Robert,
Do you know what revision you were running when it still worked?
Probably something broke or disappeared or so when you updated a few weeks ago and didnt get fixed until now...
Tom
----- Reply message -----
From: "Robert Williams" <Robert(a)CustodianDC.com>
To: "'Observium Network Observation System'" <observium(a)observium.org>
Subject: [Observium] C6K Fabric stats not being recorded any more
Date: Fri, Feb 3, 2012 10:55
Hi all,
I'm still not getting very far with this one :(
Out of desperation I deleted and re-added one 6509 chassis and now it doesn't even show the stats for the fabric, it's just missing the whole "Catalyst 6k Crossbar" section?
I'm running the latest (2854) revision.
Any ideas please?
Robert Williams
Backline / Operations Team
Custodian DataCentre
tel: +44 (0)1622 230382
email: Robert(a)CustodianDC.com
http://www.custodiandc.com/disclaimer.txt
-----Original Message-----
From: Robert Williams
Sent: 01 February 2012 16:53
To: Observium Network Observation System
Subject: C6K Fabric stats not being recorded any more
Hi Guys,
For some (unknown) reason all of our catalyst chassis (650x-E) have recently stopped recording data for their Crossbar usage on all their cards. Data is visible up to a couple of weeks ago, but then stops and nothing more has been recorded since.
The permission on the rrd are fine, and the file itself hasn't been modified since the day the data stopped being recorded.
Interestingly, the usage 'bar' next to the (blank) graph shows usage on it correctly.
So, as a test, I moved the rrd files for one chassis out of its folder then did a full redetect of the one host in question. Since then it doesn't appear to have made any effort to re-create the RRD file :(
So, any ideas on why this one element would stop working suddenly? Or alternatively, is there a way to 'force' Observium to re-create it's rrd files for specific sensors on demand?
Any input most welcome :)
Cheers,
_______________________________________________
observium mailing list
observium(a)observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hi Guys,
For some (unknown) reason all of our catalyst chassis (650x-E) have recently stopped recording data for their Crossbar usage on all their cards. Data is visible up to a couple of weeks ago, but then stops and nothing more has been recorded since.
The permission on the rrd are fine, and the file itself hasn't been modified since the day the data stopped being recorded.
Interestingly, the usage 'bar' next to the (blank) graph shows usage on it correctly.
So, as a test, I moved the rrd files for one chassis out of its folder then did a full redetect of the one host in question. Since then it doesn't appear to have made any effort to re-create the RRD file :(
So, any ideas on why this one element would stop working suddenly? Or alternatively, is there a way to 'force' Observium to re-create it's rrd files for specific sensors on demand?
Any input most welcome :)
Cheers,
Robert Williams
Backline / Operations Team
Custodian DataCentre
tel: +44 (0)1622 230382
email: Robert(a)CustodianDC.com
http://www.custodiandc.com/disclaimer.txt
Hello!
the mail notification doesn't work after a down of a BGP peer.
it works when peering is going UP or FLAP!
I tried to understand where the problem was but I didn't succeed.. :(
thanks!
Dario
--
http://www.playnet.it
phone +39 055 3226622 - fax +39 055 3226444
Brain Technology s.p.a. - AS 13284
http://www.linkedin.com/in/dariocalamai
Hello,
is there a way o configure observium to send out a notification,
whenever a port is erroring and shown as such under the Ports-tab?
Regards
Timo
Hi, it would be really helpful for Capacity Planning to include
Trending support, as far as I know, RRDTool already has support for
it. Maybe adding trending on all graphs is not the best option,
because for one graph there are many data sources, and while one can
choose what one thinks would be the most obvious candidate another
user might need to do the trending using other datasource...
Maybe adding a separate TAB where one can choose what to graph with
trending would be the most "universal" solution, or maybe a mix of
both, graph the obvious by default and enable and "advanced" tab for
trending on demand...
Regards,
--
Ciro Iriarte
http://cyruspy.wordpress.com
--
Hi, I'll try asking again splitting topics... Does collectd add new
data/features not available with SNMP polling?, should I implement
collectd alongside a new Observium installation?. Also, which version
plays nice with Observium today?, 4.10 or 5.0?
Regards,
--
Ciro Iriarte
http://cyruspy.wordpress.com
--
Sorry if this has been asked before, couldn't find the answer in a
quick search. Graphing Linux load average on a busy machine I see
integer numbers, which is expected "20" for example, but graphing an
almost idle machine (load 0.22), it's graphed as 220 m. Is there
something I need to adjust?, is a RRD limitation?
Regards,
--
Ciro Iriarte
http://cyruspy.wordpress.com
--
Hi, I managed to install Observium rev2831 on SLES11SP1, and it's
working great!!
Trying to add hosts by CLI I was getting the "cannot ping" message,
fixed it adding "$config['fping'] = "/usr/sbin/fping";" to config.php.
Now, I keep getting the same trying to use the Web UI to add a hosts,
apparently it's not using the config['fping'] variable. Is that
possible?
Regards,
--
Ciro Iriarte
http://cyruspy.wordpress.com
--