| 00:20.51 | *** join/#asterisk-dev elguero (~miguel323@ns1.nashuacs.com) |
| 01:18.06 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 07:01.37 | *** join/#asterisk-dev snuff-work (~aaa@210.9.82.197) |
| 07:01.37 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 08:02.08 | *** join/#asterisk-dev hehol (~hehol@buero-gw.dortmund.loca.net) |
| 08:20.42 | *** join/#asterisk-dev tamiel (~tamiel@213.30.183.226) |
| 08:21.12 | *** join/#asterisk-dev schmidts (~schmidts@lmlo.sil.at) |
| 08:21.15 | schmidts | good morning |
| 08:21.52 | jsmith | Good morning schmidts |
| 08:26.25 | schmidts | how are you jared? |
| 08:37.51 | *** join/#asterisk-dev goddva (~glarsen@77.40.154.242) |
| 09:15.01 | *** join/#asterisk-dev mpe (~mpe@94.127.49.1) |
| 09:20.37 | *** join/#asterisk-dev mpe (~mpe@94.127.49.1) |
| 09:32.50 | *** join/#asterisk-dev elred_ (~elred_@unaffiliated/elred-/x-5010831) |
| 10:38.08 | *** join/#asterisk-dev ccesario (~ccesario@187.75.139.188) |
| 11:11.48 | *** join/#asterisk-dev shamelessn00b (~chatzilla@58-65-172-114.nayatel.pk) |
| 11:11.54 | *** join/#asterisk-dev frawd (~francois@83.35.162.108) |
| 11:17.13 | *** join/#asterisk-dev frawd (~francois@83.35.162.108) |
| 11:44.40 | *** join/#asterisk-dev elguero (~miguel323@ns1.nashuacs.com) |
| 12:10.55 | *** join/#asterisk-dev MuffinMan (~muffinman@asterisk/issue-tracker-bot/muffinman) |
| 12:28.08 | *** join/#asterisk-dev elred_ (~elred_@unaffiliated/elred-/x-5010831) |
| 12:45.54 | *** join/#asterisk-dev sruffell (~sruffell@asterisk/the-kernel-guy/sruffell) |
| 12:45.54 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 13:00.17 | *** join/#asterisk-dev rrb3942 (~rbullock@208.34.105.161) |
| 13:03.42 | *** join/#asterisk-dev hehol (~hehol@buero-gw.dortmund.loca.net) |
| 13:21.30 | *** join/#asterisk-dev guilhermebr (~Guilherme@200.103.96.98) |
| 13:51.22 | *** join/#asterisk-dev elguero (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 13:51.26 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 13:55.17 | *** join/#asterisk-dev lftsy (~lftsy@install.deckpoint.ch) |
| 13:56.12 | *** join/#asterisk-dev lftsy (~lftsy@install.deckpoint.ch) |
| 14:25.56 | *** join/#asterisk-dev jeffg (~jeffg@pdpc/supporter/professional/jeffg) |
| 14:42.29 | *** join/#asterisk-dev datachomper (~Adium@nat/digium/x-hszsgrezjgwtvsuf) |
| 14:49.19 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 14:51.04 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 14:51.04 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 15:05.18 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 15:05.18 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 15:36.30 | *** join/#asterisk-dev elguero (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 15:43.38 | *** join/#asterisk-dev moy (~moy@UNVLON55-1176057127.sdsl.bell.ca) |
| 15:52.47 | *** join/#asterisk-dev anthm (~anthm@freeswitch/developer/anthm) |
| 15:53.53 | *** join/#asterisk-dev mog (~mog@c-68-62-169-225.hsd1.al.comcast.net) |
| 16:05.53 | *** join/#asterisk-dev rrb3942 (~rbullock@208.34.105.161) |
| 16:08.56 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 16:08.56 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 16:15.18 | *** join/#asterisk-dev marv (~timr@router.asteriasgi.com) |
| 16:25.15 | *** join/#asterisk-dev ccesario_ (~ccesario@189-29-53-173-ac.cpe.vivax.com.br) |
| 16:28.07 | *** join/#asterisk-dev Corydon76-lap (~Adium@nat/digium/x-gzlgogtzrynnwxhf) |
| 16:28.07 | *** mode/#asterisk-dev [+o Corydon76-lap] by ChanServ |
| 16:29.48 | *** join/#asterisk-dev Corydon76-lap (~Adium@nat/digium/x-afctrtmgfzexgywr) |
| 16:29.48 | *** mode/#asterisk-dev [+o Corydon76-lap] by ChanServ |
| 16:45.51 | *** join/#asterisk-dev Tim_Toady (~moi@178.128.21.207.dsl.dyn.forthnet.gr) |
