Bug 251871 - init.d/postresql startup failure
Summary: init.d/postresql startup failure
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: Fedora
Classification: Fedora
Component: postgresql
Version: 7
Hardware: All
OS: Linux
low
high
Target Milestone: ---
Assignee: Tom Lane
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2007-08-13 01:00 UTC by Tim Malnati
Modified: 2013-07-03 03:14 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2007-08-13 03:10:39 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)

Description Tim Malnati 2007-08-13 01:00:38 UTC
Description of problem: Postgresql fails to startup due to version mismatch
error.  init.d script is hard coded with "PGVERSION=8.1.a" causing script to
fail version checking.


Version-Release number of selected component (if applicable):
postgresql-server-8.2.4-1.fc7
/etc/rc.d/init.d/postresql


How reproducible: 


Steps to Reproduce:
/etc/rc.d/init.d/postgresql start (2nd and subsequent strat w/new database)
  
Actual results:
An old version of the database format was found.
You need to upgrade the data format before using PostgreSQL.
See (Your System's documentation directory)/postgresql-8.1.a/README.rpm-dist for
more information.
fails to start after new database is created (2nd & subsequent attempts)


Expected results: normal startup


Additional info:

Comment 1 Tom Lane 2007-08-13 02:14:15 UTC
[ blink ] ... Not here it doesn't.  To double-check, I pulled down both i386 and x86_64 versions of 
postgresql-server-8.2.4-1.fc7 from download.fedora.redhat.com, and they contain the expected
PGVERSION=8.2.4
So the salient question seems to be, exactly which postgresql-server RPM did you get, from exactly where?

Comment 2 Tim Malnati 2007-08-13 02:36:56 UTC
Postgres was loaded some time ago but it unclear at the moment if it part of 
the original load or updated.  It probably came form a mirror where I use yum.  
I looked for postgresql.rpmnew and there was none though.  

If you got the correct versioning, the I suspect mine was a remnant.  Although 
I have to wonder why this would be hard coded in the script anyway.

Thanks,  Tim

Comment 3 Tom Lane 2007-08-13 03:10:39 UTC
The number is inserted into the script during RPM build, which is the same time that the binary 
executables have it hard-wired into them --- I see nothing very wrong with that.

I'm going to assume that you got a corrupt download and close this WORKSFORME.  Please reopen if you 
find some evidence that the file is actually corrupt on any Fedora distribution site.


Note You need to log in before you can comment on or make changes to this bug.