> Le 12 oct. 2015 à 19:12, Michael Stahnke <[email protected]> a écrit : > > > > On Mon, Oct 12, 2015 at 9:51 AM, Fabrice Bacchella > <[email protected]> wrote: > > > Le 12 oct. 2015 à 18:01, Michael Stahnke <[email protected]> a écrit : > > > With transparency comes the ability to see things that are not perfect. We > > had a large conference last week where a very large portion of our employee > > base was in attendance. We shipped the bits because it fixed a bug that > > some people had run into. We haven't yet sent out release notes, or > > completed Jira hygiene. While the order of operations there is not perfect, > > I hardly think it's worth making sweeping generalizations about the product > > quality. > > Twice the same mistake in two weeks ? > > This wasn't a mistake. It was a choice to ship and do the paperwork later. > > pgp key conflict in RHEL 6, but not 5 and 7 ? And it tooks too long to > correct such a simple defect (https://tickets.puppetlabs.com/browse/PUP-4970) > > Originally it wasn't considered a defect. This was a design decision > originally that was changed. Originally we didn't want products and PC1 > release to be co-installable. After some feedback and use-cases it was > changed. > > cfacter barely tested with not so old RHEL 5 ? > (https://tickets.puppetlabs.com/browse/FACT-1152 and > https://tickets.puppetlabs.com/browse/FACT-1137) > > I can't comment on this because I don't know all the details. (I would say > that a 9 year old OS is old though ;) ) > > They are all minor, but the number of them I found without even looking in a > short time frame (about two month) are annoying. > > All software has bugs. Our software is a subset of all software, so ours does > as well. We strive for quality and we run many thousands of tests and suites > against or projects. You are welcome to submit fixes for items you see as > incorrect. You are encouraged to file tickets. We can't prioritize every > ticket all the time. Sometimes they will sit for a bit. A patch from a user > can speed up that process sometimes. > > Complaining about bugs/tickets that have recently been addressed/fixed is > even less helpful, as they are done. I ask that you please try to make your > interactions more deliberate and drive action rather than simply state where > things don't work for you. The community can not easily take action or > improve simply on complaints. >
I really like puppet, I think it's a good product. PC1 was a good thing, don't depend on stock puppet is a really good move. Puppet 4 is a interesting realease with good improvement in the language. And all products have bug, I know that too much. I'm not complaining about hard or difficult to reproduce bugs. I have a bug with an HP Smart Array, I see this one as a "normal" bug, I don't except puppetlabs to have every kind of hardware to test every combination of OS release. I just report it and cursed my bad luck. But they always has been minor glitch in puppet, too many. And the problem is not in submitting a patch or finding the bug in the code. It's about the small inconsistently one can found everywhere in puppet. Look at the problem about PGP. We could argue hours if the conflict was a good or a bad thing. I really don't have opinion about that. But why no conflict on RHEL5/7 and one on RHEL6 ? That the real problem for me. And that's something someone outside puppetlabs can't help to solve. It's more about internal procedure and how decision are taken. So I complain about that, hopping that puppetlabs will start to think about this problem. Your comment about a "old enough" OS is a really bad answer. Oracle, JBoss can stop to support old OS release., customers will follow them because they are top level application and we adapt our servers to fit their needs. Puppet is the opposite. It's very purpose is to run on as much operating system as it's possible, as long as there is "enough" customers that uses it. We use puppet to ease management of old, strange, heterogeneous environment. I dislike it as much as you and would like to kill my last few instance of RHEL5, but that's not a decision I can take, I and hence you to must live with that and keep support it. -- You received this message because you are subscribed to the Google Groups "Puppet Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/puppet-users/E86BB388-9E67-484D-9339-181DE3B44D92%40orange.fr. For more options, visit https://groups.google.com/d/optout.
