[Bug 637134] Review Request: bird - BIRD Internet Routing Daemon
bugzilla at redhat.com
bugzilla at redhat.com
Mon Oct 4 13:10:34 UTC 2010
Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.
https://bugzilla.redhat.com/show_bug.cgi?id=637134
Jan Vcelak <jvcelak at redhat.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Flag| |needinfo?(jgorig at redhat.com
| |)
--- Comment #2 from Jan Vcelak <jvcelak at redhat.com> 2010-10-04 09:10:33 EDT ---
MUST ITEMS
[!!] rpmlint
bird.src: W: name-repeated-in-summary C BIRD
bird.x86_64: W: name-repeated-in-summary C BIRD
Seems to be fine. (However "Internet Routing Daemon" can be used as well.)
bird.x86_64: W: no-manual-page-for-binary bird6
bird.x86_64: W: no-manual-page-for-binary birdc6
bird.x86_64: W: no-manual-page-for-binary bird
bird.x86_64: W: no-manual-page-for-binary birdc
According to guidelines, all binaries/scripts should have manual page. This is
not as important to reject the review, but you should work with upstream on
this.
bird.x86_64: E: incoherent-subsys /etc/rc.d/init.d/bird bird6
bird.x86_64: E: incoherent-subsys /etc/rc.d/init.d/bird bird6
This is bad. I suggest creating separate initscripts and bird6 subpackage.
[ok] naming
[ok] licensing
source package doesn't contain file with license
[ok] spec: American English
[ok] spec: legible
[ok] spec: sources match the upstream sources
[ok] possible to build on rawhide (f15) x86_64 in mock
[ok] buildrequires
[ok] locales
none included
[ok] shared libraries
none included
[ok] no bundled libraries
[ok] clean section
[ok] relocatable, /usr prefix
[ok] owns all directories
[ok] no duplicates in %files
[ok] permissions on files set properly
[ok] consistent use of macros
[ok] contains code or permissible content
[ok] large documentation goes in -doc subpackage
not needed
[ok] will run properly without %doc files
[ok] headers in -devel subpackage
not needed
[ok] static libraries in -static subpackage
not needed
[ok] dynamic libraries without version specification in -devel subpackage
not needed
[ok] devel package dependencies
not needed
[ok] no libtool archives
[ok] desktop file for gui application
not gui application
[ok] doesn't own already owned directories
[ok] all filenames valid utf-8
SHOULD ITEMS
[--] license file in upstream release
please, query upstream to include license file
[--] package translation for Non-English languages
IMHO this is not needed for this package
[ok] builds in mock
rawhide x86_64
[ok] package functions
installs, starts, runs
[ok] sane scriptlets
[ok] subpackage dependencies fully versioned
[ok] pkgconfig files
not included
[ok] dependencies on packages instead of binaries
[--] man pages for all binaries and scripts
INITSCRIPT ISSUES
[!!] separate bird and bird6
Please, create separate initscripts for bird a bird6. And move all IPv6
stuff into bird6 subpackage. It will make many administrators happy. ;-) It's
done the same for example in Debian.
[!!] doesn't conform with initscripts guidelines
http://fedoraproject.org/wiki/Packaging:SysVInitScript
- missing condrestart and try-restart
- validate exit codes
OTHER ISSUES
[--] description
It's not important, but I think the description is really short. Maybe you
should point out command-line interface, soft reconfiguration and powerful
language for routing (as written on upstream website).
[!!] requires for post and preun
You are missing requirements for post and preun scriptlets.
Requires(post): chkconfig
Requires(preun): chkconfig
[!!] restart after upgrade, stop before removing
You should condrestart the service when the package is upgraded.
You should stop the service when the package is being removed.
(Don't forget to add 'Requires(...): initscripts' where necessary.)
SUMMARY
following issuses have to be fixed:
- initscript(s)
- specfile: requires for scriptlets
- specfile: restart/stop the service on upgrade/removal
--
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
More information about the package-review
mailing list