[rancid] Re: "svn warning - <router> is already under version control" message while running the rancid-run -r <router>

[Atle Østbø]

Hi John

>>I think you should be asking yourself why this so slow.  Is your system's
>>directly cache too small?  Disk performance problems?  Is there contention
>>for locks on the svn database?

I am running an DELL server with the following configuration:

	CPU's:
	------
	System Specifics.....:   
	 GenuineIntel  4
	               Intel(R) Pentium(R) D CPU 3.00GHz
	 GenuineIntel  4
	               Intel(R) Pentium(R) D CPU 3.00GHz,  3000
	 3000 MHz
	                        1024348 kB RAM
	                        2097144 kB swap space
	Operating System.....:  Linux version 2.6.9-42.0.3.ELsmp (brewbuilder at hs20-bc2-2.build.redhat.com) (gcc version 3.4.6 20060404 (Red Hat 3.4.6-3)) #1 SMP Mon Sep 25 	17:24:31 EDT 2006


	Disk Configuration:
	------------------
	Filesystem           1K-blocks      Used Available Use% Mounted on
	/dev/mapper/VolGroup00-LogVol00
	                       4128448    511596   3407140  14% /
	/dev/sda3               194449     25871    158538  15% /boot
	none                    512172         0    512172   0% /dev/shm
	/dev/mapper/VolGroup00-LogVol05
	                       4128448     51048   3867688   2% /home
	/dev/mapper/VolGroup00-LogVol03
	                       4128448     50924   3867812   2% /tmp
	/dev/mapper/VolGroup00-LogVol02
	                       8256952   5071532   2765992  65% /usr
	/dev/mapper/VolGroup00-LogVol04
	                       4128448    239204   3679532   7% /var

You write something about the system's cache size, could you please come with more info on that  (- how to checkout + and fix) ?


>> Is there contention for locks on the svn database?

What did you mean ? - I don't know. 


>>You might also ask yourself if 5000 devices in a single group is a good
>>design.  Seems like that would produce rather overwhelming diffs.

Yes, I need the all config backup's in the same group. It's only 5000 files - I have read about subversion server with more that 30000 files, so I don't se any performance problem with that.



------------------------------------------------

Med vennlig hilsen/Best regards


Atle Østbø
Network Services

 

> -----Original Message-----
> From: john heasley [mailto:heas at shrubbery.net] 
> Sent: 9. desember 2006 03:32
> To: Atle Østbø
> Cc: john heasley; rancid-discuss at shrubbery.net
> Subject: Re: "svn warning - <router> is already under version 
> control" message while running the rancid-run -r <router>
> 
> Thu, Dec 07, 2006 at 11:54:10AM +0100, Atle ?stb?:
> > Hi
> >  
> > I have try to use rancid to take care of the configuration 
> changes for our routers.
> >  
> > I export an node list from our HP OpenView NNM and build an 
> list of over 5000 cisco routers.
> >  
> > The rancid starting: Tue Dec 5 23:00:01 CET 2006 - and was 
> finish: ending: Thu Dec 7 05:22:23 CET 2006
> > 
> > It's take a long time to build the svn repository - it use 
> nearly 24 hours to do that. I don't know why this take so longe time. 
> 
> I think you should be asking yourself why this so slow.  Is 
> your system's
> directly cache too small?  Disk performance problems?  Is 
> there contention
> for locks on the svn database?
> 
> You might also ask yourself if 5000 devices in a single group 
> is a good
> design.  Seems like that would produce rather overwhelming diffs.
> 
> > Is there any way to perform the rancid to get config from 
> one router without check in all of them, only get the new 
> config from the -r <router> ?
> 
> Not with the current code.
>