Nathan,
OK. My current work on laying hotproc on top of dynamic proc is here:
https://github.com/ubccr/pcp/tree/dynamic_hotproc
Right now the functionality should be the exact same as my initial
submission except that it works with the new dynamic proc framework.
Actually it was pretty easy to merge. This seems like a good way to
handle these types of pmdas. I have not done any of the fixes you
suggested in your initial review of hotproc, so its not ready for a
merge yet.
You can ignore all of this until I am done. I just wanted to get this
out there since I don't know where in your cgroup work you are. I think
the only conflict might be in proc_pid.c with my addition of an extra
parameter into some of the existing methods as I mentioned before.
A couple of other comments below.
On 11/10/14 1:26 AM, Nathan Scott wrote:
3. Couldn't find any precedent for handling long form help text for dynamic
pmdas. So I just generated a header file from the existing help file and
used that. Let me know if you want something different. Attached is the perl
script I wrote for that in case it is generally useful.
We might be able to do some remapping from hotproc -> proc variants, but I
don't think there's any precedence there I can point you towards - I'll can
take a look into that too if you like.
OK, this seems to be working OK as I have implemented it. Feel free to
make other suggestions if this looks difficult to maintain.
4. I left a few commented code sections that will show where hotprocs will
hook in. Will add another root node, and all other stuff ( help, etc ) will
be shared.
This has been now filled in and seems to work in the limited testing I
have done at the end of the day. I ran the existing pmda.proc QA tests
and the results are the same as before. The 2 tests you fixed with sort
and the third I am stumped at the moment as I mentioned in my other email.
One thing I noticed and am trying to track down:
In the:
static void
refresh_metrictable(pmdaMetric *source, pmdaMetric *dest, int id)
callback that I provide. The domain doesn't appear to be set in the
"source" at the time this call is made. I check, and it is "0". But
everything seems to work. I assume this gets set at some later time?
Thanks
Martins
|