I thought that audit logging had a significant performance penalty.
Besides that, I'm open to doing some recording for others (and myself)
to benefit.
jason
On 08/22/2011 11:42 AM, Matt W. Benjamin wrote:
Hi,
Data from callback state dump and audit log would be of value to me. We would
like to develop empirical information about callback breaks, callback
re-registration, and related.
Matt
----- "Andrew Deason"<[email protected]> wrote:
On Sun, 21 Aug 2011 11:17:26 -0400
Jason Edgecombe<[email protected]> wrote:
I was wondering. What data from production environments would be
helpful
for the OpenAFS developers?
I'm pretty sure at least Tom would be interested in the xstat data
(xstat_cm_test, xstat_fs_test). You can run them pretty often with
almost no impact (as in, on the order of once every few seconds), and
the xstat data will actually show how little impact the calls make.
Just storing the output of the commands is pretty inefficient, though.
I
believe Tom was going to release some tooling soon for (among other
things) recording and storing that data over time.
--
Andrew Deason
[email protected]
_______________________________________________
OpenAFS-devel mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-devel
_______________________________________________
OpenAFS-devel mailing list
[email protected]
https://lists.openafs.org/mailman/listinfo/openafs-devel
