[Spacewalk-list] Ubuntu errata and Spacewalk 2.7

Paul-Andre Panon paul-andre.panon at avigilon.com
Tue Sep 12 01:52:39 UTC 2017


Hi,

Last week I finished updating our Spacewalk server to 2.7 and rebuilding all our Ubuntu channels and repos. I then tried to reload all the old Ubuntu errata with the same scripts I used in the past (using Philipp Schuler's parseUbuntu.py script to generate an input file for Pedro Andujar's errata-import.py) .... and I got a tiny number of matches and errata created. This was somewhat obfuscated by the fact that the September notification e-mail started including non-breaking spaces as UTF, which caused Philipp's script to break. So the first thing was to fix parseUbuntu.py to not choke on the UTF nbsp, and then I could look into what happened on prior months.

At first I was wondering if the PR500 Ubuntu package fixes have broken whatever search function Andujar's script uses in searching for packages, however that doesn't appear to be the case. Once I got September advisories parsed,  all 4 of those advisories/errata imported fine. However most of the previous monthly advisories appear to be treated as if they exist, even though they don't show up in the SW GUI. Here's the relevant portion of the errata import for August

...
[+] Retrieving data from ubuntu-errata.xml
[+] USN-3294-2 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3370-2 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3376-1 already exist: skipping
[-] USN-3375-1 already exist: skipping
[+] USN-3377-1 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3377-2 already exist: skipping
[-] USN-3378-1 already exist: skipping
[-] USN-3378-2 already exist: skipping
[+] USN-3212-4 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3339-2 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3379-1 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3380-1 already exist: skipping
[-] USN-3381-1 already exist: skipping
[+] USN-3381-2 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3382-1 already exist: skipping
[-] USN-3383-1 already exist: skipping
[+] USN-3384-1 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3384-2 already exist: skipping
[-] USN-3385-1 already exist: skipping
[-] USN-3385-2 already exist: skipping
[-] USN-3386-1 already exist: skipping
[+] USN-3386-2 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3387-1 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3388-1 already exist: skipping
[+] USN-3389-1 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3389-2 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3390-1 already exist: skipping
[+] USN-3391-1 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3392-1 already exist: skipping
[-] USN-3392-2 already exist: skipping
[-] USN-3391-2 already exist: skipping
[-] USN-3393-1 already exist: skipping
[-] USN-3394-1 already exist: skipping
[-] USN-3395-1 already exist: skipping
[+] USN-3393-2 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3391-3 already exist: skipping
[-] USN-3396-1 already exist: skipping
[+] USN-3397-1 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3398-1 already exist: skipping
[+] USN-3399-1 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3400-1 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3401-1 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3402-1 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3199-3 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3403-1 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3404-1 doesn't exist: analyzing
[-] No related packages found: skipping
[-] USN-3404-2 already exist: skipping
[-] USN-3405-1 already exist: skipping
[-] USN-3405-2 already exist: skipping
[-] USN-3406-1 already exist: skipping
[+] USN-3406-2 doesn't exist: analyzing
[-] No related packages found: skipping
[+] USN-3407-1 doesn't exist: analyzing
[-] No related packages found: skipping

If I do a SW GUI search for errata containing "USN", I don't see anything reported between USN-3024-1 and USN-3382-1, yet the list above shows many that could not be imported in that range because they already existed.

I mostly followed http://www.hrbac.cz/2017/06/proper-way-to-delete-channelrepositorypackages-in-spacewalk/ when dropping the channels before recreating them, but unfortunately I initially deleted the repos before the channels, hoping I wouldn't need to delete the channels. A number of orphan packages were cleared out when I did the channel deletes followed by package_removeorphans, but there doesn't appear to be any way to do the same for orphan errata. The errata are presumably still in there somehow, but not associated with the new (or any) channels and re-synced packages, and are blocking the re-import. Any suggestions on how to clean that up so that those errata can be deleted and re-imported? 

Thanks,

Paul-Andre Panon
Senior systems administrator
 

Office: 604.629.5182 ext 2341 Mobile: 604.679.1617





More information about the Spacewalk-list mailing list