hping2 orphaned? Joining Extras

Michael Schwendt bugs.michael at gmx.net
Fri Jul 1 17:31:29 UTC 2005


On Sun, 19 Jun 2005 01:40:54 -0400, Pete Toscano wrote:

> Will sign the CLA on Monday, when I'm in the office.
> 
> >  2) query http://bugzilla.redhat.com for possibly open bug reports
> >     which call for your attention as package maintainer, and present
> >     an updated package if necessary (for urgent fixes, it can be
> >     imported and built quickly by an existing contributor before
> >     your cvs account would be ready)
> 
> Done.  Found one report
> (https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=149776).  I
> haven't investigated this much yet, so I could be wrong, but it looks to
> me that this guy's beef is with Salvatore Sanfilippo, the hping2
> maintainer.  He should probably report a problem such as this on the
> hping2 development lists, shouldn't he?  That said, it looks like it
> wouldn't be that hard of a problem to fix, so I might just submit a
> patch to Salvatore.

Okay. Sorry for the late reply. I'm facing a backlog of mail, as often I
mark/flag some messages and plan to return to them later.

Above bug report is from somebody, who uses an Intel compiler to
test-build all of Extras (and possibly Core, too), because he believes
Intel's compiler offers more detailed checking for warnings and errors.
He then assumes that he can report all found issues in a single tracker,
Fedora Extras bugzilla, where the many package owners then work like ants
and forward such reports to the many many many upstream projects.

It's something we have to live with. Some of the findings are real bugs,
which can lead to malfunctioning code. Some are false positives, e.g.
uninitialised variables in conjunctions with call by references, or
compiled but unused (!) code sections. 

> I want to be sure I understand where our responsibilities lie as package
> maintainers.  It's our job to take the releases (and possibly useful
> patches too) and bundle them up in RPMs for release in Extras, right?

Yes.

Plus, depending on upstream's release habits, you might want to evaluate
the latest release some time before taking it as an update.

> If a bug report comes in asking for code fixes or extra features
> (possibly aside from already-existing patches) that don't involve the
> packaging, the it's best to refer the submitter to the program's
> development list, right?

Yes, IMO.  Even if you keep contact with the developers, for feature
requests or requests for changed behaviour, it's best for end-users
to propose changes in upstream's mailing-lists or bug trackers, so
they are available in case questions are raised.

> We're not expected to actively participate in
> the program's development (although that wouldn't hurt).  We should be
> monitoring the programs we maintain for new releases and possibly for
> some important patches.  Does this about sum it up?

Yes.  You are free to participate in development or in creating patches
for bugs, which are reported to you. Some packagers believe every
oh-so-minor fix ought to be done upstream first. But I belong to those,
who disagree with that.

-- 
Michael Schwendt <mschwendt at users.sf.net>
Fedora Core release 3 (Heidelberg) - Linux 2.6.11-1.35_FC3
loadavg: 1.90 1.72 1.39




More information about the fedora-extras-list mailing list