Custom Query (101 matches)

Hi,

I run the last stable version of jobmonarch on an IA64 server with redhat Linux AS 4 and torque 2.1.8. I install yesterday jobmarch and it works fine until a long job (which was running about 12hours) finished this night. After that, the process jobmond.py consumes about 20% of "cpu system time". With the ps command , I see that jobmond.py run very frequently pbs_iff.

root 1772 4.8 0.1 67280 8096 ? S Apr25 60:21 /usr/bin/python -v /usr/local/sbin/jobmond.py -c /etc/jobmond.conf root 4146 0.0 0.0 752 320 ? R 09:58 0:00 /usr/local/torque.2.1.8.fPIC/sbin/pbs_iff localhost 15001 4

How can I correct this problem ?

Thanks,

Mickael,

I attach the content of my file /etc/jobmond.conf

Nick Galbreath created a 100% python native gmetric module:

​http://code.google.com/p/embeddedgmetric/wiki/GmetricPython

It does not require a C module, swig, nothing.

Hi Ramon,

correct, the python code is one file and "does it all" without any dependencies.  No C, No Swig, No non-standard modules.

Take a peek
http://embeddedgmetric.googlecode.com/svn/trunk/python/gmetric.py
that's really all it is.



The main difference between my code and the official gmetric is that my code only will send metrics to  ONE mutlicast or  ONE udp port.  official gmetric reads in the gmond.conf file and can send metrics to multiple ports with multiple protocols at once.  But I think 99% of the ganglia installs just sends data to 1 port.


Feel free to package it up however you like.  Its the MIT license, which just says "Don't sue me if it blows up, and preserve the copyright notice"   If you need a different license let me know ( but MIT has is least restrictive).

have fun!

--nickg

Need to implement this ASAP, as I think this will speed up jobmond quite a bit and make it more flexible.

After going through the instructions, I attempted to execute
jobmond.py. When I did that, I received the following error message:
cluster1:/usr/local/sbin # /usr/local/sbin/jobmond.py -c /etc/jobmond.conf
Traceback (most recent call last):
 File "/usr/local/sbin/jobmond.py", line 814, in ?
   main()
 File "/usr/local/sbin/jobmond.py", line 807, in main
   gather.daemon()
UnboundLocalError: local variable 'gather' referenced before assignment

An examination of the code reveals that the SGE data gathering code
was commented out on line 792. Uncommenting it had the following
effect:
cluster1:/usr/local/sbin # /usr/local/sbin/jobmond.py -c /etc/jobmond.conf
 File "/usr/local/sbin/jobmond.py", line 797
   debug_msg( 0, "fatal error: BATCH_API set to 'sge' but python
module 'sge_drmaa' is not installed' )

                               ^
SyntaxError: EOL while scanning single-quoted string

Commenting out everything but "gather = SgeDataGatherer()" gave me the
following error:
cluster1:/usr/local/sbin # /usr/local/sbin/jobmond.py -c /etc/jobmond.conf
Traceback (most recent call last):
 File "/usr/local/sbin/jobmond.py", line 814, in ?
   main()
 File "/usr/local/sbin/jobmond.py", line 800, in main
   gather = SgeDataGatherer()
 File "/usr/local/sbin/jobmond.py", line 419, in __init__
   self.initSgeJobInfo()
 File "/usr/local/sbin/jobmond.py", line 426, in initSgeJobInfo
   self.qstatparser = SgeQstatXMLParser( SGE_QSTAT_XML_FILE )
NameError: global name 'SGE_QSTAT_XML_FILE' is not defined

At this point, I decided to search my systems for references to drmaa.
I saw several references to C++ example and header files related to
it. Is the sge_drmaa module supposed to be provided by Job Monarch or
Sun Grid Engine?