| 00:40.39 | *** join/#asterisk-dev citywok (~citywok@67-134-194-33.dia.static.qwest.net) |
| 01:26.09 | *** join/#asterisk-dev seraphie (~erin@75.76.38.159) |
| 01:42.35 | *** join/#asterisk-dev bklang (~bklang@tesla.alkaloid.net) |
| 01:56.21 | *** join/#asterisk-dev mjordan (~mjordan@user-69-1-6-49.knology.net) |
| 01:56.21 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 01:58.36 | *** join/#asterisk-dev otherwiseguy (~otherwise@69.76.221.79) |
| 02:21.46 | *** join/#asterisk-dev snuff-work (~snuffy@210.9.82.197) |
| 02:21.46 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 02:31.42 | *** join/#asterisk-dev otherwiseguy_ (~otherwise@69.76.221.79) |
| 02:31.53 | *** part/#asterisk-dev otherwiseguy_ (~otherwise@69.76.221.79) |
| 04:53.19 | *** join/#asterisk-dev otherwiseguy (~otherwise@cpe-69-76-221-79.kc.res.rr.com) |
| 05:00.18 | *** join/#asterisk-dev irroot (~irroot@197.110.250.107) |
| 05:37.05 | *** join/#asterisk-dev happylife (~happylife@212.92.145.7) |
| 06:33.28 | *** join/#asterisk-dev philipp64|laptop (~chatzilla@mail.redfish-solutions.com) |
| 06:48.17 | *** join/#asterisk-dev otherwiseguy (~otherwise@cpe-69-76-221-79.kc.res.rr.com) |
| 06:53.20 | *** join/#asterisk-dev oej (~olle@ns.webway.se) |
| 06:53.20 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 07:52.03 | *** join/#asterisk-dev snuff-work (~snuffy@210.9.82.197) |
| 07:52.03 | *** mode/#asterisk-dev [+o snuff-work] by ChanServ |
| 08:06.27 | *** join/#asterisk-dev hehol (~hehol@2001:1438:1009:200:3ca8:393f:bc1a:a415) |
| 08:10.38 | *** join/#asterisk-dev Qwell (~north@pdpc/sponsor/digium/Qwell) |
| 08:10.38 | *** mode/#asterisk-dev [+o Qwell] by ChanServ |
| 08:17.53 | *** join/#asterisk-dev irroot (~irroot@pbx.distrotech.co.za) |
| 08:37.38 | *** join/#asterisk-dev citywok (~citywok@67-134-194-33.dia.static.qwest.net) |
| 08:39.47 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 08:39.47 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 08:45.11 | *** join/#asterisk-dev tamiel (~tamiel@213.30.183.226) |
| 08:54.52 | *** join/#asterisk-dev oej (~olle@static-213-115-251-100.sme.bredbandsbolaget.se) |
| 08:54.52 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 08:56.19 | *** join/#asterisk-dev irroot (~irroot@pbx.distrotech.co.za) |
| 08:56.37 | irroot | have these techies never heard of Heisenberg uncertainty principle |
| 08:56.47 | oej | hello folks |
| 08:56.57 | irroot | lo there oej |
| 08:57.10 | oej | Which techies and what principle? Feel stupid. |
| 08:58.34 | *** join/#asterisk-dev schmidts (~schmidts@Lo20-rt01.lm33.sil.at) |
| 08:58.36 | schmidts | good morning |
| 08:58.53 | irroot | looks like the EU is in da house |
| 08:58.56 | *** join/#asterisk-dev jsmith (~jsmith@fedora/jsmith) |
| 08:58.56 | *** mode/#asterisk-dev [+o jsmith] by ChanServ |
| 08:59.14 | irroot | jsmith mmm welcome aboard you home or close to it |
| 09:00.05 | schmidts | irroot :D |
| 09:00.08 | jsmith | irroot: Still in Amsterdam |
| 09:00.31 | irroot | jsmith go for a "coffee" at a er um coffee shop ? |
| 09:01.03 | irroot | jsmith we will count you in with the europeans for now :P |
| 09:01.25 | jsmith | doesn't drink coffee either |
| 09:01.33 | jsmith | starts to feel like an outcast |
| 09:01.52 | schmidts | jsmith you know the "coffee shop" in amsterdam mostly even sell coffee ;) |
| 09:02.19 | jsmith | schmidts: So I hear... |
| 09:02.32 | oej | Why leave Europe, jared? Just stay. |
