Agents not triggered at scheduled time? #2667
Comments
|
Alright I got ahold of the admin account to erase the failed jobs. That means if the agent is launched, it will send an email with the content of it's memory? |
|
I see there are a bunch of errors about postgresql I see: |
|
Alright I restarted from scratch the docker image, everything seems fine, I'll see if the error comes back. |
Huh interesting, I have never seen that error before. Do you think the postgres error prevented the Agents from being scheduled? |
|
Indeed I did just a test, that's what happened! My db (and huginn too) are installed with ansible (a provisioning tool) and sometimes they are stopped and restarted => looks like it impacts Huginn. I'm not a ruby developer so I don't know if it can be fixed in Huginn, but I can make sure Huginn is restarted every time the db is stopped. Do you know if there is a way to auto reconnect when there is a lost connection? |
|
I am pretty sure the Rails application itself handles reconnections automatically. For the background processes we would need to look into that. |
|
I've been digging into a similar issue as the top post of this issue. I realised this week when I started setting some email digests up that the times must be way out. I went in and set the timezone correctly to "London" and restarted both the web and worker containers (using the single-threaded postgres docker-compose) but to no avail. I'm thinking I might need to tear down the compose stack and restart it for that setting to actually come through. I've not checked any of the logs though as I've only spent 1 evening on this so far and did the above. I can't find the "Job Management" page either so can only assume it doesn't exist anymore. But I'll inspect the logs to see if I can find what the problem is. By the sounds of it, a full restart might be the best option to get the setting respected. |
If you don't see that your user probably isn't an admin. |
|
To inspect the logs, as I'm using the docker image, I use: (PG should be PostgreSQL for example.) I'll see if I can do some testing. @BlueHatbRit, if you haven't configured anything about the admin user, the default password should be "password" (as written here: https://github.com/Videl/huginn/blob/fb13686598596d648f3dd422288cfe87b69d8244/docker/multi-process/README.md) |
|
Hey both, thanks for the info about the admin user, I totally missed that in the docs. I've sorted that out and the guides all make a lot more sense now, also I've closed that potential security hole 😄 I don't have any logs like the ones you're seeing @Videl so I'm afraid I can't be of any help on that one. @dsander good shout, I've not used compose for a few months and went for a restart for a |
|
Just experienced the same problem (job that is scheduled to run every 12h) didn't run for over 30 hours now. Before killing and restarting the container, I have verified that connection is actually still established: both of them sit idle: The container logs are available here. Could this be related to: ? |
Hello,
I'm fairly new to Huginn :)
I have noticed many jobs do not run on schedule. Is there anything I can do to debug what's happening?
I have ensured that I have set up the right timezone because the date of 'now' in events is correct when I trigger them myself.
I have seen comments about a "Job Management" page but I can't see it (#1668)? I have seen issues talking about admin users but I didn't see any choices? I had to use the invitation code to create an account.
I'm using the docker hub huginn/huginn image, with those variables setup:
Thanks for any help :)
The text was updated successfully, but these errors were encountered: