Here's the summary of today's IRC meeting.



Place: #openvpn-meeting on irc.freenode.net
Date: Wednesday 21st Dec 2016
Time: 20:00 CET (19:00 UTC)

Planned meeting topics for this meeting were here:


The next meeting has not been scheduled yet.

Your local meeting time is easy to check from services such as



cron2, dazo, mattock and syzzer participated in this meeting.


Discussed decommissioning openvpn-testing.git repository. Agreed that it makes sense.


Discussed updating copyrights in files to 2017. Agreed that it makes sense. Dazo will push out a review branch.


Discussed the OpenVPN 2.4.0 release. Agreed to make the release slightly ahead of schedule on 27th Dec.


Discussed the "reformatting: fix style in crypto*.{c, h}" patch:


Agreed to merge this into Git "master" now, but not into "release/2.4" until 2.4.0 is out.


Discussed scheduling the next meeting. It was agreed that we probably don't need an official meeting next week (28th). If there are issues with the 2.4.0 release we can discuss and fix things informally on #openvpn-devel channel.


Full chatlog has been attached to this email.

Samuli Seppänen
Community Manager
OpenVPN Technologies, Inc

irc freenode net: mattock

(20:57:29) syzzer: evening :)
(20:57:52) cron2: not yet
(20:58:45) syzzer: oh, sorry
(20:58:52) mattock: yes, way too yearly :P
(20:58:57) mattock: we have to wait ~60 seconds
(21:00:21) syzzer: and now, and now, and now/
(21:00:38) mattock: https://community.openvpn.net/openvpn/wiki/Topics-2016-12-21
(21:00:39) vpnHelper: Title: Topics-2016-12-21 – OpenVPN Community (at 
(21:00:41) syzzer: \o/
(21:01:14) cron2: evening!
(21:01:26) mattock: ready?
(21:01:38) mattock: is dazo here?
(21:01:51) syzzer: looking at the topics, seems we need dazo
(21:02:03) cron2: awwww
(21:02:07) cron2: why is the agenda so long
(21:02:20) mattock: yes, too much to read
(21:02:41) syzzer: at least the 2.4 status page says 'All done!'
(21:02:53) cron2: ok, I've read the agenda, and I say "make it so!".
(21:03:00) cron2: are we done?
(21:03:27) cron2: :)
(21:03:45) mattock: I agree with everything
(21:03:58) mattock: so 2.4.0 release in original schedule?
(21:04:04) cron2: (why do we have release/2.3 in testing, but no other 
release/2.x branch...?)
(21:04:25) syzzer: yeah, looks good to me too
(21:04:39) cron2: mattock: yes.  m-a brought up something strange in 
#openvpn-devel regarding some freebsd-base firewall, but that can always be 
platform-fixed later on
(21:04:56) mattock: so 2.4.0 is due on 28th
(21:05:00) mattock: which is fine for me
(21:05:04) cron2: I'm running _rc1 on our corp servers for two weeks now, with 
poor man's NCP, on FreeBSD, and that all behaves
(21:05:50) ***dazo is here 
(21:05:52) dazo: just forgot time
(21:05:56) cron2: so, anectotal evidence says "this is a good release" :)
(21:06:27) mattock: there are a few openvpn-gui, openvpnserv2 and openvpn-build 
improvements in the queue which can probably make it to 2.4.0 installers
(21:06:56) mattock: but if they don't, they can probably go to 2.4.1
(21:07:03) mattock: nothing really major for the most part
(21:07:06) cron2: there's endless opportunities to fix and improve things :-) - 
but I'm not aware on anything seriously pressing
(21:07:13) mattock: yeah, same here
(21:07:33) dazo: If nobody is against it, based on the lack of really much 
noise ... I propose we doe the final release 1-2 days earlier
(21:07:53) syzzer: as long as the interactive service is installed in a nice 
way, because that's the new feature we want people to use on windows
(21:07:54) dazo: I think we should just add a few docs updates + the copyright 
updates and then call it 2.4.0
(21:07:58) mattock: 26th is a national holiday here, so 27th would work
(21:08:43) cron2: 26th is "christmas with grandparents", so I won't get 
anything useful done.  But then, with the current state of things (which is 
really good) you do not need me anyway :)
(21:08:46) ***cron2 retires
(21:08:59) syzzer: what about my crypto*.{c,h} patch?  I'd like that in 
release/2.4 too, but *after* 2.4.0 is fine too 
(21:09:12) dazo: okay, I beleive I can have everything prepared for the 26th, 
including branch push ... so you can test on 27th and I'll do the tags push 
once you confirm it's good
(21:09:27) cron2: syzzer: that's the refactoring Selva ACKed?
(21:09:32) syzzer: indeed
(21:09:35) dazo: syzzer: I might have forgotten about that
(21:09:39) syzzer: more reformatting that refactoring
(21:09:45) cron2: uh, right
(21:10:20) cron2: dazo: 
(21:10:22) vpnHelper: Title: [Openvpn-devel] [PATCH v2] reformatting: fix style 
in crypto*.{c, h} (at www.mail-archive.com)
(21:10:42) dazo: do we strictly need it for 2.4.0?
(21:10:43) dazo:  5 files changed, 356 insertions(+), 298 deletions(-)
(21:11:11) dazo: that's a fairly big one ... and I'd like to avoid a late 
regression in 2.4.0 ... I'm ready for 2.4.1 and master
(21:11:28) cron2: do a "git diff -w"
(21:13:05) dazo: it's still fairly bit
(21:13:08) dazo: big*
(21:13:09) dazo:  5 files changed, 274 insertions(+), 216 deletions(-)
(21:13:50) syzzer: yeah, because it wraps lines
(21:14:05) dazo: even though it seems safe ... it just takes a mix up with an 
argument or two ... not saying it is so, but too little time to really test it
(21:14:12) syzzer: but I'm fine with after 2.4.0 :)
(21:14:32) syzzer: would be nice to get it into master though, so I can build 
onto it
(21:14:33) dazo: That will give me peace for the Christmas holiday at leat :)
(21:14:47) dazo: yeah, we can put it into master
(21:15:12) dazo: syzzer: have you done any manual changes which uncrustify 
would undo?
(21:15:32) syzzer: not sure
(21:15:51) dazo: I have quite successfully re-run uncrustify a few times when 
testing some patches locally ... and it did not change anything it shouldn't
(21:18:33) mattock: so the "fix style in crypto*" will go to "master" but not 
to release/2.4?
(21:19:01) dazo: not yet ... not before 2.4.0 is out the door
(21:19:07) syzzer: dazo: reformat-all.sh doesn't work anymore since the great 
reformatting: error: patch failed: include/openvpn-plugin.h.in:169
(21:19:13) cron2: it should end up in release/2.4, but I'm fine with "after 
2.4.0" - otherwise working on these code bits and backporting is too annyoing
(21:19:31) syzzer: ^^ that
(21:19:45) mattock: any other topics?
(21:19:52) dazo: syzzer: right, I ran it manually ... I want to provide a 
simpler script than reformat-all.sh though
(21:20:04) mattock: this is our possibility to have the shortest meeting ever 
(not counting cancelled ones)
(21:20:06) mattock: :)
(21:20:14) cron2: we didn't really discuss openvpn-testing yet :-) - but as I 
said "makes sense"
(21:20:22) mattock: +1
(21:20:27) dazo: mattock: just to wrap it up ... run update-copyright.sh 2017 
.... wipe openvpn-testing.git is all okay=?
(21:20:39) mattock: yes, that was the consensus before you joined
(21:20:43) mattock: and still is
(21:20:48) dazo: good! :)
(21:20:51) syzzer: yes
(21:20:52) syzzer: go
(21:21:25) dazo: then I'll prepare as much as I can for the 2.4.0 release ... 
so the process will go quickly the 27th
(21:21:31) mattock: great!
