Hello all,
I face the following issues with gpfs.py. I have followed all the
instructions in the Readme as follows:
1. The nodes do have modpython.so installed and other python modules work
as expected.
2. All the files gpfs.pyconf and gpfs.py are in there respective place.
3. I made sure sudoers exists with ganglia user and nopasswd to access
mmpmon
4. gmond -m does show that it is able to see gpfs metrics.
5. When i run gpfs.py like this :
'sudo -u ganglia python gpfs.py'
I do get printed results and also respective values every 15 minutes when
i run IO on GPFS filesystems.
However i see the following issues.
1. When i start GPFS - i see the following message.
/usr/sbin/gmond[4190]: Unable to find any metric information for
'gpfs_(.+)'. Possible that a module has not been loaded.
2. Majority of the systems on ganglia web do not show that gpfs metrics
exists, though the nodes themselves show that gpfs_* in one of there
metrics in gmond -m.
3. Some nodes show that GPFS metrics exists but do not show any values .
They are always -nan.
I have tried looking at the python module to see if there is anything
explicit, but i dont see anything wrong.
Also i tried using tcpdump on the respective ganglia port on few of the
client nodes to see if they are sending any gpfs metrics. To my surprise i
dont see any gpfs named metrics in the tcpdump.
Could anyone help me out with this issue?
Thank you,
Lohit
------------------------------------------------------------------------------
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
_______________________________________________
Ganglia-developers mailing list
Ganglia-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-developers
