Oneflow in HA mode

Hi, I’ve been trying to find documentation on how to deploy oneflow in a HA frontend setup. I’ve looked into this documentation: https://docs.opennebula.org/5.6/advanced_components/ha/frontend_ha_setup.html but nothing is mentionned about oneflow.

Now I assumed oneflow should run on all my 3 nodes, but i’m not sure if it should.
I also noticed oneflow process is getting shutdown every 24h or so on some on my nodes and I don’t understand why. I see nothing in logs execpt this in systemd:

Aug 30 10:55:50 ord-onebula-003 systemd[1]: Starting OpenNebula Flow Service...
Aug 30 10:55:50 ord-onebula-003 systemd[1]: Started OpenNebula Flow Service.
Aug 30 10:55:50 ord-onebula-003 ruby[3802581]: == Sinatra (v1.4.8) has taken the stage on 2474 for development with backup from Thin
Aug 31 07:59:15 ord-onebula-003 systemd[1]: Stopping OpenNebula Flow Service...
Aug 31 07:59:15 ord-onebula-003 ruby[3802581]: == Sinatra has ended his set (crowd applauds)
Aug 31 07:59:15 ord-onebula-003 ruby[3802581]: Thin web server (v1.7.2 codename Bachmanity)
Aug 31 07:59:15 ord-onebula-003 ruby[3802581]: Maximum connections set to 1024
Aug 31 07:59:15 ord-onebula-003 ruby[3802581]: Listening on 127.0.0.1:2474, CTRL+C to stop
Aug 31 07:59:15 ord-onebula-003 ruby[3802581]: Stopping ...
Aug 31 07:59:15 ord-onebula-003 systemd[1]: Stopped OpenNebula Flow Service.

Anyone can shed light on this ?

Thank you.


OpenNebula 5.6
Centos 7

Hello, in RAFT hook are systemd commands starting and stopping flow, gate, etc… so on all 3 nodes by default runs only oned

Ok, so according to the vip.sh script onegate and oneflow should be running only on the master node. And this script is starting and stopping them as needed.

It would be nice to have more details in the documentation tu be able to understand if and why there should be only one instance of each process in the cluster.

in this case you can open issue on github