On 18/07/2007 5:44 PM, Garrett Cooper wrote:
Antony Mawer wrote:
On 18/07/2007 10:46 AM, Stephen Montgomery-Smith wrote:
I appreciate that most people won't have this problem, but it has
bitten me.
After you have made and installed a port, but don't clean it, and
then made a bunch of other ports, if you go back to the original port
and then do "make package", then +CONTENTS can be a bit messed up for
the package. This is because the creation of other ports might
disturb _LIB_RUN_DEPENDS and might put in some extra entries in
+CONTENTS.
This happens to me because I make all my ports on one machine and
then copy them as packages to other machines. Then on the other
machines, the structure of /var/db/pkg gets a bit messed up and
pkg_delete -r malfunctions.
It seems to me that the cure is to slightly change "make
actual-package-depends" so that if the port is already installed, it
just uses +CONTENTS.
I can't comment on the particular approach taken in your patch, but
can certainly attest to experiencing the same problem and it being
frustrating to identify what was going on. It was only after much
hair-pulling that I discovered that doing a 'make clean' at the
appropriate time before package building fixed the problem.
Otherwise I was winding up with plenty of seemingly OK packages that
were missing critical files (in this instance, various PHP5 extension
ports that were "installing" but missing the actual .so files!)
--Antony
Installing ports registers them on the machine as packages, by
simulating a package install via stdin. Was that forgotten?
-Garrett
The packages were definitely installed, by working through and doing
"make install" on the desired ports... I was aiming to uninstall
existing PHP5 packages on deployed servers, and then install from a
freshly updated set. The new ports were successfully installed, but for
whatever reason some of the packages created were missing the .so files.
I removed all the installed packages, make clean'd everything, then
started again and the next respin worked fine (without updating the
ports tree). Unfortunately I can't recall the exact thing that solved
it; I seem to recall a "make clean" was involved, but don't recall
whether it was explicitly running one before or after a "make package",
or whether it was *avoiding* running one...!
--Antony
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
