1 |
> On 24 Aug 2021, at 11:24, Jaco Kroon <jaco@××××××.za> wrote: |
2 |
> |
3 |
> Hi All, |
4 |
> |
5 |
> We run glibc based systems. No musl. But we don't use systemd. |
6 |
> |
7 |
> As little as a year back we still ran into issues with systemd-udev |
8 |
> variant breaking systems, the fix of course was to nuke it and install |
9 |
> eudev. Are we certain there is nothing (eg, LVM integration was our |
10 |
> biggest problem resulting in really crazy impossible to debug since we |
11 |
> can't log in due to lvn snapshot creation/removal deadlocking with |
12 |
> systemd-udev - no ask me not how, all I can tell you is that eudev never |
13 |
> exhibited this behaviour) will break? |
14 |
|
15 |
The problem is that this is a bit indirect. blueness could've easily |
16 |
ended up backporting whatever commit causes your issue, if it is |
17 |
indeed udev, because the idea wasn't to be frozen in time anyway; |
18 |
this just kind of happened accidentally because of time commitments. |
19 |
|
20 |
I appreciate this is going to be a huge pain to debug but reporting |
21 |
this upstream is the only proper fix here. |
22 |
|
23 |
> |
24 |
> Whilst I fully appreciate the difficult of all the various e* packages |
25 |
> (elogind, eudev etc ..) and I most certainly do not have the capacity to |
26 |
> maintain, and therefore I'm in full support of the concept of |
27 |
> deprecating eudev, I'm very, very worried about us suddenly being back |
28 |
> into the reboot-a-server-a-week scenario. In the worst case we've lost |
29 |
> some large filesystems almost certainly due to systemd-udev (we've had a |
30 |
> number of filesystem crashes which was recovered with fsck, but after |
31 |
> ditching systemd-udev and moving to eudev about two years back on this |
32 |
> specific host we've had ZERO further problems other than a failed drive |
33 |
> or two, none of which required a hard-reset to get back to a sane state). |
34 |
|
35 |
I don't doubt this happened as I know you're a persistent debugger, |
36 |
although my hope is that whatever issue you hit has been solved, especially |
37 |
given udev is used by Debian/Fedora/RHEL and all the rest of it. But I accept |
38 |
that if this was <= 1 year ago, that argument doesn't hold quite as much water. |
39 |
|
40 |
I suppose it'd be worth looking to see if there were any kernel or LVM2 regressions |
41 |
fixed around that time too. |
42 |
|
43 |
best, |
44 |
sam |