On Tue, Jun 11, 2019 at 7:40 AM Mike Schwab <mike.a.sch...@gmail.com> wrote:
> The problem with SSD storage is that the blocks in a matched set of > device start to fail at the same rate. Run long enough and you get > enough bad blocks in all devices and they all fail at roughly the same > time loosing the entire raid group, you don't have enough time between > failures to replace and recreate before the next drive goes. Close > monitoring of the physical failure rate is important to start swapping > and rebuilding raid. > This can happen with HDs also. We (Windows LAN) had a catastrophic failure years ago. On the weekend, the LAN backups failed. The storage admin decided "no problem I'll fix it Monday". We had an HD in the array fail. "No problem, RAID will keep us going." While the CE was on his way in to replace the failed HD, a second drive in the array failed. No array. No current backup. Result: New storage administrator. > > https://news.ycombinator.com/item?id=12046524 > > -- This is clearly another case of too many mad scientists, and not enough hunchbacks. Maranatha! <>< John McKown ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
