Difference between revisions of "INFO: task blocked for more than 120 seconds."

From Helpful
Jump to: navigation, search
m
m
Line 6: Line 6:
  
  
'''This message is not an error'''.
+
'''This message is not an error''', but you probably do care about what caused it.
 +
 
 +
It is an indication that a program has had to [[iowait]] for a very long time (or sometimes is the ''culprit'' of heavy [[iowait]]).
 +
 
 +
The stack trace can help diagnose what it was doing. {{comment|(which is not so informative of the ''reason'' - it's common that the real IO load issue comes from another process)}}
  
It is an indication that a program has had to wait for a very long time, and what it was doing. {{comment|(which is not so informative of the ''reason'' - it's common that the real IO load issue comes from another process)}}
 
  
  
Line 14: Line 17:
 
This is a kernel thread that detects tasks that stays in the [[D state]] for a while (which typically meaning it is waiting for IO).
 
This is a kernel thread that detects tasks that stays in the [[D state]] for a while (which typically meaning it is waiting for IO).
  
It complains when it sees a process has been waiting on IO so long that the whole process has not been scheduled for any CPU-time for 120 seconds (default).
+
It complains when it sees a process has been waiting on IO so long that the whole process has not been scheduled for any CPU-time for 120 seconds (default), which is an eternity unless IO is an issue.
 
+
  
 
<!--
 
<!--
Line 27: Line 29:
 
* if it happens constantly your IO system is slower than your IO use
 
* if it happens constantly your IO system is slower than your IO use
  
* most likely to happen ''to'' a process that was [[ionice]]d into the idle class. Which means it's working, idle-class is meant as an extreme politeness thing. It just indicates something else is doing a bunch of IO right now (for at least 120 seconds)
+
* most likely to happen ''to'' a process that was [[ionice]]d into the idle class. Which means it's working as intended, becauseidle-class is meant as an extreme politeness thing. It just indicates something else is doing a bunch of IO right now (for at least 120 seconds)
 
: e.g. [http://en.wikipedia.org/wiki/Locate_%28Unix%29 updatedb] (may be victim if it were ioniced, cause if not)
 
: e.g. [http://en.wikipedia.org/wiki/Locate_%28Unix%29 updatedb] (may be victim if it were ioniced, cause if not)
  

Revision as of 19:47, 8 February 2021

Under heavy IO load on servers you may see something like:

INFO: task nfsd:2252 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.

...probably followed by a call trace that mentions your filesystem, and probably io_schedule and sync_buffer.


This message is not an error, but you probably do care about what caused it.

It is an indication that a program has had to iowait for a very long time (or sometimes is the culprit of heavy iowait).

The stack trace can help diagnose what it was doing. (which is not so informative of the reason - it's common that the real IO load issue comes from another process)


The code behind this sits in hung_task.c and was added somewhere around 2.6.30. This is a kernel thread that detects tasks that stays in the D state for a while (which typically meaning it is waiting for IO).

It complains when it sees a process has been waiting on IO so long that the whole process has not been scheduled for any CPU-time for 120 seconds (default), which is an eternity unless IO is an issue.



Notes:

  • if it happens constantly your IO system is slower than your IO use
  • most likely to happen to a process that was ioniced into the idle class. Which means it's working as intended, becauseidle-class is meant as an extreme politeness thing. It just indicates something else is doing a bunch of IO right now (for at least 120 seconds)
e.g. updatedb (may be victim if it were ioniced, cause if not)
  • if it happens only nightly, look at your cron jobs
  • a trashing system can cause this, and then it's purely a side effect of one program using too much RAM
  • being blocked by a desktop-class drive with bad sectors (because they retry for a long while)


  • NFS seems to be a common culprit, probably because it's good at filling the writeback cache, something which implies blocking while writeback happens - which is likely to block various things related to the same filesystem. (verify)
  • if it happens on a fileserver, you may want to consider spreading to more fileservers, or using a parallel filesystem


if your load is fairly sequential, you may get some relief from using the noop io scheduler (instead of cfq) though note that that disables ionice)
if your load is relatively random, upping the queue depth may help