(21:22:11) dazo: Currently, this is all we have ....
(21:22:14) dazo: $ git shortlog v2.4_rc2..HEAD
(21:22:14) dazo: David Sommerseth (1):
(21:22:14) dazo:       dev-tools: Added script for updating copyright years in 
(21:22:29) mattock: well we get a few others
(21:22:36) mattock: so the changelog won't look too silly
(21:22:53) dazo: yeah, but not many at all 1-3 more commits, that's all
(21:23:33) mattock: fine by me
(21:23:36) syzzer: which is good
(21:23:38) dazo: syzzer: did you have a chance to look at the Changes.rst + 
openvpn.8 change I proposed on #openvpn-devel?
(21:23:52) syzzer: yeah, see my reply in #openvpn-devel ;)
(21:24:00) dazo: ahh :)
(21:24:04) syzzer: 19:57 <syzzer> dazo: man fix looks good to me :)
(21:24:08) dazo: thx!
(21:24:14) dazo: I'll submit the patch the ML then
(21:24:55) dazo: (2.3 and 2.4/master needs separate ones, but the contents 
should be fairly the same)
(21:25:15) mattock: anything else?
(21:26:02) syzzer: not from me
(21:26:16) dazo: have you heard anything more from OSTIF, mattock?  regarding a 
time line for the review?
(21:26:44) mattock: dazo: no news
(21:26:57) dazo: okay, just curious :)
(21:26:58) mattock: I guess they're ironing out the details
(21:27:09) mattock: with the company doing the audit
(21:27:40) dazo: seems syzzer was quite spot on earlier though ... matt green 
will do the crypto audit and QuarkLabs will do the exploit and generic software 
(21:28:46) mattock: yep
(21:29:41) mattock: are we done? I have my mouse cursor on Send button in 
Thunderbird :P
(21:29:54) ***cron2 nudges mattock to hit the mouse button
(21:29:59) dazo: hit it!
(21:30:01) mattock: great!
(21:30:10) mattock: oh, one detail
(21:30:12) mattock: next meeting?
(21:30:22) mattock: or "not scheduled yet"?
(21:30:47) dazo: I don't think we need anything before the 2.4.0 release ... 
but we could schedule one in January
(21:30:49) cron2: well, technically, Dec 28 would be the next one - but I see 
that one as a quick one
(21:30:58) cron2: "if anything came up"
(21:31:07) mattock: yeah, let's not schedule anything quite yet
(21:31:12) mattock: if we need one on 28th, let's have one
(21:31:40) mattock: agreed?
(21:31:50) dazo: well, think we should decide to either have or not have a 
meeting ... so others reading this scrollback (or on mail) can come if they 
decide to
(21:32:05) dazo: otherwise, we just solve it on #openvpn-devel on-the-fly the 
(21:32:14) mattock: on the fly is probably good enough
(21:32:22) cron2: right
(21:32:23) mattock: I doubt there are horrible issues with 2.4.0
(21:32:25) dazo: yes, I think so
(21:32:39) mattock: ok
(21:32:41) dazo: if there is ... it will impact the man page most of all :-P
(21:32:56) dazo: (provided rc2 is rock solid)
(21:33:45) dazo: oh ... regarding the copyright update ... shall we do some 
audit on that, or should I just commit it directly?
(21:34:03) dazo: I don't think we need a massive review on that one, though ... 
the script works
(21:34:04) mattock: sending it to the list would be good
(21:34:11) dazo: seriously?
(21:34:21) mattock: follow the process :)
(21:34:31) dazo: it changes just comment lines of existing lines
(21:34:47) mattock: I assume you verified all the changes?
(21:34:51) syzzer: yes, this.  It's easy enough to ACK, so just send it to the 
(21:35:02) syzzer: the overhead is minimal
(21:35:07) mattock: I just think it's quite easy miss obvious issues sometimes
(21:35:43) dazo: the patch is ~110KiB
(21:35:46) mattock: uh :P
(21:35:48) syzzer: (which is the reason I just sent the version bump to the 
