Lukasz Janyst wrote:
> 2011/10/7 Paul T. Keener <[log in to unmask]>:
>> I will look at this, but if it really is a network problem, wouldn't I see
>> it regardless of which xrootd namespace I was talking to?
>
> Probably, but on the other hand xrootd does not have control over
> hardware interrups (unless you meant syscalls :)
Right. What I am reporting is the "int" line from dstat. I don't
expect that it is the problem; I just wanted to give a more complete
list of metrics that change significantly under load.
I am more concerned about the wait states, reported under the "wai" line
in dstat. This seems to have the best correlation with the performance
problems.
Is there a way to get useful performance information out of the xrootd
process?
-paul
|