Since some days I get the following message every day: mdadm: DeviceDisappeared event detected on md device /dev/md/md3 mdadm: DeviceDisappeared event detected on md device /dev/md/md2 mdadm: DeviceDisappeared event detected on md device /dev/md/md1 mdadm: DeviceDisappeared event detected on md device /dev/md/md0 mdadm: NewArray event detected on md device /dev/md0 mdadm: NewArray event detected on md device /dev/md1 mdadm: NewArray event detected on md device /dev/md3 mdadm: NewArray event detected on md device /dev/md2 This come from cron and the command `mdadm --monitor --scan --oneshot`. (Which produce the same message.) I never had any md device under /dev/md/*. They was always under /dev/mdX. And my /etc/mdadm/mdadm.conf has the following entries: ARRAY /dev/md0 metadata=0.90 UUID=xxxxxxxx:xxxxxxxx:xxxxxxxx:xxxxxxxx ARRAY /dev/md1 metadata=1.0 name=ikki:1 UUID=xxxxxxxx:xxxxxxxx:xxxxxxxx:xxxxxxxx ARRAY /dev/md2 metadata=1.0 name=ikki:2 UUID=xxxxxxxx:xxxxxxxx:xxxxxxxx:xxxxxxxx ARRAY /dev/md3 metadata=1.0 name=ikki:3 UUID=xxxxxxxx:xxxxxxxx:xxxxxxxx:xxxxxxxx /proc/mdstat has the following: ~> cat /proc/mdstat Personalities : [raid1] md2 : active raid1 sda6[3] sdb6[2] 419430260 blocks super 1.0 [2/2] [UU] bitmap: 0/2 pages [0KB], 131072KB chunk md3 : active raid1 sda7[0] sdb7[2] 419430208 blocks super 1.0 [2/2] [UU] bitmap: 1/4 pages [4KB], 65536KB chunk md1 : active raid1 sda5[0] sdb5[2] 1073741632 blocks super 1.0 [2/2] [UU] bitmap: 0/8 pages [0KB], 65536KB chunk md0 : active raid1 sdb1[1] sda1[0] 1998976 blocks [2/2] [UU] unused devices: <none> So it is strange where this "DeviceDisappeared" is coming from. Nowhere in my system is a mentioning /dev/md/mdX.
Hi, Am Sa den 8. Jul 2023 um 14:18 schrieb Daniel Baumann: Well, first, I don't have systemd and will never ever have! But it doesn't matter if Debian removed that cron job. It is mdadm itself that prints the error: ~> mdadm --monitor --scan --oneshot mdadm: DeviceDisappeared event detected on md device /dev/md/md3 mdadm: DeviceDisappeared event detected on md device /dev/md/md2 mdadm: DeviceDisappeared event detected on md device /dev/md/md1 mdadm: DeviceDisappeared event detected on md device /dev/md/md0 mdadm: NewArray event detected on md device /dev/md0 mdadm: NewArray event detected on md device /dev/md3 mdadm: NewArray event detected on md device /dev/md2 mdadm: RebuildStarted event detected on md device /dev/md2 mdadm: NewArray event detected on md device /dev/md1 So mdadm is broken in some sort. Regards Klaus
Hi, Am Sa den 8. Jul 2023 um 14:18 schrieb Daniel Baumann: Well, first, I don't have systemd and will never ever have! But it doesn't matter if Debian removed that cron job. It is mdadm itself that prints the error: ~> mdadm --monitor --scan --oneshot mdadm: DeviceDisappeared event detected on md device /dev/md/md3 mdadm: DeviceDisappeared event detected on md device /dev/md/md2 mdadm: DeviceDisappeared event detected on md device /dev/md/md1 mdadm: DeviceDisappeared event detected on md device /dev/md/md0 mdadm: NewArray event detected on md device /dev/md0 mdadm: NewArray event detected on md device /dev/md3 mdadm: NewArray event detected on md device /dev/md2 mdadm: RebuildStarted event detected on md device /dev/md2 mdadm: NewArray event detected on md device /dev/md1 So mdadm is broken in some sort. Regards Klaus
(Klaus, did you really mean to close this bug report with your last message? I'll go ahead and reopen it since I'm also affected by this bug.) From what I understand, this bug affects entries of the form `/dev/md0` in `mdadm.conf`, and was introduced by commit 84d969be, which now always looks for `/dev/md/$(basename)`, resulting in failures when attempting to access the non-existent `/dev/md/md0`. (Until this is resolved, a simple `../md0` symlink seems to do the job as a temporary fix.)
Hi, since some days, the warning gets even worse by naging me the name are not POSIX. But my names _are_ fully POSIX conform with only ASCII characters in the name! Regards Klaus
I'm also seeing this on my trixie system. A number of other users have reported this same issue elsewhere, particularly users of OpenMediaVault after upgrading to OMV 8 (based on trixie)[1][2]. It appears to be a regression introduced in 84d969be8f6d8a345b75f558fad26e4f62a558f6 upstream[3], which was first included in 4.2+20230223-1. This was later fixed by 456c6bed0b9e0866c75f0f7c8579d5f4f3f3e966 upstream[4], which was first included in 4.5-1. So in theory at this point, only trixie is impacted by this anymore. I tested applying 456c6bed0b9e0866c75f0f7c8579d5f4f3f3e966 onto the 4.4-11 sources on my trixie box (had to massage the patch a bit), and it appears to correct the issue. I didn't try updating to 4.5, though I would suspect that to work as well.--- [1]: https://forum.openmediavault.org/index.php?thread/58088-mdadm-devicedisappeared-event-detected-on-md-device-dev-md-md0-since-upgrade-to/ [2]: https://forum.openmediavault.org/index.php?thread/58015-md-email-every-morning/ [3]: https://github.com/md-raid-utilities/mdadm/commit/84d969be8f6d8a345b75f558fad26e4f62a558f6 [4]: https://github.com/md-raid-utilities/mdadm/commit/456c6bed0b9e0866c75f0f7c8579d5f4f3f3e966