| === orndorffgrant4 is now known as orndorffgrant | ||
| === patrick_ is now known as patrick | ||
| keithzg | Hmm is there any chance some recent Linux security patches have borked some NVMe block devices? I had a server go down yesterday because suddenly the M.2 NVMe drive in it was running into errors and automatically re-mounting as read-only (oddly in an endless loop), and then I thought "well hey good excuse to take the parts for a fast new Windows machine sitting unbuilt at the office and claim it as one of my servers instead" and so I tried to set that up | 07:48 |
|---|---|---|
| keithzg | and somehow, despite running much newer parts that had never been even unpackaged before, the fresh install almost immediately ran into *the same problem*. Probably just bad luck, but! | 07:48 |
| keithzg | Both are AMD CPUs and WD SSDs, though over half a decade separates their models in time. The first server was running Ubuntu 22.04 with livepatching, the second I was freshly installing Debian 12.10. | 07:50 |
| lotuspsychje | keithzg: maybe its also a good idea to mention kernel version to volunteers, and compare a still good boot vs recent updates/kernel updates | 07:56 |
| lotuspsychje | and/or recent dpkg logs | 07:56 |
| keithzg | Yeah I was just in teh process of trying to get the logs but after a long day of battling this at work, having been interrupted right before I was gonna sleep, I'm only getting back to it now and from home | 08:07 |
| keithzg | And oddly enough `journalctl --directory=/where/I/copied/the/journal/to` from the original failing system loads without complaint but then isn't nearly to the latest date? | 08:08 |
| keithzg | Honestly in part I'm only now starting to think "wait, what if it's the kernel?" | 08:08 |
| keithzg | That I'm just unlucky and two entirely separate SSDs died in much the same way even though one wasn't plugged in at the time remains my leading theory ;) | 08:09 |
| keithzg | Aha, aha, "Journal file ./system.journal uses an unsupported feature, ignoring file.", that explains only loading old logs | 08:10 |
| keithzg | Last dpkg logs meanwhile are back on the 15th, which gestures heavily towards this NOT being any problem delivered by updates. | 08:17 |
| lotuspsychje | keithzg: wich is your current (failing) kernel version plz? | 08:25 |
| keithzg | lotuspsychje: Judging by dpkg/status, it musta been linux-image-5.4.0-212-generic | 08:36 |
| lotuspsychje | !info linux-image-generic jammy | 08:37 |
| ubottu | linux-image-generic (5.15.0.138.134, jammy): Generic Linux kernel image. In component main, is optional. Built by linux-meta. Size 3 kB / 21 kB. (Only available for amd64, armhf, arm64, powerpc, ppc64el, s390x.) | 08:37 |
| lotuspsychje | keithzg: on 22.04? | 08:37 |
| keithzg | lotuspsychje: Yeah, 22.04 jammy, I may be just reading that file wrong though too | 08:38 |
| keithzg | Wait lol yeah dumb mistake, I was in the directory where I'd backed up things but then used an absolute path...so I was looking at the host system, d'oh | 08:39 |
| keithzg | grepping the *correct* path I do see "Version: 5.15.0.136.134" for "Package: linux-image-generic" (which indeed reports as installed) | 08:40 |
| lotuspsychje | system up to date? | 08:41 |
| keithzg | I have (or I suppose had, heh) unattended-upgrades enabled in addition to the kernel livepatching, last updated ostensibly at 2025-04-15 at around noon UTC | 08:42 |
| lotuspsychje | keithzg: could you share your full dmesg in a paste, so the volunteers can take a look for you? | 09:02 |
| keithzg | lotuspsychje: sure, https://paste.ubuntu.com/p/5mq7Hw9wF9/ | 09:10 |
| keithzg | (That looks to be from the last reboot before this, back on April 10th) | 09:11 |
| lotuspsychje | kernel: nvme nvme0: missing or invalid SUBNQN field. | 09:15 |
| lotuspsychje | kernel: ahci 0000:02:00.1: SSS flag set, parallel bus scan disabled | 09:15 |
| lotuspsychje | keithzg: you got other kernel version you could try boot for tests? | 09:17 |
| keithzg | lotuspsychje: Yeah that's a solid thought, that's gonna be the first thing I try when I next trudge up to the office, there should be at least a few still installed. | 09:41 |
| keithzg | Gonna try to sleep again now though, I had been thinking "oh I should sleep already" when I noticed the outage at about 5AM local yesterday, and only got 4 hours of sleep once I finally got home again! | 09:42 |
| * keithzg is living at 03:42 at the moment, Mountain Time | 09:43 | |
| === JanC is now known as Guest4283 | ||
| === JanC is now known as Guest3211 | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!