Hellebø;

Try this.

- Create a file called ghosts (or anything you want) with all the
hostnames of systems running gmond.
- run the following command.

<pre>
for HOST in `cat ghosts `;do echo -n "Host $HOST:   ";telnet $HOST 8649\
2>/dev/null|grep "GANGLIA_XML VERSION="|awk '{print $2;}';done
</pre>

This will print a list like this:

Host host1:   VERSION="3.1.2"
Host host2:   VERSION="3.1.2"
Host host3:   VERSION="3.1.2"
...

Then you can see all versions at once.  I had the same issue after an
upgrade to 3.1.x which turned out to be a host that had been missed
during the massive (100+ nodes) upgrade and was still running 3.0.9.  If
you have different ports for some hosts you'll need to extrapolate a bit.

I hope this helps.

Brad


Bernard Li wrote:
> Hi Hellebø:
> 
> Are your gmonds also running version 3.1.2?
> 
> Thanks,
> 
> Bernard
> 
> 2009/10/14 Hellebø Knut <[email protected]>:
>> Regards,
>>
>>
>>
>> Since upgrading a cluster to Ganglia v3.1.2, I am getting sporadic XML
>> errors preventing gmetad from functioning properly. Every time I get a
>> XML_ParseBuffer error, the WEB page updating stops working and we have to
>> restart gmetad  to get things properly up and running again. There is no
>> consistency regarding the error, ie the issue seems to trigger an error on
>> different locations and different points in time. The errors appear even
>> when system load is low. I have tried to adjust the memory usage parameters
>> in php.ini and extracted XML data by doing a telnet to the
>> tcp_accept_channel port followed by an xmllint check. No luck. Here's a
>> snippet of the errors in the /var/log/messages file:
>>
>>
>>
>> Oct  1 10:25:58 ganglia-server /usr/sbin/gmetad[6802]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 50591: not well-formed (invalid
>> token)
>>
>> Oct  5 12:56:04 ganglia-server /usr/sbin/gmetad[6354]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 62476: not well-formed (invalid
>> token)
>>
>> Oct  6 23:23:37 ganglia-server /usr/sbin/gmetad[14887]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 3119: not well-formed (invalid
>> token)
>>
>> Oct  8 02:27:10 ganglia-server /usr/sbin/gmetad[6440]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 39597: not well-formed (invalid
>> token)
>>
>> Oct  8 07:28:53 ganglia-server /usr/sbin/gmetad[10433]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 59909: not well-formed (invalid
>> token)
>>
>> Oct  8 12:04:33 ganglia-server /usr/sbin/gmetad[28723]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 11002: not well-formed (invalid
>> token)
>>
>> Oct  8 15:11:08 ganglia-server /usr/sbin/gmetad[26030]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 34877: not well-formed (invalid
>> token)
>>
>> Oct  9 05:23:11 ganglia-server /usr/sbin/gmetad[2315]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 50100: not well-formed (invalid
>> token)
>>
>> Oct 10 23:07:47 ganglia-server /usr/sbin/gmetad[2397]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 1195: not well-formed (invalid
>> token)
>>
>> Oct 11 18:22:02 ganglia-server /usr/sbin/gmetad[28620]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 7919: not well-formed (invalid
>> token)
>>
>> Oct 12 15:20:01 ganglia-server /usr/sbin/gmetad[14779]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 74246: not well-formed (invalid
>> token)
>>
>> Oct 12 19:37:05 ganglia-server /usr/sbin/gmetad[18635]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 32447: not well-formed (invalid
>> token)
>>
>> Oct 13 07:28:39 ganglia-server /usr/sbin/gmetad[25855]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 296: not well-formed (invalid
>> token)
>>
>> Oct 13 10:10:54 ganglia-server /usr/sbin/gmetad[4560]: Process XML
>> (MyCluster): XML_ParseBuffer() error at line 35613: not well-formed (invalid
>> token)
>>
>>
>>
>>
>>
>> I have also tried to run gmetad in debug mode and redirect the errormessages
>> to the debug log file, but I only get informative messages and no elaborate
>> errormessages before and after the point in time when the XML error appear.
>>
>> All gmonds are running on RHEL4 and are v3.1.2.
>>
>> Suggestions ?
>>
>> Relevant server details: RHEL 4 U5 ES, ganglia 3.1.2 , libconfuse v.2.6, apr
>> v1.3.3.
>>
>> ------------------------------------------------------------------------------
>> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
>> is the only developer event you need to attend this year. Jumpstart your
>> developing skills, take BlackBerry mobile applications to market and stay
>> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
>> http://p.sf.net/sfu/devconference
>> _______________________________________________
>> Ganglia-general mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/ganglia-general
>>
>>
> 
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay 
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> _______________________________________________
> Ganglia-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/ganglia-general



-- 
Brad Hudson
SA Team Lead
The Pythian Group
Desk: 613-565-8696 x202
IM: pythianhudson

<<attachment: hudson.vcf>>

------------------------------------------------------------------------------
Come build with us! The BlackBerry(R) Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay 
ahead of the curve. Join us from November 9 - 12, 2009. Register now!
http://p.sf.net/sfu/devconference
_______________________________________________
Ganglia-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ganglia-general

Reply via email to