Upgrade failure with PG::InternalError


(Katja Durrani) #1

I got a blank page when following the upgrade link, so I tried upgrading manually with:

  cd  /var/discourse
  git pull
  ./launcher rebuild app

This failed with the following error message:

RuntimeError: cd /var/www/discourse && su discourse -c 'bundle exec rake db:migrate' failed with return #<Process::Status: pid 292 exit 1>
Location of failure: /pups/lib/pups/exec_command.rb:105:in `spawn'
exec failed with the params {"cd"=>"$home", "hook"=>"bundle_exec", "cmd"=>["su discourse -c 'bundle install --deployment --verbose --without test --without development'", "su discourse -c 'bundle exec rake db:migrate'", "su discourse -c 'bundle exec rake assets:precompile'"]}

There is also this message higher up:

I, [2015-05-29T18:57:39.834094 #38]  INFO -- : > cd /var/www/discourse && su discourse -c 'bundle exec rake db:migrate'
2015-05-29 18:57:52 UTC [302-1] discourse@discourse ERROR:  catalog is missing 2 attribute(s) for relid 16563
2015-05-29 18:57:52 UTC [302-2] discourse@discourse STATEMENT:  CREATE  INDEX  "index_topics_on_pinned_globally" ON "topics"  ("pinned_globally") WHERE pinned_globally
rake aborted!
StandardError: An error has occurred, this and all later migrations canceled:

PG::InternalError: ERROR:  catalog is missing 2 attribute(s) for relid 16563

Could somebody help me with this please? Thanks!

(Jeff Atwood) #2

Hmm, I kind of wonder if that’s a database integrity problem.

This is our standard Docker based install, yes?

(Katja Durrani) #3

Yes I followed the 30 min installation from your blog

(Jeff Atwood) #4

Hmm, I just updated discourse.codinghorror.com and talk.commonmark.org using the web updater (they are also Docker based installs on digital ocean) and no issues.

  • Is this on Digital Ocean or somewhere else?

  • Do you know roughly from what version you upgraded, was it recent or very old?

(Katja Durrani) #5

Yes, it is on Digital Ocean, and the previous version was 1.2.0.beta8

(Jeff Atwood) #6

If you can share credentials privately with @techapj from our team, he can assist. Can you private message him the information necessary?

(Katja Durrani) #7

Yes, okay, I will do that. Thanks for your help!

(Arpit Jalan) #8

I looked into this issue and it was due to corrupted Postgres database.

I would recommend everyone to enable “backup daily” site setting.

(system) #9