5.3-RELEASE: WARNING - WRITE_DMA interrupt timout - what does it mean?

[Søren Schmidt]

Robert Watson wrote:

>>It means that the disk has processed the write request (interrupt seen),
>>but that the system (the bio_taskqueue) hasn't been able to get the
>>result returned to the kernel. 
>>
>>Your disk is not involved in this problem since it has done its part,
>>but the rest of the system is either busy with something else, or there
>>are bugs lurking that prohibits the bio_taskqueue from running. 
>>
>>Either way its a WARNING not a FAILURE :) 
> 
> 
> I'm still a bit skeptical that the task queue is at fault -- I run my
> notebook with continuous measurement of the latency to schedule tasks,
> generating a warning for any latency > .5 seconds, and the only time I
> ever see that sort of latency is during the boot process when ACPI has
> scheduled a task to run, but the task queue thread has not yet been
> allowed to run:

Right, the timeout is 5 secs. I havn't looked into how the taskqueues 
are handled recently, but in case of ATA read/writes it is the 
bio_taskqueue handled by geom thats in use not the catchall ones, does 
your timing cover that as well?

There are several explanations for what happens:

1. the bio_taskqueue is not pushing requests through.
2. the disks takes long to respond and uses almost all of the 5 secs
3. timeouts are not working and fireing at random.

I cannot reproduce the symptoms on any of my HW no matter how hard I hit 
it, and I dont really belive in items 2 and 3 above, however I've been 
proven wrong before :)

-- 

-Søren