[stats-dev] FW: OGR verification

Chris Hodson nerf at slacker.com
Fri May 2 13:20:22 EDT 2003


FYI

I'm not sure about the bug he's talking about, but I'll look into it.

-----Original Message-----
From: Stanley Appel [mailto:s.appel at bigfoot.com] 
Sent: Thursday, May 01, 2003 2:14 PM
To: 'Chris Hodson'
Subject: RE: OGR verification


Nerf,

If found a small problem with the verification code. When there is no stub
in pass2 returned it goes wrong. There will be no insert because it's a null
value.

And I made some scripts for you. I saw some graphs on d.net done with
rrdtool. I also made some for my own. Anyway here are some scripts you can
you to create and use to show the completed %. 
They are really simple scripts.

create.sh : will create 2 rrd files. It will use 1st of May 2003 a fixed
start date. No arguments.

graph.sh : will create 2 .gif files also start date 1st of May 2003. If you
need a .png file just add  : --imgformat PNG

feed.pl : perl script which will retrieve the current pass1 and pass2 values
from the postgresql database. It hase 4 arguments:
	project id   eg 24
	day		 eg 01
	month		 eg 05
	year	  	 eg 2003
If you need to change the database name, user and password. Its :
	DBI->connect("DBI:Pg:dbname=DATBASE_NAME", 
		"DATABASE_USER", 
		"DATABASE_PASSWORD")

update.sh : will add the values from feed.pl to the rrd datafiles.
		it uses the current date, maybe this has to be changed.

Just run create.sh once. Then update.sh and graph.sh after the daily
verification update.


Have fun with it and let me know.

Stappel.



-----Original Message-----
From: Chris Hodson [mailto:nerf at slacker.com] 
Sent: Friday, April 11, 2003 16:32
To: Stanley Appel
Subject: Re: OGR verification

Always good to hear that someone is reading what I write, both .plans and my
code.

It's great to hear that you able to get the code to run, I tried really hard
to make it as generic as possible.  I would be very interested in any
changes you had to make to get it to run.  And, we rarely turn down
enhancements :) I would also be interested in how useful the "dummy" sql I
wrote is. That is the code that makes up for the fact that you don't have
access to the real data (like making OGR_idlookup).

As for the help I mentioned, get the latest stats.sql.  It creates a table
called ogr_complete that has much better formatting, not to mention a
history.

As for your specific questions:

> I must say the code looks sound to me. I'll do some more testing to 
> see if
I
> can find something wrong.

I certainly hope you find a bug that increases uour completion rate, but I'm
quite happy to hear that no bugs have been found.

> A few questions.
> a) why do you say in filter.pl that everything from a client <= 8012 
> is wrong? is this data not good enough for a pass 1?

Windows clients that are version <8012 produced questionable results. As for
the specifics, the developers would be better to ask.  I was just told what
I told you :)  You might also want to talk to Joel as well, as he was big in
getting this thing going and wrote the filter program originally.

> b) also you define 3 variables in filter.pl as filenames, but never 
> use them.

Yes, this goes back to the original way filter.pl was written, in which it
wrote out to three separate files.  We decided to make ogr24 and ogr25 be
all together, so this was changed, and then it was changed into being a
pipe, so it would be similar to the filter program currently used to rc5

> c) I saw that there was a reissue of OGR-24 stubs last night. is this 
> the hole stub space or just the missing stubs?

I've been manually sending a list of completed stubs to Bovine about once a
week.  I'm pretty sure that means we're only sending out the stubs that have
not completed pass2
> 
> and
> 	Do you still need help on the code?

Always!  As I mentioned, the specific issue I mentioned in my .plan is done,
but there's still work to be done, enhancements to be made.

> 
> regards,
> 
> Stanley [aka Stappel]
> 

-Nerf
-------------- next part --------------
A non-text attachment was scrubbed...
Name: create.sh
Type: application/octet-stream
Size: 330 bytes
Desc: not available
Url : http://lists.distributed.net/pipermail/stats-dev/attachments/20030502/137a36f0/create.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: feed.pl
Type: application/octet-stream
Size: 915 bytes
Desc: not available
Url : http://lists.distributed.net/pipermail/stats-dev/attachments/20030502/137a36f0/feed.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: graph.sh
Type: application/octet-stream
Size: 560 bytes
Desc: not available
Url : http://lists.distributed.net/pipermail/stats-dev/attachments/20030502/137a36f0/graph.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: update.sh
Type: application/octet-stream
Size: 173 bytes
Desc: not available
Url : http://lists.distributed.net/pipermail/stats-dev/attachments/20030502/137a36f0/update.obj


More information about the stats-dev mailing list