Re: NOHZ: local_softirq_pending 100 - is there something to worry about? [message #50868 is a reply to message #50867] |
Mon, 18 November 2013 05:29   |
|
The problem is that when the machine crashes, there's nothing in the logs, no indication. It just happens, and I am left hanging with no data to hand to developers that might aid them in debugging and understanding the cause. I don't know how to reproduce it, it happens randomly.
All I can see is those small "warnings" in the logs, such as the NOHZ error message, CPU locking messages, etc.
For example:
[42071.390012] hrtimer: interrupt took 13881 ns
Or this one:
[30754.200039] NOHZ: local_softirq_pending 100
Or this one which happened a few days ago and complete froze the machine:
[249348.095995] BUG: soft lockup - CPU#4 stuck for 67s! [flush-8:0:866]
(for all CPU's, not just CPU#4).
|
|
|