We are currently using Octopus Server v2018.8.7 and we have ~40 projects that include a trigger for automatic deployment with a filter of EventGroups = “Machine becomes available for deployment” to certain environments for certain octopus roles.
From what the documentation shows, these triggers should check for those events every 30 seconds. Unfortunately, when we scale out and ~400 new targets register with Octopus Server it seems after ~20 minutes these triggers stop working as new targets register. Sometimes they trigger 20-40 minutes later, but sometimes they don’t trigger at all.
I haven’t been able to find anything interesting in the “Auto Deploy logs” it is as if the octo process that checks for these events is failing. Is there something in the log or database that I should be searching for to try and find out what is going on? Is this a known issue with our Octopus Server version 2018.8.7?
Thanks for getting in touch! I had a look at our release notes since v2018.8.7 and there has been quite a few fixes around our triggers feature which could be responsible for this. We also have a current GitHub issue relating to Machine triggers sometimes missing machines. The GitHub issue does not have any helpful information on it yet unfortunately.
Our Compare Versions page has a complete list of our resolved issues. If you do a search for Trigger you will see multiple issues we have resolve around the feature since your current Octopus version.
I think the best option to resolve this would be to upgrade your Octopus server version. It may be a good idea to keep an eye on the above issue I linked to see when we release a fix for that. You can then upgrade to the latest Octopus Server version.
If you have any thoughts or questions on this, please don’t hesitate to let me know.
Thank you, Daniel. We’ve scheduled an upgrade to 2019.10.1 in the coming weeks and if we still have this issue after that we will watch that bug. I am hoping our bug is this one, because we are using more and more tenants now: