From mboxrd@z Thu Jan 1 00:00:00 1970 From: Brian J. Murrell Date: Wed, 24 Nov 2010 23:17:42 -0500 Subject: [Lustre-devel] Lustre tags In-Reply-To: <4CED8F11.8090606@llnl.gov> References: <4CED66BB.1010706@llnl.gov> <4CED8F11.8090606@llnl.gov> Message-ID: <1290658662.18668.151.camel@pc> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: lustre-devel@lists.lustre.org On Wed, 2010-11-24 at 14:17 -0800, Christopher J. Morrone wrote: > Christopher, You probably realize all of this but to ensure that it's absolutely clear... > I hadn't done a fetch from prime in a couple of weeks, and when I just > did it I saw that v1_8_5_RC4 had been tagged 1.8.5. Yes. This signifies that our release candidate 4 of the 1.8.5 series was what was released as the official 1.8.5. > In that case, you > can see that the change to lustre-version.ac to make it 1.8.5 actually > happened several commits before the actual 1.8.5 tag. That's right. We have to change the version (lustre-version.ac) at the start of the 1.8.5 release candidate testing phase so that when a candidate is chosen, it can be released as it was created and tested and not regenerated just for the purposes of final release. Put another way, the exact same packages that were generated for the RC and tested and passed our QA are then given to the users. There is no regeneration of packages after the chosen RC to produce "GA" packages. This is to prevent any kind of silliness from being introduced into the GA packages between the chosen RC and the GA release. > I have no way of > knowing how long v1_8_5_RC4 was in existence before it became 1.8.5. By "long" are you referring to linear time here? Are you trying to assess how much time elapsed while testing the final RC before it was declared GA? b. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 198 bytes Desc: This is a digitally signed message part URL: