Then I recommend to look into log files. For my case, after executing commands:

$ bitbake package-index -c cleansstate
$ bitbake -D -D package-index > bitbake-d-d.package-index 2> errlog

I have at least four log files.
1. bitbake-d-d.package-index in current dir
2. errlog   in current dir
3. run.do_package_index.* in
build/tmp-angstrom_v2012_05-eglibc/work/armv5te-angstrom-linux-gnueabi/package-index-1.0-r0
4. log.do_package_index.* in
build/tmp-angstrom_v2012_05-eglibc/work/armv5te-angstrom-linux-gnueabi/package-index-1.0-r0

(Actually, I skipped to generate 2., my build succeed.)
I guess you can find some clue looking into 4., 3., and maybe 1.

I attach mine to this mail, and hope they might be some help for you.
-- 
Takeshi Hamasaki

2013/2/19 Jaap de Jong <jaap.dejong@nedap.com>:
>
> On 02/18/2013 04:19 PM, Takeshi Hamasaki wrote:
>>
>> Hi,
>>
>> this may or may not help you:
>>
>> $ bitbake -e package-index > bbenv.package-index
>> $ grep PACKAGEINDEXES bbenv.package-index
>> (
>>   for me, the result was:
>>
>>   # PACKAGEINDEXES= package_update_index_ipk;
>>   PACKAGEINDEXES="package_update_index_ipk;"
>>
>> )
>>
>> Then I tried searching "package_update_index_ipk" in bbenv.package-index
>> using a text editor. It is a function, and in it, opkg-make-index
>> is called in a line:
>>
>>   flock $pkgdir/Packages.flock -c "opkg-make-index -r $pkgdir/Packages
>> -p $pkgdir/Packages -m $pkgdir/"
>>
>> What do you find in your box?
>
> For both I find the same results.
> The funny thing is that on the same machine my buildserver is running and
> that one is having no problems...
> Jaap
>
>
>
> _______________________________________________
> Openembedded-devel mailing list
> Openembedded-devel@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel