Segmentation Fault with pear on several machines and versions of FreeBSD

[eculp at bafirst.com]

Strange but effective Workaround for the problem.  See below.

Quoting eculp at encontacto.net:

> I've been updating all my ports and I started to get segmentation 
> faults with pear.  All I have to do is to type pear, let it list the 
> options and it finishes with the following error and threatens to 
> open a debugger that never opens.
>
> I now have three machines that have the problem.  There is little in 
> common between them.  One is a P4 running up to date current, another 
> is AMD Athlon also running up to date current and the last is a brand 
> new DELL P-4 Server with 2G of ECC memory that I just finished 
> installing RELENG_6, php5, php5-extensions, pear, etc. and have the 
> same problem.  All the other machines are a couple of years old.  All 
> three of the machines are cvsuping, building world, and kernels daily 
> with no problems at all.  All the systems libraries, includes, etc 
> are up to date.  I have installed world erased all more than a day 
> old and re installed world several times in the last few days on the 
> two older ones and the dell installation is less than 48 hours old 
> and has build and installed world twice.
>
> The simplest way to get the error is to just run the pear command and 
> it finishes with the following:
>
>  Type "pear help options" to list all options.
>  Type "pear help shortcuts" to list all command shortcuts.
>  Type "pear help <command>" to get the help for the specified command.
>  Segmentation Fault in 42156, waiting for debugger
>
> I have no idea what could cause the problem and much less, how to 
> troubleshoot it.  Google hasn't been my friend, yet, and that worries 
> me ;)  makes it sound more like operator error, but I have no idea 
> how.


On the new Dell Server, I decided to add debug support to php5 and 
php5-extensions to see if I could figure out what the problem was.  
Much to my surprise, I couldn't find the problem because all worked 
perfectly and had no pear "Segmentation Fault in 42156, waiting for 
debugger" anymore or any other errors.  I did recompile without debug 
support and I could no longer run pear so I compiled again and am 
running happily with debug and no errors.

The other machines were having similar problems but not as extreme as 
the Dell and compiling with debug seems to have fixed all my php 
related Seg Faults even the sig 11's with apache and the download 
popups for the ocassional php page.  All three are now running fine.  I 
still have another to test but I just want to get this in the archives 
just in case I'm not the only person who has similar porblems.

Comments still appriated.

ed

>
> Any ideas, suggestions or even confirmation that I'm not alone ;) 
> would be greatly appreciated.  I'm pretty sure that if I could get it 
> to not try to drop into the debugger I would at least be able to add 
> pear modules even though it would core dump but I have no idea how.
>
> Thanks,
>
> ed
>
> P.S. Versions of software that I have are:
>
>     php5-5.1.4
>     pear-1.4.6
>     I asked this question on the freebsd-questions list yesterday
>     but now I'm begining to think it might be more port related.
>
>
>
> _______________________________________________
> freebsd-ports at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-ports
> To unsubscribe, send any mail to "freebsd-ports-unsubscribe at freebsd.org"
>