Platform Team Week of July 21st, 2014

35 views
Skip to first unread message

Andy Parker

unread,
Jul 28, 2014, 12:29:47 PM7/28/14
to puppe...@googlegroups.com

** Next PR Triage Wednesday, July 30th @ 10:00 am Pacific. **

Priorities

  1. Puppet 3.7.0/4.0.0 with the final language polish as well as many other changes.
  2. Windows stability improvements/x64 support
  3. CFacter on the march

Commentary

The language work is nearing the end. The past week saw a large number of decisions finally reached on how resource expressions are going to work in Puppet 4. One of the notable changes was that resource defaults will *not* be changed to the same scoping rules as variables. See the thread on puppet-dev if you want to know the reasoning.

Windows 64-bit work hit some snags on win2003. I believe that there is some doubt about whether we can get it to work. Win2008r2 and newer is going along just fine, however. I'm sure that Rob, Josh, or Ethan can provide more details if anyone wants them.

Work on cfacter on windows has begun! I believe that this week also saw that client team reach the decision to concentrate on getting cfacter ready instead of putting work into ongoing development of the ruby facter codebase. What this means is that facter 2.2 is likely that last release that you'll be seeing of that line of code.
Data


3.7 Burn up. Our target date is August 27th

Things are pretty close. We should be fine to do the release candidate on the 27th, but we can't slip.
Inline image 1
--
Andrew Parker
Freenode: zaphod42
Twitter: @aparker42
Software Developer

Join us at PuppetConf 2014September 22-24 in San Francisco
Register by May 30th to take advantage of the Early Adopter discount save $349!

Rob Reynolds

unread,
Jul 28, 2014, 2:04:07 PM7/28/14
to puppe...@googlegroups.com
On Mon, Jul 28, 2014 at 11:29 AM, Andy Parker <an...@puppetlabs.com> wrote:

** Next PR Triage Wednesday, July 30th @ 10:00 am Pacific. **

Priorities

  1. Puppet 3.7.0/4.0.0 with the final language polish as well as many other changes.
  2. Windows stability improvements/x64 support
  3. CFacter on the march

Commentary

The language work is nearing the end. The past week saw a large number of decisions finally reached on how resource expressions are going to work in Puppet 4. One of the notable changes was that resource defaults will *not* be changed to the same scoping rules as variables. See the thread on puppet-dev if you want to know the reasoning.

Windows 64-bit work hit some snags on win2003. I believe that there is some doubt about whether we can get it to work. Win2008r2 and newer is going along just fine, however. I'm sure that Rob, Josh, or Ethan can provide more details if anyone wants them.

I will just add some details here for those that may not ask. If we take a look at RubyInstaller's Issue #211[1], you will see that the installer is linked to the system msvcrt.dll, where a couple of dependencies have been added to things that do not exist on Windows 2003's (possibly R2 as well) system msvcrt.dll.

We are tracking the work for this issue at PUP-2964[2].

There are really three options for us:

1) Do nothing. Do not allow install of Puppet x64 on Windows 2003. There is only a year of support left (unless it is extended). Not sure how everyone would feel about this option.

2) Attempt to link to a VC++ Runtime Redistributable package[3]. This is undesirable as it creates a dependency on having that runtime installed on boxes.

3) Backport a patch from Ruby 2.1.x that makes the check on whether these functions are available. See Ruby #9119[4] and ruby commit 58a7cda[5]. If this is all we need to do, this becomes the most desirable as it would allow for running Puppet 64 bit on Windows Server 2003. However it is possible as we get into this that we could find that there is more work to do and then move to a preference of Option 1.

 

Work on cfacter on windows has begun! I believe that this week also saw that client team reach the decision to concentrate on getting cfacter ready instead of putting work into ongoing development of the ruby facter codebase. What this means is that facter 2.2 is likely that last release that you'll be seeing of that line of code.
Data


3.7 Burn up. Our target date is August 27th

Things are pretty close. We should be fine to do the release candidate on the 27th, but we can't slip.
Inline image 1
--
Andrew Parker
Freenode: zaphod42
Twitter: @aparker42
Software Developer

Join us at PuppetConf 2014September 22-24 in San Francisco
Register by May 30th to take advantage of the Early Adopter discount save $349!

--
You received this message because you are subscribed to the Google Groups "Puppet Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to puppet-dev+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/puppet-dev/CANhgQXuhTrN57-%2BsD32Z2w%3DVcLbsb%3D%2BYu9b0%2BsUHKkQE7trroA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.



--
Rob Reynolds
Developer, Puppet Labs

Join us at PuppetConf 2014September 20-24 in San Francisco
Register by July 31st to take advantage of the Early Bird discount save $249!

Paul Stack

unread,
Jul 31, 2014, 5:50:28 PM7/31/14
to puppe...@googlegroups.com
In my opinion, I would not worry too much about Windows 2003 support for x64. The effort to get this working will far outweigh the benefit considering, as you said, there is less than 1 year's support left

As a company that has some 2003 core infrastructure, I'd rather see it die than to continue supporting it

$0.02
Reply all
Reply to author
Forward
0 new messages