[scyld-users] beostatus showing all nodes are down

Felipe Albornoz Felipe.Albornoz at pbmr.co.za
Thu Nov 11 09:37:42 PST 2004


A colleague of mine had setup a Beowulf cluster with 1 master and 7
computational nodes by using Scyld Beowulf release 29cz (29cz-2_Scyld
200404191804)
 
As a user, I would like to check the status of the processes running in
the different nodes. I am able to use "bpstat" and get information about
the status of each node. I also use "bpsh -ad ps ax" to check which are
the processes running on each node... But when I try to use "beostatus"
(the GUI interface for monitoring status), it starts but shows me that
all nodes are down. Before opening the x-window showing me that status,
I got the following messages in my terminal:
 
[root at BeowulfCluster mcnp]# beostatus
Traceback (most recent call last):
  File "/usr/share/beostatus/modules/beostatus_gui.py", line 116, in
updateStats
    nodeStatus, maxCpus = getStatus()
  File "/usr/bin/beostatus", line 145, in getStatus
    nStatus = getNodeStatus(nodeNum)
  File "/usr/bin/beostatus", line 90, in getNodeStatus
    status.ncpus = beostat.get_cpu_count(nodeNum)
beostat module error: Library function error.
Traceback (most recent call last):
  File "/usr/share/beostatus/modules/beostatus_gui.py", line 116, in
updateStats
    nodeStatus, maxCpus = getStatus()
  File "/usr/bin/beostatus", line 145, in getStatus
    nStatus = getNodeStatus(nodeNum)
  File "/usr/bin/beostatus", line 90, in getNodeStatus
    status.ncpus = beostat.get_cpu_count(nodeNum)
beostat module error: Library function error.

Could you give me some advice about what is happening, and how to solve
it? 
 
Thanks!
 
Felipe
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.scyld.com/pipermail/scyld-users/attachments/20041111/e644bf72/attachment.html


More information about the Scyld-users mailing list