[SOLVED] Connect to Grafana: /etc/pve/status.cfg file missing

[patcas]

Hi,

I want to connect my proxmox cluster (4 nodes) to grafana using an InfluxDB:
https://grafana.com/grafana/dashboards/10048

On the wiki, i read that I must change the content of the /etc/pve/status.cfg file (https://pve.proxmox.com/wiki/External_Metric_Server). I am now quite confused, because this file is missing on all my nodes. Should I just create a new file with the example content of the wiki. And if yes, on which of my nodes should I do that?

Thank you for your answer!

patcas

 

[Stoiko Ivanov]

Should I just create a new file with the example content of the wiki.

yes - and adapt that to your needs

And if yes, on which of my nodes should I do that?

That file and all other content in '/etc/pve' is store in the cluster filesystem (pmxcfs) - and is distributed across the cluster to all nodes - see:
https://pve.proxmox.com/pve-docs/chapter-pmxcfs.html

I hope this helps!

 

[patcas]

Thank you very much. Its works!

For everyone with external metric problems:

I struggeled a lot with letting the node send data to my influxdb.
1) I used the following command on my influxdb server to sniff packets:
tcpdump -n udp port 8089

2) I tried to sent packets by myself from my node to the influxdb server:
netcat -u influxdb.intranet 89
and checked, if the influxdb server recieves the packets using the above command

3) I consulted the on Proxmox Web UI under Node1>System>pvestatd>Log and saw, that proxmox is ignoring my status.cfg file, because I had an whitespace before the label influxdb (because I copied the content from the wiki):

_influxdb: influx1
___server 192.168.1.189
___port 8089

instead of:

influxdb: influx1
__server 192.168.1.189
__port 8089

(Whitespaces replaced with underscores)

Hope this helps somebody!

 

[Stoiko Ivanov]

Glad you found the issue!

please mark the thread as 'SOLVED'

 

[sfnme]

Hi,

I don't want to disturb for with opening new topic but may you tell us how did you connect your proxmox to grafana cloud ? How is your

Hi,

I want to connect my proxmox cluster (4 nodes) to grafana using an InfluxDB:
https://grafana.com/grafana/dashboards/10048

On the wiki, i read that I must change the content of the /etc/pve/status.cfg file (https://pve.proxmox.com/wiki/External_Metric_Server). I am now quite confused, because this file is missing on all my nodes. Should I just create a new file with the example content of the wiki. And if yes, on which of my nodes should I do that?

Thank you for your answer!

patcas

Hi,

I don't want to disturb with opening new topic for this issue. If is it not mind, could you explain how you connect your proxmox to grafana cloud ? I mean if you can share your /etc/pve/status.cfg so it can be very useful for us.

Thank you,
SFN

 

[patcas]

Hi SFN,

The following 3 lines are in my /etc/pve/status.cfg:


influxdb: influx1 server 10.0.0.53 port 8089

10.0.0.53 is the IP of my influxDB server.

Hope this helps!

patcas

 

[CRCinAU]

On a similar topic, are there any plans to add support via Prometheus?

 

[sfnme]

Hi SFN,

The following 3 lines are in my /etc/pve/status.cfg:


influxdb: influx1 server 10.0.0.53 port 8089

10.0.0.53 is the IP of my influxDB server.

Hope this helps!

patcas

Thank you for your care but my status.conf file has this information and i confused with ports 2003 or my proxmox port ? How should i fill ?

graphite: your-id
server your-server
port your-port
path your-path

and should i install influx db to my proxmox node ?