[WARN] Your database schema has extra migrations (2025_05_28_092148_create_device_stats, 2025_11_26_170658_set_device_stats_timestamp_nullable). If you just switched to the stable release from the daily release, your database is in between releases and this will be resolved with the next release.
[OK] SQL Server meets minimum requirements
[OK] lower_case_table_names is enabled
[OK] MySQL engine is optimal
[OK] MySQL and PHP time match
[OK] Distributed Polling setting is enabled globally
[OK] Connected to rrdcached
[OK] Active pollers found
[OK] Dispatcher Service is enabled
[OK] Locks are functional
[OK] No python wrapper pollers found
[OK] Redis is functional
[OK] rrdtool version ok
[OK] Connected to rrdcached
Hi, I’m running 5 servers in a distributed poller setup and all of them are configured to use the stable release branch. Recently 4/5 servers have installed the developer branch “master” instead for unknown reasons and I would like help to know why because it’s not supposed to happen.
When checking the logs from daily.sh it never triggered an update and the script works as intended.
Somehow the installation happened randomly at different times triggered by unknown.
Has anyone else experienced this recently?
Please advise, thanks in advance.
Pastebin with ./validate output from all 5 servers:
I haven’t seen this happen automatically. It’s always been misconfiguration or a manual update.
I’d make sure that all are back on the same version and that nothing exists in config.php with regards to updates, then also check lnms config:get update and lnms config:get update_channel to make sure your configuration is still correct
Thanks for the response, do you recommend switching to the dev branch and installing the latest matching all 5 servers, followed by configuring stable release and waiting for december update?
config.php is only changed for the distributed polling configuration and nagios-plugins, nothing in regards to updates.
lnms config:get update give me “1” as output on all 5 servers and lnms config:get update_channel is set to “release” on all 5 currently.
Yes, switch to daily, update them all then switch straight back to the stable release. It won’t take you back to 25.11.0 but will stop anymore updates until 25.12.0
Hello again, I just wanted to let you know that all servers were upgraded to run the same dev version on November 27th. Directly after installing the latest dev version we switched to the stable version with the command “lnms config:set update_channel release” and confirmed its setting.
Two days later on November 29th it seems to have happened again where some servers upgraded at a random time during the day without using the daily.sh script. I still don’t understand how the servers upgrade without admin intervention.
Is there any upgrade log outside of daily.log that I can check to see what caused the upgrade?
I would like to apologize because I have now found out that the daily.sh script is actually the cause of the two upgraded servers. Still does not explain why it chose to install dev release when configured for stable release but that’s not the biggest issue.
Thank you @laf for the help, I wish you a nice day.
