I have just registered to be able to post as I am having a serious problem with a Debian box and I am out of ideas.
I am running several Debian servers as storage servers. All of them doing pretty much the same stuff and having the same load. Some with software raid, some with hardware raid some are really really close in terms of hardware and configuration.
However there is only one box that gives me one headache after another while the others are running totally fine without issues.
Here is the headache:
Code: Select all
INFO: task jbd2/sda5-8:625 blocked for more than 120 seconds.
Not tainted 5.10.0-22-amd64 #1 Debian 5.10.178-3
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message."
I have tried to set
Code: Select all
vm.dirty_background_ratio/bytes
Code: Select all
vm.dirty_ratio
Here is a trace from dmesg:
Code: Select all
task:jbd2/sda5-8 state:D stack: 0 pid: 625 ppid: 2 flags:0x00004000
Call Trace:
__schedule+0x282/0x870
schedule+0x46/0xb0
jbd2_journal_commit_transaction+0x163/0x1ad0 [jbd2]
? sched_clock+0x5/0x10
? sched_clock_cpu+0xc/0xb0
? record_times+0x15/0x90
? psi_group_change+0x41/0x220
? __switch_to_asm+0x3a/0x60
? __switch_to+0x114/0x440
? add_wait_queue_exclusive+0x70/0x70
? lock_timer_base+0x61/0x80
kjournald2+0xab/0x270 [jbd2]
? add_wait_queue_exclusive+0x70/0x70
? load_superblock.part.0+0xb0/0xb0 [jbd2]
kthread+0x11b/0x140
? __kthread_bind_mask+0x60/0x60
ret_from_fork+0x22/0x30
Any help to further debug this issue and maybe find the cause or even a solution would be appreciated.