| 09:02.53 | oej | irroot: Just got email that referred to the same principle. I'm surrounded. |
| 09:03.59 | wdoekes2 | haha |
| 09:05.20 | mvanbaak | most coffeeshops sell coffee |
| 09:05.23 | mvanbaak | :) |
| 09:05.26 | mvanbaak | trust me |
| 09:05.43 | schmidts | mvanbaak but most people dont go there for a coffee ;) |
| 09:05.48 | mvanbaak | amongst other things ;-) |
| 09:06.10 | mvanbaak | well, I always buy coffee |
| 09:06.27 | jsmith | oej: I'd love to. You gonna pay for my divorce? |
| 09:06.30 | mvanbaak | so I can sit down at a sofa, relaxt, have a coffee while processing the other things I bought |
| 09:06.33 | mvanbaak | ;) |
| 09:07.01 | mvanbaak | jsmith: why divorce? just buy your wife a ticket |
| 09:07.29 | jsmith | mvanbaak: Someone has to watch the children... |
| 09:07.52 | jsmith | mvanbaak: And if you're suggesting I buy them tickets -- you've overestimated my financial worth :-p |
| 09:08.01 | mvanbaak | whehehehe |
| 09:08.06 | irroot | oej i have a sip server with 3DSL links SIP set up to register to the same server down all 3 links but to get it there i have 3 routing tables and default in each with a mark 3 phantom ip's not routed anywhere but dest nated on way out setting mark to appropriate routing table so that 3 accounts can register to same ip remotely one per DSL line .... its prolly not the best way but works |
| 09:08.20 | wdoekes2 | subsidised daycare in europe will pay it back, jsmith |
| 09:08.56 | irroot | wdoekes2 until austerity comes knocking with the IMF goons |
| 09:09.25 | wdoekes2 | oej's region is still free of debt |
| 09:10.57 | irroot | wants to be debt free |
| 09:13.55 | oej | my region is not free of debt. Norway is. We need to invade them and let them occupy Sweden and pay our debts. |
| 09:14.10 | oej | And we still have SAAB. The Americans would not let us sell it to China. |
| 09:14.41 | jsmith | :-( |
| 09:14.43 | irroot | oej hehe SAAB bailed out by the south african arms industry :) in a "arms" deal that raped us |
| 09:15.44 | irroot | we also bailed out the canadians in a deal for a train hehe what would you 1st world folk do without us .... |
| 09:16.36 | *** join/#asterisk-dev black187 (5d67162a@gateway/web/freenode/ip.93.103.22.42) |
| 09:18.26 | black187 | Hello, one question - I have an i7 processor with 4 cores, and I'm stress testing Asterisk on it - so far I've got to 700 concurent calls. But when I check with command "top" and press "1" - I see that Asterisk uses only one core. Is this a normal thing, or not? Because if I can get it to work on multuple core, the concurent calls would be a lot bigger. |
| 09:19.25 | schmidts | black187 what asterisk version and what type of calls do you test? |
| 09:19.45 | oej | That's not normal. Asteirsk is heavily multi-threaded. |
| 09:19.52 | oej | What type of calls/channels are you using? |
| 09:20.27 | black187 | Asterisk 1.8.7.1 - the newest stable version. I'm testing with SIPP - make a call, play music on hold for 60 seconds, hangup. I have 2 SIPP servers, who make calls to Asterisk box. |
| 09:21.00 | black187 | I know that asterisk is multi-threaded, that's why I was suprised to see, that with "top" command it uses just one processor. |
| 09:21.19 | schmidts | i dont konw top very well but take a look at htop there you can easy see how much load is on every cpu core |
| 09:21.23 | black187 | Did anybody make any test here with multi core processors, so I can see, if its only me. |
| 09:21.34 | schmidts | raise hands :D |
| 09:21.52 | black187 | @schmidts: Hmm, I will check it. Oh, I'm using Centos 5.7 64 bit. |
| 09:21.57 | schmidts | 3200 concurrent calls and a well spread load over all of my 24 cores |
| 09:22.04 | irroot | schmidts is our resident profiler :P |
| 09:22.31 | oej | just fyi black187: It's not a "stable" version. We never assign that label to any release. And it's not stable at all in my personal view. |
| 09:22.50 | schmidts | oej :P |
| 09:22.59 | oej | I have done a lot of tests with multicore and Asterisk distributes fine. |
| 09:23.09 | black187 | 24 real cores, or 12 cores with 12 virtualized |
| 09:23.11 | black187 | ? |
| 09:23.13 | oej | The SIP thread is the largest one but you still have threads on all cores |
| 09:23.18 | oej | We had 16 cores |
| 09:23.27 | oej | four quad-core |
| 09:23.37 | oej | We had over 11.000 channels on that system |
| 09:23.58 | black187 | 11.000 - damn! |
| 09:24.20 | black187 | Do you have to compile asterisk differently, or is it just normal procedure |
| 09:24.52 | schmidts | black187 2x 6 core with hyperthreading |
| 09:24.58 | oej | I was using what I consider "stable" - Asterisk 1.4 with a non-special compilation |
| 09:25.02 | black187 | I just used make -j 4, but that's just for faster make. |
| 09:27.12 | black187 | Ok, so 2 processors with 6 cores. I will check if I can get one - what kind of proc is the best for Asterisk? Intel, AMD - or it doesn't matter. We have a partnership with Dell, so we can get good prices on their servers. |
| 09:28.07 | mvanbaak | as long as it is X86 |
| 09:29.41 | oej | You need to focus on why you have all on one thread... |
| 09:29.49 | black187 | Ok, thanks for that info. And the cache must be very big for Asterisk I presume. With low latency. So Xeon would be good? |
| 09:29.54 | oej | OR all threads on one core |
| 09:30.20 | irroot | black187 we got customers on atom boxes :P |
| 09:30.33 | black187 | What linux distribution do you use? |
| 09:30.43 | mvanbaak | I do not use linux |
| 09:30.53 | black187 | @irrot: Are they stable - the atom boxes? |
| 09:31.48 | mvanbaak | my home asterisk is running on this cpu: |
| 09:31.48 | black187 | @oej, I will check, maybe its the Centos distirbution. Any idea, how to check it? |
| 09:31.49 | mvanbaak | 233 Mhz NSC SC1100 single chip processor |
| 09:32.07 | mvanbaak | soekris 4801 |
| 09:32.15 | irroot | black187 we use them for sites upto 20 users and they stable look more like gateways than servers one site with 6 staff use it as a fileserver / domain controller / firewall with samba and iptables |
| 09:32.37 | irroot | activate the integrated wifi as a hotspot they love them |
| 09:33.36 | black187 | for smaller sites we use pika warp hardware - it's great for small price - up to 70 users. |
| 09:33.55 | mvanbaak | wow, there is now a soekris with gigabit ports ..... |
| 09:33.57 | schmidts | oej this 11.000 call setup what kind of calls did you test? i mean what extension did you use? |
| 09:35.03 | irroot | black187 this box is 1800R thats about 180E |
| 09:37.35 | black187 | @schmidts and @oej: what linux distribution you use? |
| 09:37.52 | oej | centos |
| 09:37.57 | schmidts | debian squeeze |
| 09:38.09 | schmidts | but in an openvz container as a virutal machine |
| 09:38.25 | mvanbaak | freebsd ! |
| 09:41.08 | irroot | mvanbaak is it a coincedence that Berkley brought us LSD and BSD ? |
