| 1 |
On Thursday 26 November 2015 18:15:26 James wrote: |
| 2 |
|
| 3 |
> And this is an different approach suggesting for Peter's Chrony issues |
| 4 |
> so it might warrant a new thread anyway. |
| 5 |
> |
| 6 |
> <snip> |
| 7 |
> The installation-default file causes the start-stop-daemon to catch an |
| 8 |
> unexpected interrupt and report an error, even though the chronyd |
| 9 |
> process |
| 10 |
> continues to run. |
| 11 |
> |
| 12 |
> Any time I run 'strace -ff -o/tmp/chronyd.strace /etc/init.d/chronyd |
| 13 |
> start' the init process runs normally and I'm left with scores of trace |
| 14 |
> files, none of which help because the stray interrupt wasn't detected |
| 15 |
> <end/snip> |
| 16 |
> |
| 17 |
> So, a while back the excellent tools of trace-cmd and kernelshark (part of |
| 18 |
> trace-cmd) were added to the portage tree. |
| 19 |
> |
| 20 |
> Here is a little 'diddy' that might get you going [1]. Also, there is much |
| 21 |
> posted on the use of ftrace, trace-cmd and kernelshark. These very |
| 22 |
> powerful tools take a while to master. Often, I try to recreate deep |
| 23 |
> performance problems on minimal system, just to reduce the amount of data |
| 24 |
> one has to parse. |
| 25 |
--->8 |
| 26 |
> [1] |
| 27 |
> http://fabbritech.blogspot.com/2013/11/using-ftrace-and-kernelshark-part-2 |
| 28 |
> .html |
| 29 |
|
| 30 |
Thanks anyway, but it turns out to have been a spurious parameter in the |
| 31 |
init script (see my other message). |
| 32 |
|
| 33 |
-- |
| 34 |
Rgds |
| 35 |
Peter |
Archives