list, instead of bugging dazo or cron2 in the channel to do it)
(21:36:00) dazo:  211 files changed, 243 insertions(+), 243 deletions(-)
(21:36:14) mattock: is it in a Git tree somewherE?
(21:36:32) dazo: ./dev-tools/update-copyright.sh 2017
(21:36:51) dazo: git version bump is not applied yet, on my list
(21:37:14) mattock: I can run that tomorrow and have a quick look at the results
(21:37:48) dazo: okay, lets do what we did with the reformatting stuff ... I'll 
push out a review-branch ... and ack the commitish instead
(21:37:55) mattock: ok, that will also work
(21:38:19) syzzer: really?  110k isn't that bad I'd but, but whatever you 
prefer :)
(21:39:29) mattock: ok, now I will start moving my cursor towards send...
(21:39:45) dazo: when all the changes are just something like this ... I just 
find to be a useless overhead
(21:39:46) dazo: - *  Copyright (C) 2002-2010 OpenVPN Technologies, Inc. 
(21:39:47) dazo: + *  Copyright (C) 2002-2017 OpenVPN Technologies, Inc. 
(21:40:49) cron2: do we know that all changes are like this without having seen 
the actual change?
(21:40:54) mattock: +1
(21:42:08) dazo: well, you'll see it in the git log afterwards .... and if I'm 
a jerk changing anything else than comments related to copyrights  ... I'd 
probably loose all the credibility here
(21:42:26) dazo: it's not that it would be easy to detect
(21:42:29) mattock: I'm not worried about, but the script
(21:42:31) syzzer: cron2: there's a script to reproduce the result, like with 
the reformatting
(21:42:33) mattock: about you
(21:42:49) syzzer: the script is good, I ACK'ed it ;)
(21:43:22) mattock: well, if you trust the script 100% then I'm fine with 
direct push
(21:43:36) mattock: the script-ACK could be considered an ACK for the patch 
(21:43:38) cron2: I'm fine with a mail, or with a review branch, or with 
trusting the script (and verifying later)...
(21:44:09) mattock: let's let dazo do as he wishes then :)
(21:44:45) dazo: Oh .... I'd like to change my users.sourceforge.net address to 
my own address, as the sf.net address have become useless due to bad spam 
handling on sf.net ... but that's strictly "changing comments related to 
copyrights" ... which I already carry ... but can do that separately if you want
(21:45:09) dazo: the script is already ready to handle my new address
(21:46:03) mattock: I have no strong opinions one way or the other
(21:46:26) syzzer: tbh, I really don't care.  Just don't see the advantage of 
not just sending patch to the list :)
(21:46:43) syzzer: so let's see what dazo decides
(21:46:48) mattock: yep
(21:46:53) dazo: okay, you'll get the ML approach
(21:47:09) mattock: sounds good
(21:47:20) dazo: one reason to avoid the ML is actually spam crawlers 
harvesting addresses
(21:47:25) dazo: on this context
(21:47:41) dazo: *in
(21:47:56) mattock: you choose
(21:48:04) dazo: (so with ML, I will not do the mail address change)
(21:49:08) syzzer: as you please :)
(21:49:11) mattock: +1
(21:50:08) mattock: ok, maybe we're now done? :P
(21:50:12) dazo: yes :)
(21:50:21) mattock: ok, good night guys!
(21:50:28) syzzer: almost managed to stretch the meeting to an hour :p
(21:50:30) syzzer: good night!
(21:50:35) mattock: yeah, lol :)
(21:50:40) mattock: we we're basically ready at 21:20
(21:50:44) mattock: 30 minutes ago
Developer Access Program for Intel Xeon Phi Processors
Access to Intel Xeon Phi processor-based developer platforms.
With one year of Intel Parallel Studio XE.
Training and support from Colfax.
Order your platform today.http://sdm.link/intel
Openvpn-devel mailing list

Reply via email to