On Tue, Jul 09, 2013 at 06:01:08PM +0000, Edison Su wrote: > > > > -----Original Message----- > > From: Chip Childers [mailto:chip.child...@sungard.com] > > Sent: Tuesday, July 09, 2013 11:00 AM > > To: Edison Su > > Cc: dev@cloudstack.apache.org > > Subject: Re: Swift in 4.2 is broken, anybody wants it to be supported in > > 4.2? > > > > On Tue, Jul 09, 2013 at 05:56:24PM +0000, Edison Su wrote: > > > > > > > > > > -----Original Message----- > > > > From: John Burwell [mailto:jburw...@basho.com] > > > > Sent: Tuesday, July 09, 2013 7:53 AM > > > > To: dev@cloudstack.apache.org > > > > Cc: 'Chip Childers' > > > > Subject: Re: Swift in 4.2 is broken, anybody wants it to be supported in > > 4.2? > > > > > > > > Edison, > > > > > > > > As I read through this thread, we seem to be conflating the following > > topics: > > > > > > > > 1. Feature regression testing per release cycle > > > > 2. Identifying and back porting defect fixes to previous > > > > releases > > > > 3. Feature removal process > > > > > > > > To my mind, these topics are completely unrelated. We have > > > > regression test and defect triage processes to address items 1 and > > > > 2. If you feel that they can be improved, then we should discuss > > > > those improvements in a separate thread. No community or system > > > > will be perfect. I believe the best we can do is seek to do it > > > > better today than yesterday. To that end, observing that we did > > > > something poorly in the past does not justify continuing to do it > > > > poorly or > > removing a feature on which users are relying. > > > > > > > > > > > > I am concerned about item 3 -- the merge of a feature removal > > > > without community consensus. If you *think* a feature is broken in > > > > a previous > > > > > > This feature is not been tested since about one and half year ago, nobody > > knows the status of swift integration. > > > If we can't claim to support Swift in 4.0, 4.1, then why you think I am > > removing a feature? > > > > But we *do* claim that support. See [1]. > > > > Not having tested it is *not* the same as saying that it isn't supported. > > > > [1] http://cloudstack.apache.org/docs/en- > > US/Apache_CloudStack/4.1.0/html/Installation_Guide/about-secondary- > > storage.html > > IMHO, saying something is supported without tested for each release is worse > than saying not supported. > >
Fine, so we screwed up as a community. I guess it needs to be tested for all feature releases. That has nothing to do with the issue of simply "dropping" support of what might very well have been a functional feature. Additionally, I know of at least one major user of older cloudstack versions that *DO* use swift for secondary storage. Are you suggesting that we strand another group of users *on purpose* again? Here's my issue... in the rush to make changes to the architecture and / or to get a new feature in, we have now run into the situation where we have asked ourselves to simply drop a function *after the fact*. This is unhealthy for the project, crappy for our users, and a sloppy way to evolve a software system. I'm not blaming you for this at all, but the object-storage architectural changes are an example of this behaviour. We need to stop this habit. -chip
