Started new graph for the same port


I have an issue with port graphs.
For some random reason, it decided to start a totally new graph for the same port.
So now I have an old graph with good data, and a new graph, which only has a couple of hours of data and being added more.
How do I combine both graphs back into the old one and keep it updating to there?

Component | Version
--------- | -------
LibreNMS | 1.63-157-g2a6342f2c
DB Schema | 2020_04_06_001048_the_great_index_rename (165)
PHP | 7.2.9-1+b2
Python | 3.7.3
MySQL | 10.3.22-MariaDB-0+deb10u1
RRDTool | 1.7.1
[OK] Composer Version: 1.10.6
[OK] Dependencies up-to-date.
[OK] Database connection successful
[OK] Database schema correct


Whats your port association mode? You can check it in Devices -> Edit -> SNMP


It’s set to ifIndex.

Check eventlog for that device. You should see why it has been rediscovered.

By any chance…was the device rebooted before to the interface change?

Event log shows that interface changed from up to down and back up.
Device hasnt been rebooted.


It seems all 3 interfaces have generated new graphs : ptm0, ptm0.1 and ptm0.4

It’s my ISP vdsl2 router : Inteno DG400prime, and im monitoring 3 interfaces on it + 1 lan
Ptm0 (Combined of ptm0.1 and ptm0.4)
ptm0.1 (Lan+WiFi activity)
ptm0.4 (IPTV on vlan)

Maybe somehow the router changed the index.

What can you do now? Change port assoc mode to ifDescr or ifName, as they dont change.

To recover aggregate the data… your only option is to export both rrds, aggregate them and then create the rrd again

Thank you for helping.

If i don’t want to recover the new data and just continue with the old one where it stopped, should i just change the port assoc to either ifdescr or ifName and delete the new port rrd file?
I would lose the few hours it generated but thats not that important.

just cp the old rrd to new rrd name and you should be good to go. If you are using rrdcached, stop it first, then do the cp then start again.