Re: What happened to pinned bootstrap

2014-04-21 Thread Akash Chandrashekar
I would add to this, that there are equivalent changes in MAAS (maas-cli example), and its interaction. With the changes, its also from a user perspective at least- that we maintain some level of documentation for installations, in which old docs that are our reference architectures, or install

Re: What happened to pinned bootstrap

2014-04-18 Thread William Reade
I remain strongly +1 on bootstrapping with tools that exactly match the client version -- the client code near-enough directly drives those tools as we bootstrap, and if we force pinned versions at bootstrap time we're free to change how bootstrap works. If we need to deal with variation in the

Re: What happened to pinned bootstrap

2014-04-18 Thread Aaron Bentley
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14-04-18 06:28 AM, William Reade wrote: As for automatically upgrading: it's clearly apparent that there's a compelling case for not *always* doing so. But the bulk of patch releases *will* be server-side bug fixes, and it's not great if we

Re: What happened to pinned bootstrap

2014-04-18 Thread Kapil Thangavelu
On Fri, Apr 18, 2014 at 11:34 AM, Aaron Bentley aaron.bent...@canonical.com wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14-04-18 06:28 AM, William Reade wrote: As for automatically upgrading: it's clearly apparent that there's a compelling case for not *always* doing so. But

Re: What happened to pinned bootstrap

2014-04-17 Thread John Meinel
FWIW that isn't what you get if you do apt-get install foo you always get the latest version of foo that has been patched, not the original one that came with Precise. I don't have a problem with juju bootstrap growing a --version target (same as apt-get install foo=1.2.3). But I think our policy

Re: What happened to pinned bootstrap

2014-04-17 Thread Aaron Bentley
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 14-04-17 12:03 PM, John Meinel wrote: If you bootstrap, you are installing juju onto the remotel machine. The reason we created a *patched* version is to give you improvements (bug fixes, security fixes, etc). Sure, but if the user doesn't

Re: What happened to pinned bootstrap

2014-03-31 Thread John Meinel
On Mon, Mar 31, 2014 at 5:16 AM, Andrew Wilkins andrew.wilk...@canonical.com wrote: On Mon, Mar 31, 2014 at 7:17 AM, Ian Booth ian.bo...@canonical.comwrote: On 31/03/14 02:11, Kapil Thangavelu wrote: sounds like a great case being made for --upload-tools by default. --upload-tools does

Re: What happened to pinned bootstrap

2014-03-30 Thread Kapil Thangavelu
sounds like a great case being made for --upload-tools by default. On Sun, Mar 30, 2014 at 12:23 AM, John Meinel j...@arbash-meinel.comwrote: I thought at one point we were explicitly requiring that we bootstrap exact versions of tools (so juju CLI 1.17.2 would only bootstrap a 1.17.2 set

Re: What happened to pinned bootstrap

2014-03-30 Thread Jeff Pihach
This sounds like a good idea. The bug which I ran into (which I'm assuming is the source behind this query) would be very difficult to debug by users who didn't realize that there was a new juju release and they have not yet updated the client. -Jeff Kapil Thangavelu wrote: sounds like a great

Re: What happened to pinned bootstrap

2014-03-30 Thread Kapil Thangavelu
On Sun, Mar 30, 2014 at 4:17 PM, Ian Booth ian.bo...@canonical.com wrote: On 31/03/14 02:11, Kapil Thangavelu wrote: sounds like a great case being made for --upload-tools by default. --upload-tools does happen automatically on bootstrap, but only if no matching, pre-built tools are