Confluence upgrade 6.13.0 to 6.15.2: Db collation 'C' is not supported Edited

[Steve Allen]

Updating confluence after email: Confluence Security Advisory - 2019-04-17

Pre-production test.
Upgrade from 6.13.0 --> 6.15.3 (atlassian-confluence-6.15.2-x64.bin)
Debian 8 with PostgreSQL 9.6.10

Upgrade proceeded without error. Upon loading browser for first time:

============================================

Preparing to start Confluence
This page is for Confluence administrators. If you're seeing this page, your Confluence administrator is probably working to restore the service.
We discovered a problem while preparing to start Confluence. If you're a Confluence admin, you can either fix this problem now or ignore it, and start Confluence anyway.
Database: The database collation 'C' is not supported by Confluence. You need to use 'utf-8'.

=============================================

I selected 'ignore it'.
I need to resolve prior to production upgrade. Please advise :-)

[Petr Vaníček · Accepted solution]

Hi,

just follow this article: https://confluence.atlassian.com/confkb/how-to-fix-the-collation-of-a-postgres-confluence-database-966676071.html 

[Steve Allen]

Thanks and I choose 'method 2'.
If it should help anyone, the following worked for me. Debian 9, Postgreaql 9.6, Confluence 6.13.

From command line.

1) $ sudo systemctl stop confluence
2) Backup - as owner user ie '<myUser>':
      $ pg_dump confluencedb > dumpfile1
3) Delete original db via psql: DROP DATABASE confluencedb;
4) Make new databse via psql:
      CREATE DATABASE confluencedb WITH OWNER <myUser> ENCODING 'UTF-8' LC_COLLATE 'en_GB.UTF-8' LC_CTYPE 'en_GB.UTF-8' TEMPLATE template0;
      GRANT ALL PRIVILEGES ON DATABASE confluencedb to <myUser>;
5) Restore:
      $psql confluencedb < dumpfile1
6) $ systemctl start confluence

I was then able to update to 6.15.2 without error.

[Deleted user]

Yes, this work also in my instances, but after that, now I have a lot of 400 Errors like:

java.lang.IllegalArgumentException: Invalid character found in the request target. The valid characters are defined in RFC 7230 and RFC 3986
 org.apache.coyote.http11.Http11InputBuffer.parseRequestLine(Http11InputBuffer.java:474)
 org.apache.coyote.http11.Http11Processor.service(Http11Processor.java:294)
 org.apache.coyote.AbstractProcessorLight.process(AbstractProcessorLight.java:66)
 org.apache.coyote.AbstractProtocol$ConnectionHandler.process(AbstractProtocol.java:836)
 org.apache.tomcat.util.net.NioEndpoint$SocketProcessor.doRun(NioEndpoint.java:1747)
 org.apache.tomcat.util.net.SocketProcessorBase.run(SocketProcessorBase.java:49)
 java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
 java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
 org.apache.tomcat.util.threads.TaskThread$WrappingRunnable.run(TaskThread.java:61)
 java.lang.Thread.run(Thread.java:748)

In the version before, I did not have this problems, but after changing collation, I have. Any ideas how to fix confluence?

[Robert Swift]

Thanks @Steve Allen this is the sort of simple, accurate information that makes community forums so worthwhile :)