Opened 16 years ago

Closed 16 years ago

Last modified 14 years ago

#13634 closed defect (fixed)

need to run ranlib on cregistry.a

Reported by: jeffadams78@… Owned by: jmpp@…
Priority: Normal Milestone: MacPorts 1.7.0
Component: base Version: 1.6.0
Keywords: Cc:
Port:

Description

My port selfupdate command failed, as did downloading the source and trying to "make" it. Both failed because it said I "need to run ranlib on cregistry.a".

Manually running it and then trying the make again fixed it.

However that change is described in the below changeset, so I'm not sure why it is still an issue, unless this changeset is still in development? Anyway it seems pretty important since I couldn't selfupdate to, or build from source, the current released version (1.6.0).

http://trac.macports.org/projects/macports/changeset/32105

I am running Panther and had MacPorts 1.5.2

Change History (7)

comment:1 Changed 16 years ago by wsiegrist@…

Component: infrastructurebase
Owner: changed from wsiegrist@… to macports-tickets@…

Sending to base and macports-tickets as this is not an infrastructure issue. (The referenced changeset is in base also).

comment:2 Changed 16 years ago by jmpp@…

Milestone: MacPorts base bugs

Known bug that sadly didn't get fixed in time for 1.6.0 as I only caught it after releasing. It's fixed in trunk and will make the 1.6.1 release, so I'll be closing this ticket as fixed pretty soon.

Nevertheless, it intrigues me why you get this error on Panther, as I only experienced it while building MacPorts through the MacPorts Portfile *and* in universal fashion, which needless to say is not an issue on Panther. Even building MacPorts through the MacPorts Portfile but with universal flags turned off (which is basically the same as building MacPorts manually and/or selfupdating) succeeded. If you can help us, mind telling me what's the output of file base/src/cregistry/cregistry.a (with the proper path to the base directory adjusted in, of course) just after the failure? It'd be great to get such output both in the selfupdate case and when building MacPorts manually. Thanks!

-jmpp

comment:3 Changed 16 years ago by ryandesign (Ryan Carsten Schmidt)

Owner: changed from macports-tickets@… to jmpp@…

comment:4 Changed 16 years ago by jeffadams78@…

Here's the actual error, just in case it is helpful:

cc -dynamiclib -L/usr/local/lib  registry.o util.o entry.o entryobj.o -o registry.dylib -L/System/Library/Frameworks/Tcl.framework/Versions/8.4 -ltclstub8.4 -L/usr/local/lib   ../cregistry/cregistry.a -lsqlite3
ld: archive: ../cregistry/cregistry.a has no table of contents, add one with ranlib(1) (can't load from it)
/usr/bin/libtool: internal link edit command failed
make[2]: *** [registry.dylib] Error 1
make[1]: *** [all] Error 1
make: *** [all] Error 1

Here's the output of the file command:

jeff$ file src/cregistry/cregistry.a 
src/cregistry/cregistry.a: current ar archive

This is when building from source. Let me know how badly you need the output from a selfupdate... in order to get you that, I'll have to move aside my (now working) 1.6 install, download and install 1.5 (probably from source I guess, depends what archives are available for download), and then run selfupdate again. Unless there's an easy way to restore my old 1.5, but I doubt it since I "make install"ed the 1.6 version over top of it (didn't want to have to reinstall all my ports).

Also FYI, doubt it is related but just in case... I'm having trouble getting the cmake port installed (using 1.6). It fails:

[ 64%] Building CXX object Source/CMakeFiles/CMakeLib.dir/cmDependsJavaParser.o
*** malloc_zone_malloc[4222]: argument too large: 3609800964

cc1plus: out of memory allocating 3609800964 bytes after a total of 0 bytes
make[2]: *** [Source/CMakeFiles/CMakeLib.dir/cmDependsJavaParser.o] Error 1
make[1]: *** [Source/CMakeFiles/CMakeLib.dir/all] Error 2
make: *** [all] Error 2

Error: Status 1 encountered during processing.

comment:5 Changed 16 years ago by jmpp@…

Resolution: fixed
Status: newclosed

I don't think the cmake issue is related in any way, it seems to me like a compiler issue instead (either an Apple bug if using the default Mac OS X compiler, or a poor Portfile compiler choice if enforced at that level); please file it as a separate issue. As for the MacPorts build failure, it is indeed the same one I fixed in r32105, though surfacing due to a different reason this time round, it seems. But in any case, the fix will be in 1.6.1 so I'm going to close this ticket as fixed. Thanks for the report and feedback!

-jmpp

comment:6 Changed 15 years ago by tobypeterson

Milestone: MacPorts base bugsMacPorts Future

Milestone MacPorts base bugs deleted

comment:7 Changed 14 years ago by jmroot (Joshua Root)

Milestone: MacPorts FutureMacPorts 1.7.0
Note: See TracTickets for help on using tickets.