Bug 294462 – postgresql to postgresql-{base,server} migration/build issues

• Product: Gentoo Linux
• Status: RESOLVED FIXED

[Description Carsten Lohrke (RETIRED) 2009-11-24 18:14:09 UTC]

I did the migration using 8.2.14 (which shouldn't really matter). After building postgresql-base, building postgresql-server failed:


make -C port all
make[2]: Entering directory `/var/tmp/portage/dev-db/postgresql-server-8.2.14/work/postgresql-8.2.14/src/port'
i686-pc-linux-gnu-gcc -O2 -march=athlon64 -pipe -mfpmath=sse -msse3 -funit-at-a-time -D_FORTIFY_SOURCE=2 -fstack-protector -Wall -Wmissing-prototypes -Wpointer-arith -Winline -Wdeclaration-after-statement -Wendif-labels -fno-strict-aliasing -fwrapv  -I../../src/port  -I/usr/include/postgresql-8.2/server -I/usr/include/postgresql-8.2/internal -D_GNU_SOURCE  -I/usr/include/postgresql-8.2/ -c strlcpy.c -o strlcpy_srv.o
strlcpy.c:35:15: error: c.h: No such file or directory
strlcpy.c:45: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘strlcpy’


Building postgresql-base again brought me past this issue. I can only imaginge the soft blocked and therefore late unmerged libpq to be the culprit. No idea, if this is considered to be a Portage (2.2_rc51) bug or if libpq should be blocked hard.


Afterwards postgresql-server failed to test, with the message


FATAL:  could not create lock file "/var/run/postgresql/.s.PGSQL.55432.lock": No such file or directory


in src/test/regress/log/postmaster.log.

Of course the test should not access /var. It'll be triggered by FEATURES+="userpriv usersandbox".The old dev-db/postgresql ebuilds never failed their test suite like that.

[Comment 1 Carsten Lohrke (RETIRED) 2009-11-24 18:47:33 UTC]

More issues:

/etc/init.d/postgresql from dev-db/postgresql has not been preserved, so I couldn't stop the running server the "Gentoo way". Moreso, the postgresql rc symlink has not been cared for - it's dangling. Just do not delete the init script, please. That's up to the administrator hands or via emerge --config maybe.



"/etc/init.d/postgresql-8.2 stop" spits out


/usr/lib/postgresql-8.2/bin/pg_ctl: invalid option -- 't'
Try "pg_ctl --help" for more information.

[Comment 2 Aaron W. Swenson 2010-03-24 00:06:27 UTC]

(In reply to comment #1)
> More issues:
> 
> /etc/init.d/postgresql from dev-db/postgresql has not been preserved, so I
> couldn't stop the running server the "Gentoo way". Moreso, the postgresql rc
> symlink has not been cared for - it's dangling. Just do not delete the init
> script, please. That's up to the administrator hands or via emerge --config
> maybe.
> 
> 
> 
> "/etc/init.d/postgresql-8.2 stop" spits out
> 
> 
> /usr/lib/postgresql-8.2/bin/pg_ctl: invalid option -- 't'
> Try "pg_ctl --help" for more information.
> 

Addressed in bug #311047

[Comment 3 Patrick Lauer 2010-06-02 17:22:38 UTC]

+  02 Jun 2010; Patrick Lauer <[e-mail]>
+  +postgresql-base-7.4.29-r1.ebuild, +postgresql-base-8.0.25-r1.ebuild,
+  +postgresql-base-8.1.21-r1.ebuild, +postgresql-base-8.2.17-r1.ebuild,
+  +postgresql-base-8.3.11-r1.ebuild,
+  +files/postgresql-base-8.4-9.0-heimdal_strlcpy.patch,
+  +postgresql-base-8.4.4-r1.ebuild, +postgresql-base-9.0_beta1-r1.ebuild:
+  Fixes for #313765, #251046, #294462, #300793, #274836, #296714, #238817,
+  #278228, #263096, #246397, #285953. Thanks to Aaron Swenson for collecting
+  the fixes and testing.