[Linux-ha-dev] LRM logging

[Dejan Muhamedagic]

Hi,

What worries me is that these are obviously logging outbursts
which may introduce another performance penalty.

Not only the logging but also calling external programs (such as
ps) might sometimes aggravate situations when we are short on
resources (I mean computing resources).

Cheers,

Dejan

On Wed, Aug 09, 2006 at 04:02:32PM +0200, Andrew Beekhof wrote:
> Its time to bring this up again... can someone please do some work on  
> the LRM logging?
> 
> There is verbose and then there is:
> 
> Aug 09 15:53:30 Running test SimulStop (c001n06)        [7]
> Aug 09 15:56:53 BadNews: Aug  9 15:54:19 c001n03 lrmd: [28984]:  
> ERROR: cl_log: 293 messages were dropped
> Aug 09 15:56:53 BadNews: Aug  9 15:54:19 c001n02 lrmd: [31993]:  
> ERROR: cl_log: 141 messages were dropped
> Aug 09 15:56:55 BadNews: Aug  9 15:54:53 c001n07 lrmd: [11102]:  
> ERROR: cl_log: 108 messages were dropped
> Aug 09 15:56:55 BadNews: Aug  9 15:54:53 c001n06 lrmd: [11166]:  
> ERROR: cl_log: 70 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:54 c001n07 lrmd: [11102]:  
> ERROR: cl_log: 106 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:54 c001n04 lrmd: [12569]:  
> ERROR: cl_log: 18 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:54 c001n07 lrmd: [11102]:  
> ERROR: cl_log: 153 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:54 c001n04 lrmd: [12569]:  
> ERROR: cl_log: 124 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:54 c001n04 lrmd: [12569]:  
> ERROR: cl_log: 493 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:54 c001n06 lrmd: [11166]:  
> ERROR: cl_log: 849 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:55 c001n04 lrmd: [12569]:  
> ERROR: cl_log: 104 messages were dropped
> Aug 09 15:56:56 BadNews: Aug  9 15:54:55 c001n04 lrmd: [12569]:  
> ERROR: cl_log: 156 messages were dropped
> Aug 09 15:56:57 BadNews: Aug  9 15:54:55 c001n06 lrmd: [11166]:  
> ERROR: cl_log: 22 messages were dropped
> Aug 09 15:56:57 BadNews: Aug  9 15:54:55 c001n06 lrmd: [11166]:  
> ERROR: cl_log: 5 messages were dropped
> Aug 09 15:56:57 BadNews: Aug  9 15:54:55 c001n06 lrmd: [11166]:  
> ERROR: cl_log: 245 messages were dropped
> Aug 09 15:56:57 BadNews: Aug  9 15:54:56 c001n04 lrmd: [12569]:  
> ERROR: cl_log: 22 messages were dropped
> Aug 09 15:56:57 BadNews: Aug  9 15:54:56 c001n04 lrmd: [12569]:  
> ERROR: cl_log: 208 messages were dropped
> Aug 09 15:56:57 BadNews: Aug  9 15:54:56 c001n06 lrmd: [11166]:  
> ERROR: cl_log: 16 messages were dropped
> Aug 09 15:57:57 Running test SimulStop (c001n06)        [8]
> 
> And thats only with 6 CTS nodes.
> 
> This is a real problem.  The CRM is horribly verbose and still  
> doesn't deluge the logging queue the way the LRM does.
> 
> --
> Andrew Beekhof
> 
> "Would the last person to leave please turn out the enlightenment?" -  
> TISM
> 

> _______________________________________________________
> Linux-HA-Dev: Linux-HA-Dev at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
> Home Page: http://linux-ha.org/