drive is this and what did you do when you replaced the drive (considering you had 3 healthy drives in the mag you pulled to access the failed drive.
The views and opinions expressed are my own and do not necessarily reflect those of my current or previous employers.
Re: P10000/V400 can't replace disk?
This is the standard 600GB 15k disks recommended by HPE. Same part # as what I am replacing.
I did a servicemag start -pdid 6. It ran fine. I replaced the disk and tried servicemag resume 0 2. The 3 good disks i left alone in the magazine. Servicemag resume failed. I put back in the original, failed disk and tried to resume. That servicemag resume failed.
I then tried to dismisspd on 6. It referenced a bunch of chunklets I manually moved off.
I then tried dismisspd again and it worked. I then put back in the new disk and tried to admitpd/admithw. It adds a new disk, with the same ID and the same WWN. I try to do a servicemag resume and it fails with the mentioned errors.
I did a servicemag start -pdid 6. It ran fine. I replaced the disk and tried servicemag resume 0 2. The 3 good disks i left alone in the magazine. Servicemag resume failed. I put back in the original, failed disk and tried to resume. That servicemag resume failed.
I then tried to dismisspd on 6. It referenced a bunch of chunklets I manually moved off.
I then tried dismisspd again and it worked. I then put back in the new disk and tried to admitpd/admithw. It adds a new disk, with the same ID and the same WWN. I try to do a servicemag resume and it fails with the mentioned errors.
Re: P10000/V400 can't replace disk?
mujzeptu wrote:This is the standard 600GB 15k disks recommended by HPE. Same part # as what I am replacing.
I did a servicemag start -pdid 6. It ran fine. I replaced the disk and tried servicemag resume 0 2. The 3 good disks i left alone in the magazine. Servicemag resume failed. I put back in the original, failed disk and tried to resume. That servicemag resume failed.
I then tried to dismisspd on 6. It referenced a bunch of chunklets I manually moved off.
I then tried dismisspd again and it worked. I then put back in the new disk and tried to admitpd/admithw. It adds a new disk, with the same ID and the same WWN. I try to do a servicemag resume and it fails with the mentioned errors.
Why are you trying to do servicemag if the disk was successfully added to the array with admitpd/admithw? What does checkhealth say? My guess is that you shrinked and expanded the array with one drive and need to do tunesys to get data back onto the drive.
Is this a 3.5" FC drive or a 2.5" SAS drive using FC<->SAS adapter?
The views and opinions expressed are my own and do not necessarily reflect those of my current or previous employers.
Re: P10000/V400 can't replace disk?
I am trying to bring that magazine back into production and make all 4 disks active again? It shows the servicemag status as still active so I am attempting to resume it and let the data migrate back over?
This array isn't in the best shape... Node 0 currently has a battery failure and PS failure. I attempted another admithw and it put my LD's/PD's in degraded state..... I then waited and did another admitpd and it found it again. I tried another servicemag resume and this time it took and is rebuilding?! No idea, but I am going to let it finish and see what happens.... thank you for all the help!
This array isn't in the best shape... Node 0 currently has a battery failure and PS failure. I attempted another admithw and it put my LD's/PD's in degraded state..... I then waited and did another admitpd and it found it again. I tried another servicemag resume and this time it took and is rebuilding?! No idea, but I am going to let it finish and see what happens.... thank you for all the help!