Re: [Ganglia-developers] Problems with cpu_speed_func()

[matt m. <ma...@cs...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

lars-

can you give a little more details about how ganglia is failing?  the left 
thing to do is for us to try and fix the cpu speed function to not crash.  
what cpu(s) are you running?
what os?
what does you /proc/cpuinfo look like (i guess you are running linux)?
do you have any ideas how the current function is failing to parse it?

sorry there are more questions here than answers.  
- -matt

Today, Lars Kellogg-Stedman wrote forth saying...

> From: Lars Kellogg-Stedman <la...@la...>
> To: "gan...@li..."
>     <gan...@li...>
> Date: Tue, 10 Jun 2003 10:11:33 -0400 (EDT)
> Subject: [Ganglia-developers] Problems with cpu_speed_func()
> 
> [ This was originally posted to ganglia-general, but I suspect the
> developers list would be a more appropriate forum.]
> 
> I'm using a small user-mode-linux cluster to test out some clustering 
> tools.  Ganglia looks like it may solve some issues I've been ruminating 
> about...but under UML, gmond segfaults in cpu_speed_func().
> 
> The segfault occurs because when the code is built, __i386__ is defined, 
> but at runtime /proc/cpuinfo doesn't contain the expected information.
> 
> Is this important enough to warrant a solution in the code?  One possible 
> solution would be to allow the user to specify a fixed value in 
> gmond.conf, so that cpu_speed_func looks like:
> 
>   if (user provided a value)
>     return user_provided_vlaue;
>   else
>     do normal cpu speed detection
> 
> I don't know if the value of cpu_speed_func() is used for anything by 
> ganglia, but another possible solution would be to simply return 0 if any 
> of the various strstr()/strchr()/etc. functions fail -- this is what I've
> done locally.
> 
> -- Lars
> 
> 
> 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE+5j/HVmIXr0CKtmERAr2QAKCMrYNVHrNPRZjcpeNXWln6EvZsnQCfYdHR
JY7EgI8EsykQdAo7uhCmZBU=
=hK8g
-----END PGP SIGNATURE-----