Thanks for that, I got the spec file and might have added the patches properly...
Having a problem when trying to build the rpm with mock though.
(x86_64 / mock-0.6.13-1.el5.centos.1)
I installed the perl src.rpm with rpm -Uvh, modified the spec file and am using a default mock cfg (copied).
When I run mock -r el5-x86_64 perl-5.8.8-10.el5_0.2.src.rpm I continue to get this error
Code: Select all
Traceback (most recent call last):
File "/usr/libexec/mock-yum", line 12, in ?
...
File "/usr/lib/python2.4/site-packages/yum/rpmtrans.py", line 355, in _instOpenFile
OSError: [Errno 2] No such file or directory: '/var/lib/mock/epel-5-x86_64/root/var/cache/yum/os/packages/python-2.4.3-21.el5.x86_64.rpm'
Error performing yum command: /usr/sbin/mock-helper yum --installroot /var/lib/mock/epel-5-x86_64/root install buildsys-build
...
Even though...
Code: Select all
-rw-r--r-- 1 root mock 6236478 Jun 15 07:01 /var/lib/mock/epel-5-x86_64/root/var/cache/yum/os/packages/python-2.4.3-21.el5.x86_64.rpm
I tried downloading a fresh copy of the python rpm to that location, but no diff.
When I installed the perl src.rpm I got errors about user and group mockbuild not existing, but assumed those were ok to ignore - is that right?
Is the mock cfg here a good example of what I should use (wondering about the repos in particular)? -
http://www.atomicorp.com/Tutorials/Building%20RPMS/.
So, the patches... Comparing the existing patch files to what I got from
http://use.perl.org/~nicholas/journal/37274 I see the existing files have a different diff format... all I did was paste the diffs linked to from the perl.org page into a file for each (perl-5.8.8-bsXXXX.patch in SOURCES)... is that it? Also, any patch naming convention I'm violating?
Thanks!
[UPDATE]
Now that I've looked in my logs... it seems grsec was blocking rpm -q for that user. Now that's out of the way the build process completes but one of the tests fails
Code: Select all
lib/Benchmark.............................# Failed test 'is 960 within 0.4 of estimate (1461.33)'
# in ../lib/Benchmark.t at line 80.
# Looks like you failed 1 test of 194.
FAILED at test 13
I'm not sure what the Benchmark test tests, but imagine my patches didn't work or the performance bug wasn't fixed with them?
Will look into it later, now for some r&r