Scheduler not running after migrating from v10->v11 about docker-erpnext HOT 7 CLOSED

I can confirm: With the current updated images, the scheduler is started as expected.

from docker-erpnext.

Hello @paulator ,

Thank you for reporting this.
It's very possible that the issue is more related to the ERPNext migration itself than this docker container, but I will try to help you find out the root cause.

First thing that comes to my mind might be a silly question, but better safe than sorry: when you updated the docker-compose.yml, you updated all the containers version and not just the erpnext_app, right ? I'm not sure how ERPNext would behave if the app is in v11 but the schedulers in v10.

Next, is it possible for you to share the logs of the containers with docker-compose logs ?
If the scheduler is not running, I'm expecting to see an error message in the docker logs explaining the reason why it crashed.

Looking forward to your answer.

from docker-erpnext.

Regarding your questions:

Yes, all images are at the same version.

After resetting my complete test, I started with a blank v10 image / database and started the system with docker-compose up. In the logfile I found the following message from erpnext_app at the end of the initialization:

[app] [2019-10-21T18:58:24] Apps python packages installed
 [app] [2019-10-21T18:58:24] Setup logs folders and files owner to frappe...
 [app] [2019-10-21T18:58:24] Checking diagnostic info...
 -----Checking scheduler status-----
 Scheduler disabled for my.site
 Workers online: 0
 -----None Jobs-----
 [app] [2019-10-21T18:58:32] Everything seems to be good with your Frappe environment and background workers.
 [app] [2019-10-21T18:58:32] Starting app on port 8000...

from the scheduler I got the following logs:

 [scheduler] [2019-10-21T18:58:17] Waiting apps...
 [scheduler] [2019-10-21T18:58:27] Waiting for docker container init...
 [scheduler] [2019-10-21T18:58:27] Waiting for frappe current site to be set...
 [scheduler] [2019-10-21T18:58:27] Starting scheduler...

from docker-erpnext.

One more information: with the command
docker-compose run --rm erpnext_app bench --site all enable-scheduler
I'm able to start the scheduler...

from docker-erpnext.

Hello @paulator ,

Thank you for your answers.
We have been trying to reproduce this issue with the steps you provided without any success so far, but (luckily) we've had a test environment this week that had the exact same behavior.

We are currently testing a fix that should enable schedulers automatically after a successful bench doctor and call it after backup, restore and migrate.

We will let you know as soon as the fix as been deployed to the docker containers.

from docker-erpnext.

I will reopen this until we can add enhancements for the custom restore function that allow to set the private and public files and automatically execute bench doctor and enable schedulers.

from docker-erpnext.

Hello @paulator,

The restore function in the docker-frappe has been improved to accept to set the full path of the SQL, private and public files, so everything should be good now:

docker-compose run --rm erpnext_app bench restore /path/to/backup.sql /path/to/public.zip /path/to/private.zip

I'm closing this issue but feel free to report any new issues you may find.

from docker-erpnext.