| 16:54.44 | pabelanger | So, under FreeBSD (8.1) we need to pass 'ASTCFLAGS=-march=i686' to make, otherwise Asterisk will fail to compile (march=i386 not supported for __sync_fetch_and_add_4). Could we not do this by default in the Makefile? |
| 16:56.06 | kpfleming | we don't try to pass any optimization flags by default, but that can cause this sort of problem. on linux gcc usually defaults to -i686, so we don't see this issue. |
| 16:56.19 | tzafrir | pabelanger, FreeBSD does not support anything < i686? |
| 16:56.40 | Corydon76-lap | FreeBSD supports i386, too |
| 16:56.49 | *** join/#asterisk-dev obnauticus (~obnauticu@about/windows/regular/obnauticus) |
| 16:56.50 | Corydon76-lap | i386, specifically |
| 16:57.18 | Corydon76-lap | though you can tune the kernel out of the slower processors, too |
| 16:57.57 | kpfleming | you can't use atomic ops in gcc on i386 though, which is the issue here, and the FreeBSD gcc defaults to -i386 instead of -i686 |
| 16:58.32 | kpfleming | so i think making i686 the default on FreeBSD would be fine as long as the user didn't already provide an -march or -mcpu parameter themselves |
| 16:59.17 | kpfleming | well... actually if you just specify it in the Makefile, anything the user specifies will override it, so it'd be all good |
| 16:59.47 | pabelanger | great, I'll see what I can come up with |
| 17:00.12 | kpfleming | put it in _ASTCFLAGS, since ASTCFLAGS overrides it |
| 17:00.23 | pabelanger | kpfleming: will do |
| 17:17.27 | *** join/#asterisk-dev mpe (~mpe@0xd99d3f8f.customer.cybercity.dk) |
| 17:49.02 | *** join/#asterisk-dev miguel3239 (~miguel323@h-68-167-124-171.cmbrmaor.static.covad.net) |
| 18:19.39 | pabelanger | Running into an issue with the testsuite on FreeBSD, http://asterisk.pastebin.com/6Wx7RdfR |
| 18:21.35 | pabelanger | looks like a bug in python, fixed in 2.7+ |
| 18:21.41 | pabelanger | http://bugs.python.org/issue1068268 |
| 18:26.37 | Marquis42 | OK, so how am I really supposed to get post-review now that it's not in repotools? The README.post-review says to use 'easy_install', but I have no idea what that's supposed to mean. |
| 18:27.13 | Qwell | easy_install is a python thingie |
| 18:27.28 | russellb | python-setuptools in debian/ubuntu i think |
| 18:27.42 | seanbright | http://tinyurl.com/29wmd87 |
| 18:28.01 | Marquis42 | Alright, funny guy... :P |
| 18:28.10 | seanbright | weeeeeee |
| 18:29.32 | russellb | how very seanbright of you |
| 18:31.17 | Marquis42 | Yay, thanks guys. |
| 18:32.12 | seanbright | russellb: now with more seanbrightiness |
| 18:37.12 | file | that seanbright guy is L-A-M-E |
| 18:40.23 | tzafrir | aptitude install python-rbtools |
| 18:40.45 | russellb | even fancier |
| 18:41.04 | Marquis42 | tzafrir: That doesn't exactly work on Fedora. ;) |
| 18:41.14 | Marquis42 | But python-setuptools is there, so I got it going. |
| 18:41.22 | russellb | yay |
| 18:41.40 | tzafrir | Marquis42, which only means nobody has packaged it yet there. Now, are you going to blame that on seanbright as well? |
| 18:41.57 | russellb | i would |
| 18:42.04 | Marquis42 | tzafrir: I might! :D |
| 18:42.27 | Marquis42 | Although I was talking about aptitude more than python-rbtools (which also doesn't exist). |
| 18:43.02 | Marquis42 | russellb: BTW, that review you told me I'd do just got posted. r1030 |
| 18:43.09 | russellb | R1030 |
| 18:43.13 | MuffinMan | Failed to retrieve review request: You don't have permission for this |
| 18:43.13 | Marquis42 | Errr... Told me that *youd* would review.... |
| 18:43.20 | russellb | have you published it? |
| 18:43.30 | Marquis42 | Doh, no give me just a minute. |
| 18:43.43 | russellb | starts the 1 minute timer |
| 18:44.59 | Marquis42 | Done |
| 19:15.21 | *** join/#asterisk-dev Gugge (~gugge@vlan2.dlxhosting.dk) |
