Broken upgrade - forum down, unhelpful error

Just got this while upgrading Discourse:

Oops

The software powering this discussion forum encountered an unexpected problem. We apologize for the inconvenience.

Detailed information about the error was logged, and an automatic notification generated. We’ll take a look at it.

No further action is necessary. However, if the error condition persists, you can provide additional detail, including steps to reproduce the error, by posting a discussion topic in the site’s feedback category.

What notification was generated? I haven’t received any email.

The advice to post a discussion topic doesn’t make sense - the forum is down.

It would be useful to list where the error information was logged.

I couldn’t tell what broke, so I rebooted the server. Now I get 502 Bad Gateway.

1 Like

If you remove all third-party plugins and rebuild, the error should go away. Once the site is up, go to /logs and see which plugin was causing the issue.

1 Like

You can also try safe mode - go to your site URL with /safe-mode appended to the URL. The logs you can look for at /logs. You can also try Announcing Discourse Doctor from the command line to look for trouble. Sometimes just rebuilding a second time from the command line brings the site back up.

The message you are seeing there is intentionally unspecific so it doesn’t cause alarm for your community members or confuse them. You are right to come here to ask for guidance on troubleshooting steps - in short order you will learn them and know what to do to figure things out.

Good luck and have fun! :sparkles:

1 Like

I’ll have to look up how to remove plugins. I only administer this forum once in the blue moon (thanks to the stability of Discourse).

In the meantime I tried ./launcher rebuild app which failed with an out of disk space error. After running cleanup per the prompt (14G free now), it failed with this:

Pups::ExecError: cd /var/www/discourse && su discourse -c ‘bundle exec rake db:migrate’ failed with return #<Process::Status: pid 430 exit 1>
Location of failure: /pups/lib/pups/exec_command.rb:112:in `spawn’

exec failed with the params {“cd”=>“$home”, “hook”=>“db_migrate”, “cmd”=>[“su discourse -c ‘bundle exec rake db:migrate’”]}
aa028fe633a223429da4368d3e6339a75329cdb0465361804372bc54b4fc8f35

** FAILED TO BOOTSTRAP ** please scroll up and look for earlier error messages, there may be more than one.
./discourse-doctor may help diagnose the problem.

Full log (66.2 KB)

1 Like
/var/www/discourse/plugins/discourse-tagging/plugin.rb:13:in `block in activate!'
/var/www/discourse/lib/plugin/instance.rb:430:in `block in notify_after_initialize'

Be sure to remove the discourse-tagging plugin ASAP. This plugin was deprecated about 3-4 years ago when we moved the functionality to core.

7 Likes

./discours-doctor seems to be having a syntax error:

Replacing: SMTP_PASSWORD
Replacing: LETSENCRYPT_ACCOUNT_EMAIL
Replacing: DEVELOPER_EMAILS
Replacing: DISCOURSE_DB_PASSWORD
Replacing: Sending mail to

==================== DONE! ====================
Would you like to serve a publicly available version of this file? (Y/n)
./discourse-doctor: line 320: [: ==: unary operator expected
Publicly available log not generated.

You can examine the output of this script with 
LESS=-Ri less /tmp/discourse-debug.txt

That was it, thanks for the prompt response @sam.

5 Likes

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.