#Upgrade ONTAP with broken/failed drive?
1 messages · Page 1 of 1 (latest)
awesome.. i thought that needed to be the case..
We are updating large clusters with some broken disks (old nodes in cluster) all the time, works, but not recommended for obvious reasons.
With many 1000s of disks, the time window of no broken disks is very narrow.
I've seen it turn into a pile of hurt more times than I've seen it work. YMMV
our advice is to not perform ontap upgrades with broken disks present
you can do the update with broken disks. The worst thing that could happen is that the reconstruct of the broken disk might re-start from scratch but that is rarely a problem
Again, my customers experiences have not been good doing it. Perform ontap upgrades with failed disks present at your own risk
Unlike the big I storage vendor which refused to do an upgrade with a broken drive. More than once, I had a drive fail the night before an upgrade maintenance window and had to cancel my change.
I was mid-headswap once where one of the root aggr disks failed. that was fun.
20 node cluster minor update done for SMB signing issue. 12 broken disks, maybe 1 failed after the update. Only issues were LIFs failing over to wrong (not HA partner) nodes and one node stuck in LOADER, simple as always.