| 19:30.58 | *** join/#asterisk-dev crowbar7 (~JKLOBUCAR@12.2.84.2) |
| 19:41.22 | pabelanger | russellb: We'll need to bump the python requirement for the testsuite to atleast 2.6.6 to fix a python bug with multi-processor machines. |
| 19:43.04 | russellb | alright. |
| 19:43.15 | russellb | I think the linux slaves should have a new enough version already ... |
| 19:43.30 | russellb | I set the low requirement back when I was still testing on the last LTS |
| 19:45.04 | pabelanger | Cool |
| 19:45.26 | pabelanger | Also, do you have some time to provide feedback? R1026 |
| 19:45.36 | pabelanger | r1026 |
| 19:45.51 | pabelanger | heh |
| 19:45.52 | MuffinMan | [R1026] Global debug and verbose settings [Pending] Submitted by pabelanger => https://reviewboard.asterisk.org/r/1026/ |
| 19:46.06 | pabelanger | MuffinMan: have a muffin |
| 19:51.57 | russellb | pabelanger: can you make runtests.py fail gracefully if the version is not sufficient? |
| 19:52.35 | pabelanger | russellb: for python? |
| 19:53.23 | russellb | yes |
| 19:54.27 | pabelanger | Yup, I'll add a check for that under dependency section. Ideally, we should use a configure script? |
| 19:55.29 | russellb | meh |
| 19:55.50 | pabelanger | one day |
| 20:00.15 | seanbright | what's merge order these days? |
| 20:00.25 | seanbright | 1.4 -> 1.6.2 -> 1.8 -> trunk? |
| 20:00.29 | russellb | YES |
| 20:00.30 | russellb | err, yes |
| 20:00.34 | seanbright | THANKS |
| 20:00.40 | russellb | i made a hot wiki page for that |
| 20:00.52 | russellb | https://wiki.asterisk.org/wiki/display/AST/Subversion+Usage |
| 20:01.02 | russellb | it has a flow chart! |
| 20:03.25 | seanbright | hot is right |
| 20:04.21 | russellb | i had too much fun with wiki markup on that one ... |
| 20:05.24 | seanbright | like riding a bike... that being said |
| 20:06.05 | seanbright | the dahdi docs suggestion using dahdi_monitor to record the dahdi channel to see (hear?) whats going on before/after a call |
| 20:06.15 | seanbright | how would one identify polarity reversal in a recording? |
| 20:06.19 | seanbright | does it look like something magical? |
| 20:06.27 | seanbright | it does it just sound like a click? |
| 20:06.30 | russellb | i'm not sure that's something you'd hear ... |
| 20:06.33 | russellb | sruffell: ^^^ |
| 20:09.45 | citywok | Thanks russell, when i get back in the office tomorrow i'll re-run all of the testing for all of the options and make sure everything works properly. |
| 20:09.52 | russellb | cool |
| 20:10.06 | citywok | I left early, feel like hell. i was in need of some dayquil :P |
| 20:10.16 | Qwell | dayquil? pfft. |
| 20:10.23 | Corydon76-lap | seanbright: you shouldn't hear a simple polarity reversal in a recording. It's way too short. |
| 20:10.30 | citywok | i was getting realllll tired of blowing my nose every 30 seconds! |
| 20:10.32 | Qwell | s/d/ny/ is always appropriate |
| 20:10.37 | Qwell | err |
| 20:10.40 | Qwell | s/da/n/ |
| 20:10.51 | citywok | hhaha, i already slept for 13 hours last night! |
| 20:11.11 | Qwell | you don't sleep on nyquil |
| 20:11.11 | citywok | i was around sick people all weekend, and we went skiing yesterday. :P |
| 20:11.18 | Qwell | you go about your day, and it becomes hilarious. |
| 20:11.26 | sruffell | looks. |
| 20:11.41 | citywok | hahaha, nice. i took a couple muscle relaxers before going to bed. skiing hurts. lol |
| 20:11.57 | citywok | ever tried to function after taking ambien? ambien + beer = comedy gold |
| 20:12.12 | seanbright | well i've recorded a call with dahdi_monitor |
| 20:12.14 | russellb | sruffell: it was about seanbright's question about dahdi_monitor and polarity reversals. |
| 20:12.22 | Corydon76-lap | citywok: how about caffeine and beer? |
