Hello,
I have problem when I add Interface statistique on a host who already graph some othe interface, all the older one stop graphing
Do you have any issue of this probleme ?
In advance thank you for your help
Probleme when add interface on Create Graphs for this Host
Moderators: Developers, Moderators
Hi,
I have a similar problem.
After updating from cacti 0.8.5x to 0.8.7b, when I add a new interface graph (using Interface - Traffic template), old ones seems to stop graphing.
I checked in the mysql database "cacti", table "poller_item" : i can only see my new graph ?
Old graphs are still present in other tables (ie host, graph_local, host_graph)
I also do a "visual" test on a "migrated" device.
Before adding a new interface graph, i have 24 rows (it's a basic cisco switch).
Then, I add a new graph - the old graphs don't appear "faded" on the Create Graph list ! - and thus, I got only 2 rows in my table "poller_item" !!
Does anyone had the same problem and found any solution ?
Thanks for your help
I have a similar problem.
After updating from cacti 0.8.5x to 0.8.7b, when I add a new interface graph (using Interface - Traffic template), old ones seems to stop graphing.
I checked in the mysql database "cacti", table "poller_item" : i can only see my new graph ?
Old graphs are still present in other tables (ie host, graph_local, host_graph)
I also do a "visual" test on a "migrated" device.
Before adding a new interface graph, i have 24 rows (it's a basic cisco switch).
Then, I add a new graph - the old graphs don't appear "faded" on the Create Graph list ! - and thus, I got only 2 rows in my table "poller_item" !!
Does anyone had the same problem and found any solution ?
Thanks for your help
Hello,
I have that kind of message in my cacti.log file:
I have to admit I don't really understant what it means, but the "local_data_id" given in this ERROR message relies to one of the "old" graphs (graph created on cacti 0.8.5 and migrated to 0.8.7b).
Thanks for your help.
I have that kind of message in my cacti.log file:
Code: Select all
CMDPHP: Poller[0] ERROR: SQL Assoc Failed!, Error:'1064', SQL:"select snmp_query_graph_rrd.snmp_field_name, data_template_rrd.id as data_template_rrd_id from (snmp_query_graph_rrd,data_template_rrd) where snmp_query_graph_rrd.data_template_rrd_id=data_template_rrd.local_data_template_rrd_id and snmp_query_graph_rrd.snmp_query_graph_id=In/Out Bits and snmp_query_graph_rrd.data_template_id=41 and data_template_rrd.local_data_id=46613"
Thanks for your help.
I'm embarassed ... it was not 0.8.5, but 0.8.6 ...
To be honest, in 2006, a colleague installed cacti from debian sarge repos.
He assured me that the version installed was 0.8.5... i didn't check ...
But let's go straight to the interesting point :
A week ago, he decided to upgrade the hardware - and the OS, from sarge to lenny - and upgrade cacti as well (from lenny repos, 0.8.7b.), due to hardware limitations and the number of hosts ...
He told me this morning that no error occured when he imported the entries from the old DB to the new one, neither while importing old rra's, and "old" graphs were still getting data.
But since this upgrade, the original problem of this post happened.
It only occurs on interfaces graphs.
It looks like adding a new interface graph on an existing device (imported from the 0.8.6 version) do a "flush" on the poller_item entries related to this host - only interface graphs.
I'm trying to get more informations from my colleague (he is on holiday ... :p), but it's critical for my company to make it work correctly.
Thanks for your help ...
To be honest, in 2006, a colleague installed cacti from debian sarge repos.
He assured me that the version installed was 0.8.5... i didn't check ...
But let's go straight to the interesting point :
A week ago, he decided to upgrade the hardware - and the OS, from sarge to lenny - and upgrade cacti as well (from lenny repos, 0.8.7b.), due to hardware limitations and the number of hosts ...
He told me this morning that no error occured when he imported the entries from the old DB to the new one, neither while importing old rra's, and "old" graphs were still getting data.
But since this upgrade, the original problem of this post happened.
It only occurs on interfaces graphs.
It looks like adding a new interface graph on an existing device (imported from the 0.8.6 version) do a "flush" on the poller_item entries related to this host - only interface graphs.
I'm trying to get more informations from my colleague (he is on holiday ... :p), but it's critical for my company to make it work correctly.
Thanks for your help ...
- gandalf
- Developer
- Posts: 22383
- Joined: Thu Dec 02, 2004 2:46 am
- Location: Muenster, Germany
- Contact:
Is this really what he did? Please refer to the upgrade installations for cacti, see "documentation" tab above.Moox wrote:He told me this morning that no error occured when he imported the entries from the old DB to the new one, neither while importing old rra's, and "old" graphs were still getting data.
It is NOT recommended to import an old database into the new system. In fact, that will ALWAYS break cacti
Reinhard
Who is online
Users browsing this forum: No registered users and 3 guests