| 09:41.22 | mvanbaak | no |
| 09:41.27 | mvanbaak | and I like both ;-) |
| 09:41.48 | mvanbaak | gheh |
| 09:41.58 | mvanbaak | and it is closely related to my email signature tagline |
| 09:41.59 | mvanbaak | "Why is it drug addicts and computer aficionados are both called users?" |
| 09:41.59 | irroot | black187 source code bootsraped off slackware v3 |
| 09:42.04 | black187 | Ok, i've made htop command with something like 960 concurent calls -> And as far as I can see, I use mostly just on core. Screenshot -> http://imageshack.us/photo/my-images/831/htop.png/ |
| 09:42.07 | irroot | linux 3.1 now |
| 09:43.13 | black187 | Thanks for the linux distribution info. |
| 09:48.38 | black187 | Does the "htop" screenshot show, that I only use one core? |
| 09:49.11 | schmidts | black187 thats how it looks on my system with 4000 concurrent calls and just playback of tt_monkeys: http://imageshack.us/photo/my-images/35/loadtest.png/ |
| 09:49.52 | oej | wdoekes:still around? |
| 09:50.42 | wdoekes2 | sure |
| 09:52.50 | oej | In the review of externip, https://reviewboard.asterisk.org/r/1471/, you said something about ast_sockaddr_stringify I don't understand. |
| 09:53.28 | oej | Ahh. can't use it twice in the same printf |
| 09:53.35 | wdoekes2 | :) |
| 09:53.52 | black187 | schmidts - I see, that you have in use multiple cores. Ok, so the problem is with my system. Is it possible that the Centos distribution is not good at working with multiple cores...? |
| 09:55.17 | black187 | Or I just didn't compile it wrong, I will try with hoard, to see if there will be any difference |
| 10:15.53 | schmidts | is it normal that asterisk schedules a sip_poke_peer for peers with qualify=no? |
| 10:27.35 | schmidts | irroot i can reproduce my problem \o/ 100 cps for music on hold and i set a bunch of block on ast_read |
| 10:28.05 | irroot | schmidts cool so now to figgure out why |
| 10:28.35 | irroot | what flavor music on hold ?? is it a stream or a file .... |
| 10:28.46 | schmidts | a file |
| 10:29.13 | schmidts | oh its even 50 cps and i see this beginning at 200 concurrent calls |
| 10:29.23 | *** join/#asterisk-dev happylife (~happylife@212.92.145.7) |
| 10:29.46 | schmidts | btw it looks like oej load test will stay the best for a while, this 1.8 version just stopped creating pbxs on about 7500 concurrent calls :D |
| 10:35.44 | irroot | schmidts wonder if its not a IO thing i prefer streaming the MOH |
| 10:36.17 | schmidts | irroot i dont know but it really looks like the same lock/block problem i had on production state |
| 10:36.56 | irroot | yeah it just freezes up trying to read frames |
| 10:38.37 | schmidts | but i dont see a difference when using timerfd or pthread timing |
| 10:45.38 | schmidts | wow never seen asterisk crashing for a core show locks :D |
| 10:47.33 | schmidts | hmm 397 threads all locking in any way together sounds not like fun :( |
| 10:49.22 | schmidts | can someone take a look at this: http://pastebin.com/MqWtwvj1 |
| 10:50.32 | schmidts | this looks like a deadlock to me |
| 10:53.18 | irroot | schmidts looking at it |
| 10:55.12 | irroot | its trying to find a timer on line 13 calling the container |
| 10:55.28 | schmidts | yes and at this point it gets stuck |
| 10:55.51 | schmidts | and all other threads trying to call the same container and get stuck too cause they wait on this lock |
