A few days ago my machine locked up hard, requiring pressing the reset button. This morning I noticed that a resync is going on. Here are the relevant messages in the system log (after boot): ----- Feb 23 21:16:09 e7 kernel: md/raid:md127: not clean -- starting background reconstruction Feb 23 21:16:09 e7 kernel: md/raid:md127: device sdg1 operational as raid disk 5 Feb 23 21:16:09 e7 kernel: md/raid:md127: device sdf1 operational as raid disk 4 Feb 23 21:16:09 e7 kernel: md/raid:md127: device sdd1 operational as raid disk 2 Feb 23 21:16:09 e7 kernel: md/raid:md127: device sdh1 operational as raid disk 6 Feb 23 21:16:09 e7 kernel: md/raid:md127: device sde1 operational as raid disk 3 Feb 23 21:16:09 e7 kernel: md/raid:md127: device sdc1 operational as raid disk 1 Feb 23 21:16:09 e7 kernel: md/raid:md127: device sdb1 operational as raid disk 0 Feb 23 21:16:09 e7 kernel: md/raid:md127: raid level 6 active with 7 out of 7 devices, algorithm 2 Feb 23 21:16:09 e7 kernel: md127: bitmap file is out of date, doing full recovery Feb 23 21:16:09 e7 kernel: md127: detected capacity change from 0 to 117187522560 Feb 23 21:16:09 e7 kernel: md: resync of RAID array md127 Then finally Feb 24 17:40:27 e7 kernel: md: md127: resync done. Feb 24 17:40:28 e7 mdadm[1194]: RebuildFinished event detected on md device /dev/md127 fsck of the fs on md127 was clean. I thought that the whole point of the logfile/bitmap is to avoid this and allow a quick recovery, and as such these files are written very carefully and safely. In the past, following a similar lockup/reset recovered quickly. bitmap was probably valid: Oct 3 21:25:21 e7 kernel: md/raid:md127: not clean -- starting background reconstruction Oct 3 21:25:21 e7 kernel: md/raid:md127: device sdf1 operational as raid disk 2 Oct 3 21:25:21 e7 kernel: md/raid:md127: device sdd1 operational as raid disk 4 Oct 3 21:25:21 e7 kernel: md/raid:md127: device sdi1 operational as raid disk 6 Oct 3 21:25:21 e7 kernel: md/raid:md127: device sdh1 operational as raid disk 0 Oct 3 21:25:21 e7 kernel: md/raid:md127: device sde1 operational as raid disk 3 Oct 3 21:25:21 e7 kernel: md/raid:md127: device sdg1 operational as raid disk 1 Oct 3 21:25:21 e7 kernel: md/raid:md127: device sdc1 operational as raid disk 5 Oct 3 21:25:21 e7 kernel: md/raid:md127: raid level 6 active with 7 out of 7 devices, algorithm 2 Oct 3 21:25:21 e7 kernel: md127: detected capacity change from 0 to 20003251814400 Oct 3 21:25:21 e7 kernel: md: resync of RAID array md127 Oct 3 21:25:23 e7 kernel: md: md127: resync done. kernel is 5.16.9-100.fc34.x86_6 The array uses an internal bitmap. What can lead to 'bitmap file is out of date'? -- Eyal Lebedinsky (eyal@xxxxxxxxxxxxxx)
