[Orca-users] Pushing for a 1.0 release

[Tankersley, Jon]

After a bit of mulling over the question....
You really have had 1.0 running since about 0.27.
But... Here are some wants

1) hostname imbedded in the file - for the off chances of misplaced files.  Something like LoclTime_hostname in the header comes to mind.  We've had occasion when this has happened and procollator-YYYY-mm-dd isn't good enough.  But we don't need the hostname but one time per file, so something like LoclTime_hostname satisfies that.

2) Considering the way we 'match' files, the idea of procallator-YYYY-mm-dd being an endall solution isn't valid.  Procallator_Vx-YYYY-mm-dd would be better, with V1 having no _Vx.  This allows what Kai suggested.

3) Maybe a 'comment' line in the data file that is ignored by most of Orca - that could satisfy #1 and #2 above.  #COMMENT Version X, Hostname XXXX, Interval 300, ....

4) incorporation of hrule and vrule.  I think the hrule is about ready, but we've never had the time to make 'color' an option instead of #CCYYMM.  Vrule is a harder nut to crack, but we've been 'hand coding' it and it is a nice feature to show solitary events on the graphs.

5) Alternate Date format for rrdtool.  I'm STILL trying to get the autoconf/etc. to work so we can 'define' alternate date.  This generates 'Mon dd' instead of 'Wk ww' on two of the graphs.

6) Possibly the generation of more arbitrary 'data retention'.  I've been asked a couple of times about 18mo and 2yr.  I haven't tried to figure out all the stuff I need to for that.  Also arbitrary data ranges - 1 week being 7 days instead of 10, etc.

7) Min/Max/Avg value retention.  The code is geared to Average.  The changes for Min/Max/Avg and combinations for that isn't that hard.  I've got a working model.  This requires some changes to the cfg file, but so far, I've got it set up so that old cfg files retain just Average.  New cfg files would be able to request min/max/avg or any combination.

8) graph statistics retention.  It would be nice to have the graph metrics at the bottom (min, max, avg, current) ALSO be saved as an ASCII file - which would be overwritten.  We've started this to get around the requrement to reread the RRD OR look at the PNG and copying down data.  The statistics are VERY nice since they do represent the displayed interval.

9) full integration of newer rrdtool.  The Holt-Winters stuff would be nice for some things.

10) a real FAQ.  There are LOTS of the same question showing up time and time again.  Not that having a really good FAQ will prevent it.  People would still have to read it.

11) A real best-practices area.  Do this, don't do this, Etc.

>-----Original Message-----
>From: orca-users-bounces+jon.tankersley=eds.com at orcaware.com 
>[mailto:orca-users-bounces+jon.tankersley=eds.com at orcaware.com]
> On Behalf Of Kai Siering
>Sent: Friday, September 23, 2005 2:36 AM
>To: Orca Users Mailing List
>Subject: Re: [Orca-users] Pushing for a 1.0 release
>
>Blair Zajac wrote:
>
>> I'd like to get a 1.0 release out in a week or so.
>> 
>> Are there any outstanding issues they would like to see fixed?
>
>Actually, yes ;)
>
>I'm graphing data of a weatherstation with Orca (version information
>below) and are rather confused as to how orca processes my .cfg.
>
>I've implanted RRDtool 1.2 into Orca snapshot 480 because of 
>the smoother graphs produced by RRDtool 1.2 -- otherwise it's 
>a standard install on an Dual-P3 running Fedora Core 4:
>
>Orca at Fri Sep 23 09:16:14 2005 running on:
>     Orca version 0.28.0 (dev 480)
>     RRDs version 1.2011
>     Perl version 5.008006
>
>I'm doing some tests on that box regarding the new procallator 
>-- in the company we have some 200 Linux boxes running with (locally
>modified) orginal procallator for several years now, therefore 
>I am not that happy with the changes introduced into latest 
>proc- allator; since the new format is not a drop-in 
>replacement, I change the pattern of the output file for new 
>procallator to "proc2-*", this way new boxes (2.6-based or 
>equipped with new
>procallator) can be processed as well as all those legaca ones.
>
>I'd therefore vote for Orca 1.0 to rename the output file for 
>new procallator as well ...
>
>But that's not the Issue I'm after at this point. Please see 
>the graphs at
>
>	http://azrael.vpn.uu.org/orca/weatherstation/
>
>and the configuration file at
>
>	http://azrael.vpn.uu.org/orca/weatherstation.cfg
>
>and sample data at
>
>	http://azrael.vpn.uu.org/orca/WSxx00-2005-09-22-002
>
>and tell my what I did wrong ;) E. g. if you check (attached) 
>http://azrael.vpn.uu.org/orca/weatherstation/zaphod/_Ti-daily.png,
>the red line is the max, the blue one current and the green 
>one the min value -- I think, from the config file, it should 
>be blue= min, green=current, red=max?
>The "current" value in the legend (which should be in "Ti" as 
>spe- cified in the .cfg)) is a flat line, it seems that 
>"Timin" is used by Orca here instead of "Ti".
>
>Question is: did I mix something up (after pondering a day 
>about it I still doubt it) or is Orca doing something the 
>wrong way around? If you need further input, please let me know ...
>
>> If you'd like to try the latest orca version, you can download it at
>> 
>> http://www.orcaware.com/orca/pub/snapshots/
>
>I haven't have time to check with latest snapshot yet, 
>hopefully next week. Regards,
>			kai
>
>-- 
>Kai 'wusel' Siering                       eMail @ home: wusel at uu.org
>Traveller on the Information Highway doing full-time administration.
>                »Do, wat Du wullt, de Lüüd snakt doch.«
>
>Asterisk @ Home: FWD #476654 / +1-360-227-5028 / +49-2222-948-000847
>