I have multiple PCs with Fedora 35 and RAID-1 MDADM arrays.
Among them, only one is facing a weird problem:
If I leave it shutdown for more than 2~3 days, it will definitely perform an automatic check at the next reboot.
Here is the output result of journalctl -b | grep -E "(md1|/dev/md)":
Mar 14 23:06:49 zx-E5430 kernel: md/raid1:md127: active with 2 out of 2 mirrors
Mar 14 23:06:49 zx-E5430 kernel: md127: detected capacity change from 0 to 209582080
Mar 14 23:06:51 zx-E5430 systemd-fsck[586]: /dev/md127: clean, 236394/6553600 files, 3281182/26197760 blocks
Mar 14 23:06:51 zx-E5430 kernel: EXT4-fs (md127): mounted filesystem with ordered data mode. Opts: (null). Quota mode: none.
Mar 14 15:06:57 zx-E5430 kernel: EXT4-fs (md127): re-mounted. Opts: (null). Quota mode: none.
Mar 14 15:07:03 zx-E5430 systemd[1]: Started Timer to wait for more drives before activating degraded array md126..
Mar 14 15:07:04 zx-E5430 kernel: md/raid1:md126: active with 2 out of 2 mirrors
Mar 14 15:07:04 zx-E5430 kernel: md126: detected capacity change from 0 to 1743542272
Mar 14 15:07:04 zx-E5430 systemd[1]: [email protected]: Deactivated successfully.
Mar 14 15:07:04 zx-E5430 systemd[1]: Stopped Timer to wait for more drives before activating degraded array md126..
Mar 14 15:07:06 zx-E5430 systemd-fsck[847]: /dev/md126: clean, 24857/54493184 files, 4511205/217942784 blocks
Mar 14 15:07:06 zx-E5430 kernel: EXT4-fs (md126): mounted filesystem with ordered data mode. Opts: (null). Quota mode: none.
Mar 14 15:07:09 zx-E5430 mdadm[914]: RebuildStarted event detected on md device /dev/md/fedora_localhost-live00
Mar 14 15:07:09 zx-E5430 kernel: md: data-check of RAID array md126
Mar 14 15:07:09 zx-E5430 mdadm[914]: RebuildFinished event detected on md device /dev/md/fedora_localhost-live00
Mar 14 15:07:15 zx-E5430 kernel: md: delaying data-check of md127 until md126 has finished (they share one or more physical units)
Mar 14 15:07:15 zx-E5430 mdadm[914]: RebuildStarted event detected on md device /dev/md/fedora_localhost-live00
Mar 14 15:52:15 zx-E5430 mdadm[914]: Rebuild20 event detected on md device /dev/md/fedora_localhost-live00
Mar 14 16:40:17 zx-E5430 mdadm[914]: Rebuild40 event detected on md device /dev/md/fedora_localhost-live00
Note that the time gap within the logs was caused by dual systems (Windows 7 usually uses local time instead of UTC time) and thus should have nothing to do with this problem.
And here is the output result of mdadm -D /dev/md126:
/dev/md126:
Version : 1.2
Creation Time : Mon Aug 9 18:45:19 2021
Raid Level : raid1
Array Size : 871771136 (831.39 GiB 892.69 GB)
Used Dev Size : 871771136 (831.39 GiB 892.69 GB)
Raid Devices : 2
Total Devices : 2
Persistence : Superblock is persistent
Intent Bitmap : Internal
Update Time : Mon Mar 14 17:21:03 2022
State : active, checking
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Consistency Policy : bitmap
Check Status : 57% complete
Name : fedora_localhost-live00
UUID : 55508b5a:e2b47b55:aae847c6:0704ee20
Events : 16456
Number Major Minor RaidDevice State
0 8 2 0 active sync /dev/sda2
2 8 18 1 active sync /dev/sdb2
I've checked /etc/cron.d and found no entry related to MDADM. The SMART status of both 2 disks is all right as well.
At present, I can only cancel the checks manually by following commands:
sudo -i
echo idle > /sys/devices/virtual/block/md126/md/sync_action
echo idle > /sys/devices/virtual/block/md127/md/sync_action
Could someone please tell me that what is the reason that MDADM behaves like that?
