Thread: Urgnt. - problem starting postgres

Urgnt. - problem starting postgres

From
mohan ananthiah
Date:
hi,

   I have Red Hat Linux 7.2 with Postgres 7.1.3
 installed after compiling the source code with
 blocksize changed to 12K.  The installation goes
 smooth and initialization of db is done without
 any
 problems.  I started the postmaster and imported
 the
 data from an export. when I restart the postmaster
 again it gives me the following messages :

 DEBUG:  database system was shut down at
2002-02-25
 15:48:31 IST
 DEBUG:  ReadRecord: invalid magic number 0800 in
 logfile 0 seg 56 off 2408448
 DEBUG:  Invalid primary checkPoint record
 DEBUG:  ReadRecord: invalid magic number 0800 in
 logfile 0 seg 56 off 2408448
 DEBUG:  Invalid secondary checkPoint record
 FATAL 2:  Unable to locate a valid CheckPoint
 record
 /usr/local/pgsql/bin/postmaster: Startup proc 3058
 exited with status 512 - abort

 I am unable to figure it out.  Could someone
 please
 help me this is urgent.

 Thanx for your help in advance.

 Mohan A


__________________________________________________
Do You Yahoo!?
Yahoo! Sports - Coverage of the 2002 Olympic Games
http://sports.yahoo.com

Re: Urgnt. - problem starting postgres

From
Tom Lane
Date:
mohan ananthiah <techi_mohan@yahoo.com> writes:
>    I have Red Hat Linux 7.2 with Postgres 7.1.3
>  installed after compiling the source code with
>  blocksize changed to 12K.

Why in the world would you do that?

I don't believe that non-power-of-2 blocksizes will work.  Certainly the
only blocksizes that receive any meaningful testing are 8K and 32K.

            regards, tom lane