| 20:12.29 | seanbright | someone picked up their polycom, dialed their cell phone, then answered their cell phone and hung up |
| 20:12.30 | citywok | fourloco baby! |
| 20:12.35 | seanbright | asterisk/dahdi didn't recognize the hangup |
| 20:12.51 | sruffell | seanbright: Yeah...dahdi_monitor won't have any indication of a polarity reversal (would probably manifest as a click). The way to see those is to enable debugging in the driver, then you'll see them on the kernel log. |
| 20:13.01 | seanbright | sruffell: sweet |
| 20:13.05 | seanbright | i shall do that |
| 20:13.08 | seanbright | thanks |
| 20:13.17 | sruffell | let me know if you have problems...I'll see what I can do. |
| 20:13.25 | seanbright | is that a simple options .... debug=yes? |
| 20:13.31 | sruffell | which driver are you using? |
| 20:13.51 | seanbright | wctdm24xxp |
| 20:14.04 | sruffell | ok..one sec...I'll grab what you'll want and what you need to look for. |
| 20:14.15 | seanbright | thank you, sir |
| 20:15.50 | sruffell | "modprobe wctdm24xxp debug=1" then you'll see the following lines in dmesg: |
| 20:15.50 | sruffell | https://github.com/sruffell/dahdi-linux/blob/master/drivers/dahdi/wctdm24xxp/base.c#L1755 |
| 20:16.14 | seanbright | well looky there |
| 20:16.24 | seanbright | perfect |
| 20:16.42 | Corydon76-lap | Does debug=15 give you any more debug? |
| 20:16.58 | datachomper | oh no, he cranked it up past 11 |
| 20:17.03 | sruffell | lol |
| 20:17.11 | seanbright | super noob question: will that modprobe have an affect on an already loaded module or do i have to unload it first? |
| 20:17.22 | datachomper | Corydon76-dig: Debug is a bitfield, not an integer as you might expect with asterisk -rvvvvvvvvvvvvvvvvvvv |
| 20:17.46 | sruffell | you'll need to unload it...but....you're in luck...since that's read each time, you can just "echo 1 > /sys/modules/wctdm24xxp/parameters/debug" if you have a newish kernel |
| 20:17.56 | russellb | <PROTECTED> |
| 20:18.05 | seanbright | sruffell: nice |
| 20:18.16 | Corydon76-lap | datachomper: that's why I said 15 and not 16 |
| 20:18.22 | sruffell | so if you echo to sysfs, you don't need to unload the driver |
| 20:18.48 | Corydon76-lap | Just like t1e1override |
| 20:18.59 | sruffell | fears t1e1override |
| 20:20.05 | datachomper | Yea, I tried to rewrite t1e1override last week but it kicked my butt. Its going to have to wait until our dahdi_cfg revamp |
| 20:20.10 | seanbright | just did it and it didn't take the server down |
| 20:20.14 | seanbright | that's a win if you ask me |
| 20:20.19 | Corydon76-lap | btw, does t1e1override extend into 8 bits if you have two 4-port cards? |
| 20:20.53 | datachomper | Yea, each bit should be one of the registered spans |
| 20:20.56 | seanbright | sruffell: http://pastebin.com/TX2MV96A |
| 20:20.57 | datachomper | In order that they register |
| 20:21.07 | datachomper | very confusing |
| 20:21.30 | seanbright | sruffell: although there is exclaiming going on, i'm assuming the BATTERY messages are not problematic |
| 20:21.46 | Corydon76-lap | That's not confusing, at allâ¦. speaking from the perspective of one who has done assembly and can do hexadecimal to decimal conversions in his head... |
| 20:22.57 | Corydon76-lap | What would be confusing is if the bits in t1e1override had to match up with the cards, in left to right order⦠i.e. MSB allocated first |
| 20:23.40 | Corydon76-lap | even more confusing when you switch from a 32-bit system to a 64-bit system, and back again |
| 20:29.10 | seanbright | what irc client are you using that converts ... to the unicode ellipsis character? |
| 20:29.12 | *** join/#asterisk-dev i_heart_asterisk (~chatzilla@firewall-a.buf.ny.i-evolve.net) |
| 20:29.40 | Corydon76-lap | Adium |
| 20:29.57 | seanbright | ah |
| 20:30.00 | Corydon76-lap | and I did not know of that conversion |
| 20:30.14 | seanbright | ⦠|
| 20:30.30 | Corydon76-lap | . .. ⦠â¦. ..... |
| 20:30.52 | Corydon76-lap | They don't appear to be any different, to me |
| 20:31.05 | Corydon76-lap | IESH5 |
| 20:31.23 | Corydon76-lap | Oh, wait, that's EISH5 |
| 20:32.40 | seanbright | hanguponpolarityswitch=yes makes a world of difference |
| 20:32.42 | seanbright | sruffell: thanks! |
| 20:35.21 | sruffell | seanbright: heh..sorry about the silence there, I walked away from my desk. But it looks like you're all set? |
| 20:35.36 | seanbright | i believe so |
| 20:35.53 | seanbright | so now asterisk will treat reverse polarity as a hangup |
| 20:35.57 | sruffell | cool.... |
| 20:36.20 | seanbright | as far as how it signals hangup, will that just work? |
| 20:36.20 | datachomper | Corydon76-lap: Planning on moving that parameter into the dahdi config file, so it's easy to configure on a per span basis |
| 20:36.47 | russellb | pbx_load_config() in pbx_config.c makes me want to cut myself |
| 20:37.08 | sruffell | seanbright: as far as I know. Honestly, I don't deal with too many installations where the defaults don't just work, and most of the exceptions predate me.... |
| 20:37.23 | seanbright | gotcha |
| 20:37.30 | seanbright | this is belize |
| 20:37.43 | sruffell | but I like chasing down those types of problems because that's how I learn. |
| 20:37.51 | seanbright | so far it looks similar to mexico, but this is the first time i've really dealt with non-domestic analog |
| 20:38.39 | sruffell | is there only one provider in Belieze? |
| 20:38.43 | seanbright | yeah |
| 20:38.49 | sruffell | ok...I'll file that fact away. :) |
| 20:38.51 | seanbright | BTL |
| 20:39.45 | datachomper | I volunteer to fly to Belize to help troubleshoot |
| 20:39.48 | seanbright | heh |
| 20:39.54 | seanbright | me first |
| 20:40.20 | seanbright | i'm tweaking remotely and i have a guy down there picking up and dialing every few minutes to test |
| 20:49.05 | datachomper | The last company I worked for had just landed a contract to put a voice polygraph server in Vodafone India's network, accessible to the users via one of those *DIAL type mobile numbers. The engineer before me had gotten so far as to fly to India and install the server physically before he quit. |
| 20:49.49 | datachomper | It was left to me to figure out how to get this server talking to their network over ss7 and it was sitting behind a firewall that allows zero outgoing connections. |
| 20:50.54 | datachomper | I ended up doing a vpn over the telnet port to get internet access to the server for OS updates etc. But the fun part was testing out the voice polygraph. I didn't know anybody in India so I started dialing random indian numbers based off the contact numbers I had for Vodaphone and trying to keep people on the phone for long enough to collect some voice polygraph statistics. |
| 20:51.53 | datachomper | Turns out, I was actually dialing random executives at Vodaphone and trying to keep them up at what was about 1am at night for them |
| 20:55.35 | *** part/#asterisk-dev citywok (~kvirc@67-134-194-33.dia.static.qwest.net) |
| 20:55.43 | *** join/#asterisk-dev citywok (~kvirc@67-134-194-33.dia.static.qwest.net) |
| 20:56.15 | Qwell | datachomper: so, does the polygraph stuff know the difference between tired and lying? |
| 20:56.36 | Qwell | "It's 1am and I'm sleeping!" "...huh. that registered as a lie." |
| 20:57.54 | datachomper | It was total snake oil |
| 20:58.48 | datachomper | But it was a licensed 3rd party analyzer from a reputable researcher in the area |
| 21:00.52 | Corydon76-lap | datachomper: also doesn't help to talk to people who don't fear being discovered to be lying |
| 21:01.10 | *** join/#asterisk-dev Gugge (~gugge@vlan2.dlxhosting.dk) |
| 21:02.48 | citywok | Any of the digium guys on the AA50 team i found a problem in the software. |
| 21:02.58 | Corydon76-lap | Or the mentally ill. "I'm not Napoleon" "The machine says you're lying" |
| 21:03.11 | citywok | I had to change the password on a trunk, but it wouldn't let me in the interface. i had to go in to users.conf and fix it there by hand. |
| 21:03.52 | citywok | actually, it's just the save button doesn't work at all. lol. |
| 21:05.18 | Corydon76-lap | citywok: you need to restart for anything to take effect |
| 21:05.28 | citywok | Yes, but it won't let me PRESS save. lol |
| 21:05.36 | Corydon76-lap | Are you logged in? |
| 21:05.40 | citywok | like, the action of pressing save doesn't work |
| 21:05.48 | citywok | yes, i can press cancel, then click edit again, and repeat |
| 21:06.00 | Corydon76-lap | Browser? |
| 21:06.07 | citywok | currently testing in I.E. to make sure it isn't a chrome issue |
| 21:06.16 | Corydon76-lap | Use Firefox only |
| 21:06.26 | citywok | i don't have 50gb of memory :P |
| 21:06.35 | Corydon76-lap | It was the primary browser used for development. |
| 21:06.50 | Corydon76-lap | and for the longest time, even IE didn't work correctly |
| 21:07.21 | citywok | yup, it works in IE8 now, without compat mode |
| 21:07.44 | Corydon76-lap | Chalk it up to a JS incompatibility |
| 21:08.19 | citywok | kk. at least it's not as bad as Aastra's web interface |
| 21:09.02 | Corydon76-lap | You're welcome to report it as a bug, but for AstGUI on the open source tracker |
| 21:09.29 | citywok | i ditched firefox for chrome about 6 months ago, it's unbelievable how mcuh more responsive it is. i was using firefox since before i.e. had tabs. |
| 21:09.56 | Corydon76-lap | IE doesn't run on either Mac or Linux, which is why it's not an option here |
| 21:10.37 | Corydon76-lap | Chrome could become an option. I'll look at it when I have some spare time (maybe this weekend) |
| 21:16.55 | seanbright | so looking at some of the tone zones defined in tonezone.c |
| 21:18.09 | seanbright | i don't understand why a given zone would have a ring cadence of, say, 2000/4000 and then a ringtone which was 1200/4650 |
| 21:18.20 | seanbright | shouldn't they be the same? |
| 21:29.05 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 21:29.05 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 21:42.38 | pabelanger | Now that we are creating a configs directory for the testsuite, I propose any .config file that binds to an interface, be remapped to 127.0.0.1. We also have the ability to utilize more of the 127.0.0.0/8 address block. |
| 21:43.12 | pabelanger | This would help solve firewall issue, since 127.0.0.0/8 is rarely in ACLs |
| 21:45.21 | *** join/#asterisk-dev anthm][ (~anthm@cpe-184-58-189-244.wi.res.rr.com) |
| 22:04.06 | Corydon76-lap | pabelanger: how is it binding to an interface? |
| 22:04.34 | *** join/#asterisk-dev snuff-work (~aaa@210.9.82.197) |
| 22:04.34 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 22:04.53 | pabelanger | Well, not an interface but 0.0.0.0 is used by default for most sample configs |
| 22:06.21 | Corydon76-lap | Oh, and 127.0.0.0/8 is frequently in firewalls⦠just to forbid from any interface other than lo |
| 22:07.17 | Corydon76-lap | I'm fine with the remap, but how are you ensuring that the bind is changed? |
| 22:08.02 | Corydon76-lap | A sed on each known conf file? |
| 22:08.56 | Corydon76-lap | A library layer between Asterisk and the kernel? |
| 22:09.13 | pabelanger | Corydon76-lap: I've created a configs/ directory within the testsuite. Basically creating its own default config files, and using #include statement to allow tests to override settings |
| 22:10.06 | Corydon76-lap | I'm not sure about the #include statement. |
| 22:11.29 | pabelanger | about using it or it working? |
| 22:11.43 | Corydon76-lap | I'm not sure how you're using it |
| 22:12.16 | Corydon76-lap | Implicit paths that tests need to override? |
| 22:13.13 | pabelanger | not sure I understand the question |
| 22:13.15 | Corydon76-lap | Are those paths that get passed to the tests on invocation? |
| 22:13.45 | Corydon76-lap | You've said you're using #include to allow tests to override settings. How? |
| 22:14.07 | pabelanger | Here is an example: http://svnview.digium.com/svn/testsuite/asterisk/trunk/configs/ |
| 22:14.44 | pabelanger | If a test was to have a copy logger.logfiles.conf.inc, those settings would be included into logger.conf |
| 22:15.00 | Qwell | unfortunately, #include fails (intentionally) when the file doesn't exist. |
| 22:15.20 | Corydon76-lap | Right, that's the part that I wasn't clear on |
| 22:15.22 | Qwell | You might be able to do "#exec cat somefile" though |
| 22:15.38 | Corydon76-lap | So you're going to have a massive listing of possible #include files |
| 22:15.47 | Corydon76-lap | All (mostly) empty |
| 22:15.49 | pabelanger | Qwell: Yes, so by default I create an empty .conf file within configs/ to get around that |
| 22:16.01 | pabelanger | Corydon76-dig: Yes, that is the down side |
| 22:16.34 | Corydon76-lap | pabelanger: I think I prefer having tests write their own configs |
| 23:23.39 | *** join/#asterisk-dev infobot_ (~infobot@rikers.org) |
| 23:23.39 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- http://xkcd.com/303/ |
| 23:24.21 | Corydon76-lap | pabelanger: I just measured. The worst performance for creation/destruction of inodes is going to be on OS X. Solaris just yawns at the task. |
| 23:25.37 | Corydon76-lap | Linux is slightly worse, but OS X really chews on the task |
| 23:30.43 | jdoe | Corydon76-lap: ext3 I assume? |
| 23:30.57 | Corydon76-lap | jdoe: correct |
| 23:33.57 | *** join/#asterisk-dev infobot (~infobot@rikers.org) |
| 23:33.57 | *** topic/#asterisk-dev is Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- http://xkcd.com/303/ |
| 23:42.01 | pabelanger | Corydon76-dig: It should also help that we are not going to do make uninstall-all, install, samples each test. Just once before launching testsuite |
| 23:42.23 | pabelanger | Corydon76-lap: ^ |
| 23:44.54 | Corydon76-lap | pabelanger: yep |
| 23:46.01 | Corydon76-lap | pabelanger: but in terms of improving performance, parallelizing the tests and adding a framework to allow the tests to handover the cleanup tasks to the testsuite master, is going to provide the best possible speed improvement |
| 23:46.21 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 23:46.21 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 23:46.29 | pabelanger | I would agree |
| 23:46.39 | pabelanger | plus how cool would that be |
| 23:46.42 | Corydon76-lap | A registration for cleaning a directory, registration for ensuring an Asterisk instance dies |
| 23:47.24 | pabelanger | the only issue for parallel testing to to make sure CPU performance does not affect the results |
| 23:48.37 | Corydon76-lap | Fair enough. |
| 23:49.41 | Corydon76-lap | Maybe the testsuite master should monitor the amount of available system memory and make judgements on when to launch additional tests, based upon that. |
| 23:50.24 | Corydon76-lap | That allows systems with more memory to take advantage and parallelize more tasks |
| 23:50.45 | Corydon76-lap | Just need to ensure that tests don't collide with port selections |
| 23:59.33 | pabelanger | Corydon76-lap: Ya, that would be the other thing to monitor... having the master some how control the allocation ports so clients to collide |