← Back to team overview

desktop-packages team mailing list archive

[Bug 1392057] Re: pg_createcluster on an existing data directory can't be repeated

 

That's the documented behaviour of pg_createcluster, and it tells you
what happens too. I'll definitively not just copy the file and leave the
original, that would be utterly confusing and prone to getting out of
sync. A symlink could work though.

** Changed in: postgresql-common (Ubuntu)
   Importance: Undecided => Wishlist

** Changed in: postgresql-common (Ubuntu)
       Status: New => Triaged

** Summary changed:

- pg_createcluster on an existing data directory can't be repeated
+ pg_createcluster on an existing data directory: symlink config files instead of moving them

-- 
You received this bug notification because you are a member of Desktop
Packages, which is subscribed to postgresql-common in Ubuntu.
https://bugs.launchpad.net/bugs/1392057

Title:
  pg_createcluster on an existing data directory: symlink config files
  instead of moving them

Status in postgresql-common package in Ubuntu:
  Triaged

Bug description:
  My release is 14.04. postgresql-common is 154.

  When running pg_createcluster to create a cluster for which the data
  directory already exists, pg_createcluster copies, and then deletes
  the configuration files from the data directory. The deleting seems
  unnecessary, and it seems like a symlink would be better suited here.
  At the very least, I see no reason to delete the originals.

  Specifically, this is a glaring flaw if I have established a database
  on a different partition from my Ubuntu installation and I want to
  wipe and reinstall Ubuntu while migrating this data. This is simple in
  theory, but deleting the config files from the data directory creates
  a frustrating barrier if I want to do this more than once.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/postgresql-common/+bug/1392057/+subscriptions


References