⚠️ Caution: redis:maintenance
will be deprecated soon, so the instructions on this page will have to be updated once better documentation is available.
https://devcenter.heroku.com/articles/data-maintenance-cli-commands#installation
states:These [data maintenance] commands include and expand on the functionality of the older pg:maintenance and redis:maintenance commands. Heroku plans to deprecate those commands in favor of the commands in this plugin.
The notes athttps://devcenter.heroku.com/articles/heroku-redis-version-upgrade may be helpful until Heroku updates their documentation.
...
You might get an email from heroku that begins:
...
(See Heroku developer setup for instructions on setting up heroku command line, including with -r production
configuration).
...
Disable staff access, so staff can’t trigger ingest with bg jobs that might not be able to be enqueued: heroku set LOGINS_DISABLED=true
Temporarily disable our hirefire auto-scaling manager, so it will allow our worker count to be scaled down to zero. Just toggle the “enabled” toggle for
worker
at https://manager.hirefire.io/Scale down workers to zero:
heroku ps:scale worker=0 -r production
Wait until worker is actually scaled down, you can see with
heroku ps -r production
Disable staff access, so staff can’t trigger ingest with bg jobs that might not be able to be enqueued:
heroku config:set LOGINS_DISABLED=true -r production
Because we use Heroku preboot, it can take 2-3 minutes to take effect, check to make sure you are really locked out of staff UI?
Run the maintenance now per heroku instructions, on production, eg:
...
heroku redis:maintenance --run REDIS --force -r production
When it’s finished, you should get an email, you can also check on status with
heroku redis:
...
info -
...
r production
Enable staff logins again:
heroku config:set LOGINS_DISABLED=false -r production
Scale workers back up to their default, probably 2 (if you get it wrong hirefire will fix it):
heroku ps:scale worker=2 -r production
Turn on hirefire manager again at https://manager.hirefire.io/
Because of heroku preboot, it could take 2-3 minutes for staff logins to be enabled again. Don’t leave until you confirm they are!
While this process should avoid it – if you wound up with any redis-related errors in Honeybadger despite yourself, go “resolve” them.
...