Commit fail at boot

Hello everyone,

First i’m a student in I.T in internship and i have to configure a cluster between 2 servers.
I’m here today to expose a problem that i’m encountering and that i can’t fix.
I’m sorry in advance for my english. I 'll try to explain correctly.

I’m trying to configure a cluster until like a week, and its fine, everything work except one thing.
The wrong thing is that at the start the configuration of the cluster disappear.
I searched on google, my logs and it look like to be a problem due to the interface starting.
Apparently the cluster is faster than interfaces. So after a reboot, the cluster config disappear, but the config is still on the config.boot.
So i have to relaunch the config to have the cluster up.

Do you have an idea to automate this things. Something like, modify the timer of the cluster start, or a script who auto relaunch the config?

I really thank you in advance.

Hello,
please advise which version you using,
also, as per description it looks like you just run live version from cd, without installing it,
so in that mode config will not survive a reboot

You need to install system, please refer to user guide on main web

Thanks!

We’ve got a similar issue, however ours running out of disk space. When we tried to se whats taking up the space, no files are larger than 500MB. We have 10GB total disk space. It looks like a bug, but are you able to advise if this is the case? We want to restart it and see if it reports the correct size. Unless something is hidden that we are unable to see.

I have the same problem with Lithium, PPPoE and NAT. It seems that NAT loads before PPPoE initialization, so it can not apply some rules, configuration fails and i can not see the configuration with “show configuration” despite /config/config.boot exists!
Here the bug
https://phabricator.vyos.net/T129