Closed skandragon closed 5 months ago
PartialVolume
commented
5 months ago That's quite an old version. Are you able to build 0.35 or 0.35.6 (master) and test again. There's been a few updates to the serial number code since 0.31 which was released June 2021.
skandragon
commented
5 months ago Gotcha. This is from an Ubuntu apt repo, so I'll check to see if I want to upgrade it or not. I admit to being kinda lazy here, as my goal is really just to wipe these disks as a test to ensure they are still viable before throwing them into my home lab's Ceph.
PartialVolume
commented
5 months ago No problem, are you ok if I close this then, as I'm pretty sure that is fixed for SAS in the most recent version.
skandragon
commented
5 months ago Tested with latest source, works as expected. Would be great to get this in the ubuntu package system for those of us who are lazy. :)
Firminator
commented
5 months ago Yeah I was wondering, too, if we can get the nwipe package updated. PartialVolume do you happen to know the maintainers for the main distros?
PartialVolume
commented
5 months ago Only Debian (Martijnvanbrummelen) who I've been talking with about the next 0.36 release which is going into Debian Sid very soon and Fedora but I don't know if anybody updates nwipe in Ubuntu or it just feeds in from Debian.
A Ubuntu PPA might be useful for nwipe but I'm too overloaded to take on that too. Maybe somebody else could do that.
It looks like the serial number fetching is returning something odd, and it displays as gibberish in the UI.
sd[degh] are SAS drives connected to a SAS backplane, while the other disks are SATA and also attached to the same SAS backplane.
smartctl is able to retrieve the serials. hdparm returns an error:
SG_IO: bad/missing sense data, sb[]: 72 05 20 00 00 00 00 18 02 06 00 00 cf 00 00 00 03 02 00 01 81 0a 00 00 00 ff ff ff ff 00 ff ff