Unintelligible errors: Unknown block error, I/O error, critical target error

[amano-kenji]

Disk errors
1 2022-10-31 22:36:26 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
2 2022-11-01 01:08:50 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
3 2022-11-01 02:39:32 +0000 error: dev=0:0, sector=-1, nr_sector=0, error='I/O error', rwbs='N', cmd='',
4 2022-11-01 02:39:32 +0000 error: dev=0:0, sector=-1, nr_sector=0, error='I/O error', rwbs='N', cmd='',
5 2022-11-01 02:39:32 +0000 error: dev=0:0, sector=-1, nr_sector=0, error='critical target error', rwbs='N', cmd='',
6 2022-11-01 22:44:22 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
7 2022-11-02 01:41:27 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
8 2022-11-02 02:02:44 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
9 2022-11-02 22:55:22 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
10 2022-11-03 01:15:27 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
11 2022-11-03 22:43:47 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',
12 2022-11-04 01:19:09 +0000 error: dev=0:2048, sector=371038120, nr_sector=144, error='unknown block error', rwbs='R', cmd='',

Do these errors mean anything? I don't know what to do about those errors.

There is a new unknown block error whenever I boot my computer.

[msoltyspl]

Can confirm that in the 0.6.8 version as well, just noticed this on one of our proxmox hosts:

Disk errors
1 2022-12-09 12:22:29 +0100 error: dev=0:2080, sector=-1, nr_sector=0, error='unknown block error', rwbs='N', cmd='', 
2 2022-12-09 12:22:29 +0100 error: dev=0:2064, sector=68197552, nr_sector=184, error='unknown block error', rwbs='R', cmd='', 
3 2022-12-09 12:22:29 +0100 error: dev=0:2080, sector=-1, nr_sector=0, error='unknown block error', rwbs='N', cmd='', 
4 2022-12-09 12:22:29 +0100 error: dev=0:2080, sector=11786494616, nr_sector=8, error='unknown block error', rwbs='R', cmd='', 
5 2022-12-09 12:22:29 +0100 error: dev=0:2080, sector=839989992, nr_sector=16, error='unknown block error', rwbs='R', cmd='', 

There is nothing in dmesg, journal and /var/log that would imply disk errors at that timestamp either.

In rasdaemon's log there's:

Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: Calling ras_mc_event_opendb()
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]:           <idle>-0     [008]     0.000003: block_rq_complete:    2022-12-09 12:22:29 +0100
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: diskerror_eventstore: 0x55e415913138
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: register inserted at db
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]:           <idle>-0     [008]     0.000003: block_rq_complete:    2022-12-09 12:22:29 +0100
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: diskerror_eventstore: 0x55e415913138
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: register inserted at db
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]:           <idle>-0     [008]     0.000003: block_rq_complete:    2022-12-09 12:22:29 +0100
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: diskerror_eventstore: 0x55e415913138
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: register inserted at db
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]:           <idle>-0     [060]     0.000003: block_rq_complete:    2022-12-09 12:22:29 +0100
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: diskerror_eventstore: 0x55e415913138
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]: rasdaemon: register inserted at db
Dec 09 12:22:33 proxmox-node-gp3 rasdaemon[7081]:           <idle>-0     [066]     0.000003: block_rq_complete:    2022-12-09 12:22:29 +0100

[mchehab]

There is an improvement on v0.7.0 that should likely address it. It requires Kernel 5.18 at least to work.

[answer9030]

I also encountered the same problem. Because the rasdaemon.service file has both ExecStart and ExecStartPost options, both processes run simultaneously, and ExecStartPost starts the block_ rq_complete event first, and then ExecStart adds a filter, resulting in no filtering out of useless information during this period. The problem no longer occurs after removing the ExecStartPost option from the rasdaemon.service file.

Why does the rasdaemon.service file have the ExecStartPost option.

[amano-kenji]

With linux 6.1.31 and rasdaemon 0.8.0, this issue seems gone. Can anyone else confirm this?