I am experiencing the same problem. Nothing else is running on Octopus other than an update of 2 linux servers. The “tentacle busy” message is actually happening on the retention policy piece. It seems to stop for no reason for about 5 minutes. We’re also seeing that tasks that should go to 2 servers in parallel are being deployed 1 node at a time.
We are currently on version 3.2.21 but I do not see anything in the release notes about fixes to this problem.
Thanks for getting in touch and reporting this. I had a chat to another member of the team and SSH tasks are currently restricted to running sequentially and it’s a known limitation. I’ve created a Github issue to get this fixed and you can follow it’s progress at the following URL.
In the short term, I think there are two options to try to resolve the issue.
Restart the Octopus service as per the previous ticket to reset any ‘stuck’ tasks.
Upgrade to the latest Octopus version. There was one fix in Octopus 3.3.1 that removes a file lock that could potentially be related.
That said, can you tell me if the the task ever completes successfully?
Thanks Rob. Yes it does complete successfully, it’s just that the long pauses make it take much longer than expected.
Just to be clear on the sequential issue with linux. We are deploying to 2+ different servers and we are seeing this behavior. We’re not trying to deploy in parallel to the same server. Is that the issue you documented?
We recently updated to version 3.3.9 and were expecting that this issue was resolved but it still occurs. A build is pushed to 2 linux machines but the deployments occur serially. Is this not resolved?