A hard drive controller on the Pinboard search server is shuffling off its mortal coil, making its way to the Great Cloud in the sky. Search may be affected while I move it over to a less drama-prone machine
-
-
Replying to @Pinboard
Controller failures are the best, I mean, the worst, especially if they initially present as weird IO slowdowns or a HD dropping out of RAID.
1 reply 0 retweets 0 likes -
Replying to @getyrtrouserson
My symptom is some cryptic ata2.0 failure in the syslog, and processes starting to hang on I/O until the server becomes unresponsive. I think I'm going to need to call for help on here and hire someone to help me figure out what to replace
1 reply 0 retweets 2 likes -
Replying to @Pinboard
That sounds like the one time the SATA/SAS controller in a Supermicro 4U went bad on me. I had a cold spare so could swap across the drives and rebuild/verify the array, but it was a long night at the datacenter. Or per
@SwiftOnSecurity it might be fucking cables going bad.1 reply 0 retweets 2 likes -
In that server, the controller was a daughterboard and a relatively easy swap once everything was offloaded, but that was in the Before Times when you could get a part overnighted.
1 reply 0 retweets 2 likes
Thank you, that is helpful! The failing drives are a RAID-1 mirror and claim to be healthy, so I suspect the controller even more. Now to find out what on earth it is, and how to order a suitable replacement
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.