Hide Forgot
Trying to build/install iwhd from source on RHEL v6.1 fails with the following when trying to run './autogen.sh' against the cloned git repo source. autoconf v2.65 isn't available from what I see in the official RHEL RPM repos nor the Aeolus Project repos. configure.ac:19: error: Autoconf version 2.65 or higher is required Just wanted to at a minimum document this dependency and possible get the ball moving on getting the RPM into the right repo, whether that is official RHEL or Aeolus-RHEL.
Hi Dave, Thanks for the report. However, this is not a bug in iwhd. We need to distinguish two types of failure to build from source (FTBFS). The one you may be thinking of is failure to build from a distribution tarball. That does deserve a bug report, and is currently a known problem with RHEL6.x that is due to the lack of a few dependencies. However, building from cloned sources (as you do when you run ./autogen.sh) typically requires cutting-edge tools that you might not even find in the latest version of Fedora. That said, it would be well worth it for ease of development on RHEL6.x to upgrade to autoconf-2.68. (currently it's at the lamentably-old 2.63-x from 2008-09-09)
Sounds like we'll need to add "autoconf 2.65+" to the build requirements for people wanting to set up a dev environment. As an alternative thought, would it be feasible to adjust iwhd to work with the older autoconf that's in RHEL 6.1?
I've just filed http://bugzilla.redhat.com/699622 explaining why RHEL6.x should upgrade to autoconf-2.68. In the mean time, people should be able to get away with relaxing the requirement: perl -pi -e s/2\.65/2.63/ configure.ac
Cool. Should we update configure.ac in iwhd's git repo as well?
I've fixed it with this commit: http://git.fedorahosted.org/git?p=iwhd.git;a=commit;h=4fa67c27607643e0fa45c4a2139965972443c02b
Jim, The correct step after committing your change is to move the bug to modified or on_qa. Thanks!
Wes, I nearly closed as NOTABUG, but decided that was not called for. From your reaction, perhaps it was. The correct step is not to file a bug for fails-to-build-from-clone in the first place, as explained above. IMHO, this change should not involve QA.
Hey Jim, not arguing w/ your point there, a bug should have been opened on RHEL not this component. I only ask that development never close bugs, moving them to on_qa is enough to take it off your plate. If you want/need further explanation lets chat outside of bugzilla. Thanks
Dave, please confirm that the work around Jim noted works.. ***************** In the mean time, people should be able to get away with relaxing the requirement: perl -pi -e s/2\.65/2.63/ configure.ac *****************
I think I was being a pita on this one... Jim opened a bug that should cover the issue. http://bugzilla.redhat.com/699622 closing as not a bug, w/ qe ack
perm close