From: | colum(at)mail(dot)com (colm) |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | RedHat6.2 - postgres 7.1.2 lib confusion |
Date: | 2001-12-04 15:42:39 |
Message-ID: | 22bee7f2.0112040742.745515e8@posting.google.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Hello all,
I've trawled the archives and found stuff similar to my problems but
(as usual) not quite...
Basically, I am trying to move from 6.5.3 to 7.1.2 on RedHat 6.2 with
as little mucking around with RH as possible. I have an upgraded rpm
(3.0.5) so I was able to fetch the 6.2 rpms from the ftp area and
kicked off with:
postgresql-7.1.2-4PGDG.i386.rpm
postgresql-docs-7.1.2-4PGDG.i386.rpm
postgresql-libs-7.1.2-4PGDG.i386.rpm
postgresql-server-7.1.2-4PGDG.i386.rpm
Immediately calling rpm -Uvh on these gives a dependency
problem:error: failed dependencies:
libcrypto.so.0 is needed by postgresql-7.1.2-4PGDG
libssl.so.0 is needed by postgresql-7.1.2-4PGDG
libcrypto.so.0 is needed by postgresql-libs-7.1.2-4PGDG
libssl.so.0 is needed by postgresql-libs-7.1.2-4PGDG
libcrypto.so.0 is needed by postgresql-server-7.1.2-4PGDG
libssl.so.0 is needed by postgresql-server-7.1.2-4PGDG
I find this odd because I have libcrypto.so.02 and libssl.. Heres my
/usr/lib :
lrwxrwxrwx 1 root root 14 May 23 2001
/usr/lib/libcrypto.so -> libcrypto.so.0*
lrwxrwxrwx 1 root root 18 May 16 2001
/usr/lib/libcrypto.so.0 -> libcrypto.so.0.9.6*
-rwxr-xr-x 1 root root 876k Apr 18 2001
/usr/lib/libcrypto.so.0.9.6*
-rw-r--r-- 1 root root 244k Nov 10 2000
/usr/lib/libssl.a
lrwxrwxrwx 1 root root 11 May 23 2001
/usr/lib/libssl.so -> libssl.so.0*
lrwxrwxrwx 1 root root 15 May 16 2001
/usr/lib/libssl.so.0 -> libssl.so.0.9.6*
-rwxr-xr-x 1 root root 198k Apr 18 2001
/usr/lib/libssl.so.0.9.6*
So I carried on with a --nodeps. But now I felt a little unsure. Is
this yet more evidence of older versions of rpm being a complete PITA?
Perhaps its the version of the rpm executable conflicting with the
actual rpm archives..
Anyway, calling the /etc/rc.d/init.d/postgres start seemed to
initialise the database but failed on starting the service.. I
therefore thought I would start postmaster manually (assuming that the
init script had already called initdb) and got this upon calling
postmaster -D /usr/share/pgsql :
postmaster does not find the database system.
Expected to find it in the PGDATA directory "/usr/share/pgsql",
but unable to open file "/usr/share/pgsql/global/pg_control": No such
file or directory
All that is in here are the sample versions of the various conf files
i.e.
drwx------ 2 postgres postgres 4096 Dec 4 13:44 backup
-rw-r--r-- 1 root root 770 Jun 13 22:36 global.bki
-rw-r--r-- 1 root root 28 Jun 13 22:36
global.description
-rw-r--r-- 1 root root 7399 Jun 13 22:36
pg_hba.conf.sample
-rw-r--r-- 1 root root 1118 Jun 13 22:36
pg_ident.conf.sample
-rw-r--r-- 1 root root 3137 Jun 13 22:36
postgresql.conf.sample
-rw-r--r-- 1 root root 196935 Jun 13 22:36 template1.bki
-rw-r--r-- 1 root root 27412 Jun 13 22:36
template1.description
Do I need to manually alter these? I tried to manually run the initdb
but it complains that the directory should be empty... Should I just
delete the whole dir and re-call postgres start?
Other things: I am running the various manual scripts as user
postgres. I had a few test databases on my old data dir in
/var/lib/pgsql but I don't need the data - should I just delete these
and the backup directory in /usr/share/pgsql? I don't have any
postgres-related environment variables set.
I would be most grateful for any advice...
Many thanks,
colm
From | Date | Subject | |
---|---|---|---|
Next Message | Serkan BEKTAŞ | 2001-12-04 16:07:28 | FW: Localization - Turkish - Tr |
Previous Message | Johnny Jørgensen | 2001-12-04 15:28:47 | plpgsql commenting broken? |