From: | "David Esposito" <pgsql-general(at)esposito(dot)newnetco(dot)com> |
---|---|
To: | <pgsql-general(at)postgresql(dot)org> |
Subject: | Using an alternate PGDATA on RHEL4 with SELinux enabled |
Date: | 2006-07-27 17:10:47 |
Message-ID: | 200607271659.k6RGxVX6028250@nncomail1.nnco.local |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
I just tried installing Postgres 8.1.4 (RPMs from postgresql.org web site)
on a clean RHEL4 Update 2 machine that had SELinux enabled.
When I created a /etc/sysconfig/pgsql/postgresql config file with
PGDATA=/data/pgdata
I was unable to get the start script (/etc/init.d/postgresql) to populate
the data directory nor launch postgres (after I manually did an initdb on
the directory)
The file permissions were
drwx------ 11 postgres postgres 4096 Jul 27 12:51 pgdata
But the pgstartup.log was still reporting that initdb coudln't create write
to /data/pgdata (Permission denied)
When I manually copied and pasted the line that was in
/etc/init.d/postgresql to the command line, it ran just fine (executing as
root)
runuser -l postgres -c "/usr/bin/initdb --pgdata='/data/pgdata'
--auth='ident sameuser'" >> "/var/lib/pgsql/pgstartup.log" 2>&1 < /dev/null
But there's some voodoo going on when this is executed inside of the start
script ...
I started monkeying around with cutting the start script down to the point
where I finally got this error to appear (I believe it was removing the
</dev/null redirect):
Your default context is user_u:system_r:unconfined_t. Do you want to choose
a different one? [n]
If I just hit enter, the script would continue and successfully create the
data directory and launch postgres
To be perfectly clear: If I don't set a custom PGDATA in
/etc/sysconfig/pgsql/postgresql, everything works fine ... the data
directory is created in /var/lib/pgsql/data as expected ... it's only with
the custom PGDATA ..
Long story short, I have disabled SELinux on this box because this isn't the
first time SELinux stuff has burned hours of my day and this is an internal
box so I don't feel it's worth the battle right now ... but figured that
maybe someone else out there might benefit from reading this ...
-Dave
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Eisentraut | 2006-07-27 17:22:17 | Re: UTF-8, upper() and Chinese characters yielding blank result |
Previous Message | Karen Hill | 2006-07-27 17:07:50 | PostgreSQL theoretical maximums. |