[Orca-dev] Orcallator.se timestamp patch

[Blair Zajac]

Dmitry Berezin wrote:

> Hi Blair,
> 
> 
>>What other changes did you want to make to Orca?
> 
> 
> There are a few things that I'd like to change.
> 
> 1. I have submitted an "orca friendly" SE class based on Adrian's workload
> class and a patch to integrate it with orcallator. Any interest in that?
> 
> 2. There seems to be a bug in SE with first_proc/next_proc. I have posted my
> findings and tests on this earlier, and a suggested alternative. This might
> be irrelevant though, if No.1 (above) is accepted and process counting
> functionality is moved into workload code.
> 
> 3. This is probably the number one priority - disk statistics. I have
> suggested earlier either to remove RAWDISK code completely or to rewrite
> some of it to make it more "dynamic". I don't see much agreement in removing
> it, so perhaps I could try improving it.
> 
> 4. Change orcallator.se to be able to monitor multiple web log files. 

Dmitry,

I'm interested all of this.

How about we get you set up with Subversion access and then you can run with 
these as your own speed without having to go through me.

Just to let you know though, I can be a pretty picky code reviewer, and will 
review and modify commits after they go in :)

As an FYI, right now I'm looking to get people Orca Subversion commit access so 
that the project can move forward quickly.  To do this, I'm being a little more 
relaxed with giving out Subversion access than I will be in the future, where 
I'm going to raise the bar to the model they use on the Subversion team (where 
people submit multiple patches over time and they get recommended for commit 
access).

Regards,
Blair

-- 
Blair Zajac <blair at orcaware.com>
Plots of your system's performance - http://www.orcaware.com/orca/