+++ This bug was initially created as a clone of Bug #452476 +++ Description of problem: From rhn-satellite: Hi again and once again, thanks for the tips. I have been experimenting with the script, and find that for our Lab Satellite here we get the following traceback for certain channels: Channel: Red Hat Desktop (v. 3 for x86) Label: rhel-i386-desktop-3 Traceback (most recent call last): File "repomd-precache.py", line 28, in ? repo.get_repomd_file() File "/usr/share/rhn/server/repomd/repository.py", line 326, in get_repomd_file self.repository.get_updateinfo_xml_file(), File "/usr/share/rhn/server/repomd/repository.py", line 106, in get_updateinfo_xml_file viewobj.write_updateinfo() File "/usr/share/rhn/server/repomd/view.py", line 305, in write_updateinfo self.fileobj.write('\n'.join(self._get_erratum(erratum))) File "/usr/share/rhn/server/repomd/view.py", line 289, in _get_erratum output.append(" <description>%s</description>" TypeError: cannot concatenate 'str' and 'NoneType' objects Version-Release number of selected component (if applicable): 520 How reproducible: the bug is obvious Steps to Reproduce: 1. sync some channel with errate which has some synopsis and empty description (or reverse: some description and none synopsis) 2. regenerate repomd cache either by requesting data from it by client or regenerate using rhnChannel.channel_info API call. Actual results: Traceback Expected results: Works as charm. -- Additional comment from msuchy on 2008-06-23 05:24 EST -- Fixed in trunk rev. 174335
Commited to git repo as b1966fff0f8cade81ff49d88da3b177b53a30f6d
Does it make sense to verify this against spacewalk? (satellite-sync isn't really commonly used is it?) I understand we can use it to populate from a mounted internal share of packages to possibly replicate, could anyone confirm if syncing say RHEL 5 Server i386 (via this method) would include errata required to cover this?
Spacewalk is released for long time.