search

curationexperts-deprecated / ansible-hydra

Ansible playbook & roles to build a production-style Hydra Head.
http://www.curationexperts.com
Other
13 stars 13 forks source link

Rescue-Pool is in unknown state #268

Closed mark-dce closed 6 years ago

mark-dce commented 7 years ago

It feels like this error happens periodically. Capturing it here in case others have a solution. Otherwise, we should consider trying to refactor the role in a way that makes it more resilient.

TASK [sufia_dependencies/install : start resque] *********************************************************************************
 [WARNING]: The service (resque-pool) is actually an init script but the system is managed by systemd

fatal: [52.72.119.114]: FAILED! => {"changed": false, "failed": true, "msg": "Service is in unknown state", "status": {}}
    to retry, use: --limit @/Users/mark/Documents/workspace/_no_backup_/deploying_hydra/tufts-cm/tufts-ubuntu.retry
acozine commented 7 years ago

Thanks for including the full error message. What OS and version are you seeing this on? Is it Ubuntu 16?

mark-dce commented 7 years ago

OS + Version: Welcome to Ubuntu 16.10 (GNU/Linux 4.8.0-41-generic x86_64)

acozine commented 7 years ago

I'm seeing this from time to time as well. So far I can't tell what distinguishes runs where this task succeeds from those where it fails. Ultimately the answer may be to use sidekiq or some other background-job utility. Related to #206.

mark-dce commented 6 years ago

We've adopted sidekiq for all work going forward.