During upgrade, I did cd /var/discourse; git pull; ./launcher rebuild app.
I get “Upgrade Complete” and “UPGRADE OF POSTGRES COMPLETE”. The latter says to “rebuild again”, though it seems to make no difference.
Despite upgrade being complete, I can’t start the app. Doing ./launcher start app gives the following error:
Unable to find image 'local_discourse/app:latest' locally
/usr/bin/docker: Error response from daemon: pull access denied for local_discourse/app, repository does not exist or may require 'docker login'.
See '/usr/bin/docker run --help'.
Can’t restart my app again, and not sure how to proceed.
./discourse-doctor doesn’t help (it just detects app isn’t running, tries to rebuild and restart, ends up with same error).
Can’t attach file, as a new user, so here’s what I think it the relevant snippet from debug log:
Failed to rebuild app.
Checking your domain name . . .
Connection to community.gingkoapp.com succeeded.
You should probably remove any non-standard plugins and rebuild.
Attempting to restart existing container. . .
+ /usr/bin/docker run --shm-size=512m -d --restart=always -e LANG=en_US.UTF-8 -e RAILS_ENV=production -e UNICORN_WORKERS=2 -e UNICORN_SIDEKIQS=1 -e RUBY_GLOBAL_METHOD_CACHE_SIZE=131072 -e RUBY_GC_HEAP_GROWTH_MAX_SLOTS=40000 -e RUBY_GC_HEAP_INIT_SLOTS=400000 -e RUBY_GC_HEAP_OLDOBJECT_LIMIT_FACTOR=1.5 -e DISCOURSE_DB_SOCKET=/var/run/postgresql -e DISCOURSE_DB_HOST= -e DISCOURSE_DB_PORT= -e DISCOURSE_DEVELOPER_EMAILS=REDACTED -e DISCOURSE_HOSTNAME=community.gingkoapp.com -e DISCOURSE_SMTP_ADDRESS=smtp.mailgun.org -e DISCOURSE_SMTP_USER_NAME=postmaster@community.gingkoapp.com -e DISCOURSE_SMTP_PASSWORD=REDACTED -h community-app -e DOCKER_HOST_IP=172.17.0.1 --name app -t -p 80:80 -p 2222:22 -v /var/discourse/shared/standalone:/shared -v /var/discourse/shared/standalone/log/var-log:/var/log --mac-address 02:fc:04:ee:f5:11 local_discourse/app /sbin/boot
Unable to find image 'local_discourse/app:latest' locally
/usr/bin/docker: Error response from daemon: pull access denied for local_discourse/app, repository does not exist or may require 'docker login'.
See '/usr/bin/docker run --help'.
Failed to restart the container.
I am in a space-constrained environment (I’m constantly being asked attempt to clean up images), so that sounds like a plausible reason for this failure.
Not sure how to do that without the web interface (or curl to the API), given that it’s down. But will look into it. I’m just going to upgrade my droplet disk size first and see if that helps.
If that is a VPS it isn’t necessary to upgrade the current server. Spinnning a new one running 20.04 and moving your site to that new server is way faster and easier.
Yes, that’s what I’m doing at the moment. Should be much faster than upgrading current… cross my fingers that I can restore from backup files (which I can still access on the old server).
Who said it’s supported?
Any docker version that is over a couple years old isn’t even supported by it’s developer then how do you expect it to be supported by discourse?
All’s well now. I spun up a fresh Digital Ocean droplet and restored the backup. The latest backup was corrupt, I think, but at least I only lost one or two posts and restored successfully from the backup before that.