# curl http://porkchop.redhat.com/nightly/RHEL5.3-Server-20080919.nightly/tree-i386/.discinfo 1221803353.466553 Red Hat Enterprise Linux Server 5.3 i386 ALL Server/base Server/RPMS Server/pixmaps # curl http://porkchop.redhat.com/nightly/RHEL5.3-Server-20080919.nightly/tree-i386/.treeinfo [general] family = Red Hat Enterprise Linux Server timestamp = 1221802175.49 totaldiscs = 1 version = 5 discnum = 1 packagedir = Server arch = i386 ... Should the fix that changed the .discinfo version should be carried forward into the version in .treeinfo?
Just to clarify, is this what you would expect in .treeinfo? [general] family = Red Hat Enterprise Linux Server timestamp = 1221802175.49 totaldiscs = 1 version = 5 update = 3 discnum = 1 packagedir = Server arch = i386 Currently the .treeinfo file is created by anaconda and distill just copies it. I could add logic to distill to tweak the .treeinfo file, but I would be in favor of making the change in anaconda itself. What do you think?
I'd prefer keeping the existing fields (and not adding a new key 'update') ... but I know bpeck may want something different? [general] family = Red Hat Enterprise Linux Server timestamp = 1221802175.49 totaldiscs = 1 version = 5.3 Yeah Dennis, I think you are right in that anaconda should just write the product info that is passed in to builddistro. Is anaconda mk-images being supplied with the right information in order to write out the version string as noted above? Or are additional parameters required?
I pass "--version '5'" to buildinstall. I could modify that to 5.3 or include a new parameter. If I just change it to "--version '5.3'" then we _may_ be able to get away without making any changes to buildinstall. However, I don't know enough about the buildinstall scripts to say if that change would cause breakage elsewhere.
Adding jgranado as Joel instrumented the anaconda .treeinfo parsing for RHEL5.2
The "5.3" argument should be fine. Be aware that you will also be changing the /root/.buildstamp file.
So far rhel5.4 builds are not showing the correct version. Any ETA on this? version = 5.4 would be fine
change checked into distill.
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHEA-2009-1400.html