-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Andy,
Thanks for your comments on my mail. >> I'm wondering what the state of the development and especially >> bug fixing of Facter is: > > For the state of puppet development we've been trying to keep the > tickets in redmine as up to date as possible. We are using the > target version field as a way of communicating what we are hoping > to get in a release, but it isn't a guarantee that it will actually > make it in. We have been getting a bit of work on Puppet done, but > Facter has been falling by the wayside. Yes, this is also a bit the impression that I have. >> Currently I have on my (very own) list 3 really nasty bugs, that >> make it really hard to rely on values of core facts (e.g.: >> virtual, is_virtual, ipaddress etc.) within manifest/module >> development, as currently they are broken and we had to rollout >> hot-fixes or give special guidelines to puppet users. >> >> All of these bugs have been reported months ago, patches have >> been proposed, but as how I - personally - see things nothing >> have been done to review the pull requests nor to fix the bugs. >> Although, these values are - in my opinion - very crucial for the >> usefulness of facter and hence puppet. >> > > This is an area that we've been really week in: keeping on top of > patches from people outside of Puppet Labs. We've been trying to > have a (rotating) person dedicated to responding to those, but I'll > admit that I've often had to pull them off in order to work on > other things. Without that person looking at pull requests they end > up falling by the wayside. This is really not a good situation, > because a huge about of what makes puppet work in all of these > environments is the contributions from others. > > So yeah...this is a big problem right now. I'm open to suggestions > about how we could work differently to fix it. Maybe giving out > commit access to more people outside of Puppet Labs? This is definitely not an easy problem to solve and I don't really have a good proposal to fix it. What I wanted to add, is that having Pull-Requests not reviewed/merged for a longer time period makes it also hard to contribute back. Because: Usually after months a person is way out of context of the specific bug that the person tried to fix months ago. So if there are minor things that needs to be discussed or improved in their patch, people need to a) find time to look at your comments in detail, but even more important b) find their way back into the bug's context. As it is the case for your dedicated person, that she/he is less productive if she/he keeps being pulled off the tickets to review, it is even more severe for a person which daily job is not to dig around in puppet's/facter's codebase. Also keep in mind that probably (my guess) most people outside of Puppet Labs who are contributing are not professional software developers, they are usually sysadmins that try to fix a problem that tempers them from working productively. It might have took them quite some effort to dig into the codebase and do that fix. I know that often it took also me some time to respond to comments/requests that were made to pull requests/reports/whatever that I did. But at the time I got feedback I was already quite far away from that certain problem, maybe even in a different project than I encountered the reported problem. For me, it is then usually quite some effort to get back into that context to provide a proper answer. And for such a bigger step I need to find time and maybe also an environment where I can reproduce it again. Which usually takes even more time, so the whole feedback-loop takes even more time, also from side. And hey, I internally rolled out a patch that kept me from being bugged by things. However, this might certainly be easier if I would have still been in that context/project/... And in my opinion it is even worse for first time contributer if their fixes/pull requests or even bug reports are laying around for months without a comment/review: Is Puppet Labs actually interested in my contribution? Was it worth to make it? Should I make it again the next time or just roll my own patched version? So personally I think what is missing, is that people get feedback when things are still fresh and present in their mind. This would probably reduce the roundtrip time within the feedback loops. Also I would not be sad, if people would just take the idea of a proposed pull request and implement a more proper solution that is then merged more instantly, than things are pushed back to me. I know that others might see that differently, that's also why I'm writing my personal opinion here. But I'm more interested in things getting fixed than having exactly my commits in the codebase. ~pete -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://www.enigmail.net/ iEYEARECAAYFAlB/rEgACgkQbwltcAfKi39jEgCgqlSnlYu4xQyfLQJvqhir06aC SOwAoK/GZ0ecQKW6QEfxlOOeQtcGt5pd =eB/l -----END PGP SIGNATURE----- -- You received this message because you are subscribed to the Google Groups "Puppet Developers" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/puppet-dev?hl=en.
