[torqueusers] momctl error

David Jackson jacksond at clusterresources.com
Thu Feb 24 17:34:16 MST 2005


Clifton,

  Can you run the command 'momctl -d 4' when you are on node r11n02?

Dave

On Thu, 2005-02-24 at 17:47 -0600, Clifton Kirby wrote:
> Torque seems to be running fine on my Apple cluster running OSX but I get
> some interesting results from momctl as follows,
> 
> # ./momctl -d 4 -h r11n02
> simpleget: Premature end of message
> ERROR:    query[0] 'diag4' failed on r11n02 (errno: 0:5)
> startcom: diswsi error Protocol failure in commit
> 
> But pbsnodes reports,
> 
> # pbsnodes -a r11n02
> r11n02
>      state = free
>      np = 2
>      ntype = cluster
>      status = arch=darwin,uname=Darwin r11n02.mach5.roc 7.6.0 Darwin Kernel
> Version 7.6.0: Fri Oct 29 15:50:52 PDT 2004; semeria:BUILD/obj/RELEASE_PPC
> Power Macintosh,sessions=? 15205,nsessions=? 15205,nusers=?
> 15205,idletime=522399,totmem=? 15201,availmem=?
> 15201,physmem=3670016kb,ncpus=2,loadave=0.00,netload=?
> 15201,rectime=1109266449
> 
> Is this something I should be concerned with?
> 
> Now my server is on a different subnet as the client so if momctl uses UDP
> to communicate I could understand the error but is it causing other
> problems?
> 
> 
> 



More information about the torqueusers mailing list