16:00:11 #startmeeting Network Meeting 16:00:11 Meeting started Wed Aug 10 16:00:11 2016 UTC. The chair is gundalow. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:11 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:11 The meeting name has been set to 'network_meeting' 16:00:18 Hi! 16:00:25 #chair Qalthos 16:00:25 Current chairs: Qalthos gundalow 16:00:33 Hi stianvi 16:01:05 #info Agenda, as always is https://github.com/ansible/community/issues/110 Feel free to add things 16:01:13 #chair Qalthos privateip 16:01:13 Current chairs: Qalthos gundalow privateip 16:02:48 #info I've reviewed the items that are alread on the agenda. They are two sets 1) Ansible Core need to do final review and merge 2) Creator has updates to make. There doesn't seem to be anything extra that needs reviewing 16:03:01 stianvi: Did you have anything 16:03:01 * gregdek lurks 16:03:21 gregdek: :) 16:03:22 just the py24 thingy :) #16434 16:03:35 it states on the website: Currently Ansible can be run from any machine with Python 2.6 or 2.7 installed 16:03:46 can't see that network modules should run anywhere but the control machine 16:03:49 #topic Python24 support for Modules 16:03:55 privateip: you around? 16:04:00 yep 16:04:04 never got an answere why :) 16:05:08 privateip: with Networking modules can you/do people always run on the machine running ansible-playbook, or could they sometimes be delegated to a ssh jump box, or someother machine that might only have py2.4 on? 16:05:14 caphrim007: hi :) 16:05:19 i was late 16:05:21 i blame the bus 16:05:47 correct the cli transport is built to take advantage of delegate_to 16:06:06 whats the question? 16:06:31 why does the network modules need to work with python2.4 if ansible requires python2.6? 16:06:36 privateip: https://github.com/ansible/ansible/pull/16434 & https://github.com/ansible/ansible-modules-core/pull/4032 16:06:46 ansible supports py2.4 today 16:07:07 that is going away (not soon enough) 16:07:19 http://docs.ansible.com/ansible/intro_installation.html 16:07:19 > Currently Ansible can be run from any machine with Python 2.6 or 2.7 installed (Windows isn’t supported for the control machine). 16:07:38 LOL. No, not soon enough. 16:08:06 thats the control machine req 16:08:17 yes :) 16:08:18 stianvi: "can be run from" = The machine that you are executing "ansible" or "ansible-playbook" on 16:08:34 is there a problem supporting py24 in your PR? 16:08:42 machine with Ansible -> network switch 16:08:55 machine with Ansible -> Some other machine (via deligate_to) -> network switch 16:09:05 It is possible to note in the module that py24 isn't supported by that module. 16:09:12 ok 16:09:24 In the 2nd case the "other machine" may only have py2.4 on 16:09:35 got the design now, thanks gundalow. 16:09:47 :) 16:09:58 From the module checklist: "Modules must be written to support Python 2.4. If this is not possible, required minimum python version and rationale should be explained in the requirements section in DOCUMENTATION." 16:10:09 http://docs.ansible.com/ansible/developing_modules.html 16:10:38 I can always fork out and run my version for now. I'll see if I can rewrite a bit, but juniper seem to require with statement that was introduced in py25 for the feature 16:11:01 pyez? 16:11:47 not whole pyez, but the configure mode part of it 16:12:05 im looking at 4032 and dont see any issue with it support py2.4 as is 16:12:09 did i miss something? 16:13:27 https://github.com/ansible/ansible/pull/16434 16:14:13 https://github.com/Juniper/py-junos-eznc/blob/master/lib/jnpr/junos/utils/config.py 16:14:29 oh this is in the shared module ... got it now 16:15:02 https://github.com/Juniper/py-junos-eznc/blob/master/lib/jnpr/junos/utils/config.py#L622 16:15:19 according to doc in pyez, requires with. 16:15:33 but I might be wrong 16:15:39 my python is pretty fresh 16:18:20 ok to avoid turning this into a mountain ... lets caveat it in the junos_config module 16:19:08 fair enough 16:19:14 this only affects one argument 16:19:53 please update the docstring as gregdek pointed out ^^^ 16:20:36 I think abadger1999 actually posted the 2.4 deprecation timeline the other day, if that's helpful. 16:20:44 * gregdek rummages about. 16:20:54 #action stianvi to update docstring in requirements 16:20:55 * privateip celebrates (never to early to start) 16:21:19 The release after march 2017 will no longer have python-2.4 compat 16:21:20 March 2017. 16:21:35 want me to rewrite 4032 and close #16434 16:21:48 correct? 16:21:50 releases before March 2017 will continue to support py2.4 as long as we keep making maintainance releases for them. 16:23:42 stianvi: just 4032 will be fine 16:23:50 ack 16:24:04 thanks guys 16:24:24 stianvi: Thanks for your time to work though the issues 16:24:39 caphrim007: Did you have anything? 16:24:55 gundalow: yes, this module should be ready now https://github.com/ansible/ansible-modules-extras/pull/998 16:25:30 my other prs should just require an updating of the doc fragment and then they should also be ready 16:25:42 caphrim007: Can you please put "ready_for_review" on that to kick the bot to move it on 16:25:47 ok 16:25:48 #topic F5 modules 16:26:12 Thanks 16:26:47 caphrim007: "At least one of value and state C(reset) are required" 16:26:48 I might be missing it (it might be inside f5.bigip, which I don't have), however I don't see a test to ensure at least of of those parameters are passed 16:26:52 Did that get resolved? 16:27:08 lemme check 16:27:32 You've squashed the commits (which is fine), but it means I can't quickly check :) 16:32:53 gundalow: there is now a check for it and an updated pr 16:33:17 * gregdek ducks out 16:33:23 cool, I'll review then I'll add it to the merge pile 16:33:29 thx 16:33:43 Thanks for the modules :) 16:34:47 #topic Open Floor 16:34:58 privateip: Qalthos Anything else from you? 16:35:21 just a quick update on 2.2 refactor ... things are progressing along 16:35:42 i'm hoping to start pushing module updates within a week or so 16:36:02 to fix a lot of the nagging issues brought up as a result of the refactoring activity 16:36:24 privateip: is there anything that module authors should be aware of to fix in their own modules? 16:36:50 not yet 16:37:05 this is mostly reorganizing code to take advantage of ansiballz 16:37:48 i missed the memo on that...is that the ziploader? or something else? 16:38:03 ansiballz = ziploader 16:38:04 s/ziploaders/ansiballz 16:38:10 hehe. ok 16:39:40 Anything else? 16:40:34 Any events we should be getting to? 16:41:16 Shameless plug, Ansible Fest brooklyn in 11th October 16:41:21 https://www.ansible.com/ansiblefest 16:45:51 I guess we are done then 16:45:53 #endmeeting