| 10:56.06 | irroot | and hanging locking the container of timers ? |
| 10:56.34 | irroot | can you find the thread that the container lock is held |
| 10:56.45 | schmidts | you mean the first i will try |
| 10:57.05 | irroot | well not first but the one that has the lock |
| 10:57.45 | irroot | __ast_read needs to ack the timer and is been held up |
| 10:58.29 | schmidts | i will put the whole gdb output + locks output online |
| 10:58.32 | schmidts | give me a second |
| 10:59.44 | irroot | schmidts remember im a amatuer here :P but really keen on helping you out |
| 11:05.53 | schmidts | irroot :P |
| 11:05.54 | schmidts | http://sst.sil.at/lock/ |
| 11:06.06 | schmidts | sorry but its just to big for pastebin |
| 11:06.19 | irroot | you can zcat to ftp ? |
| 11:06.36 | irroot | np got |
| 11:08.37 | irroot | schmidts can you th apply all bt full ? |
| 11:09.52 | schmidts | go on the web page |
| 11:10.02 | schmidts | there you have both, bt full and also thread apply all bt |
| 11:11.11 | irroot | yeah but not the big "thread apply all bt full" |
| 11:12.28 | irroot | also with a lock file grep -E "^==" the out put then its easier to work with |
| 11:13.09 | schmidts | ok will do it, give me a minute ;) |
| 11:16.15 | irroot | ive done it here np just 411 |
| 11:16.42 | schmidts | irroot reload the webpage, should be allready up there |
| 11:18.10 | irroot | im looking for where it is locked cant see it |
| 11:18.49 | schmidts | they all wait for the same lock |
| 11:18.55 | schmidts | Thread 292 |
| 11:19.13 | irroot | let me look was looking at core show |
| 11:20.03 | schmidts | looking at locking :D |
| 11:24.13 | *** join/#asterisk-dev pnlarsson (~niklas@fw1.gml.g.icnet.infracom.se) |
| 11:30.01 | irroot | schmidts thread 274 |
| 11:32.18 | irroot | schmidts you using the pthread timer again |
| 11:37.26 | schmidts | yes, i have switched cause i have tried to change something in the code |
| 11:42.14 | *** join/#asterisk-dev tzafrir (~tzafrir@local.xorcom.com) |
| 11:42.14 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 11:55.25 | schmidts | irroot what would happen when i dont use timerfd or pthread timing? |
| 11:58.04 | schmidts | ok, for what does i need this timing stuff? |
| 11:58.17 | schmidts | pbx-03*CLI> core show channels count |
| 11:58.17 | schmidts | 7000 active channels |
| 11:58.17 | schmidts | 7000 active calls |
| 11:58.17 | schmidts | 7000 calls processed |
| 11:58.17 | schmidts | <PROTECTED> |
| 11:58.58 | irroot | if timing fails no dahdi/timerfd/pthreads asterisk wont load |
| 11:59.22 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 11:59.22 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 11:59.28 | schmidts | i havent compiled dahdi on this machine, its a openvz and i have a noload for timerfd and pthreads |
| 11:59.39 | irroot | its running ? |
| 11:59.53 | schmidts | see above 7000 concurrent calls on musiconhold |
| 11:59.59 | irroot | ok cool |
| 12:00.28 | schmidts | but i dont know if its really working or if its just loosk like its working ;) |
| 12:00.45 | irroot | if it barks like a dog its a cat |
| 12:00.52 | schmidts | :P |
| 12:03.25 | irroot | timing test on the cli ? |
| 12:03.52 | schmidts | give me a second i am just testing the quality |
| 12:04.39 | irroot | things like moh need timing |
| 12:05.49 | schmidts | it works fine |
| 12:06.03 | schmidts | Attempting to test a timer with 50 ticks per second. |
| 12:06.04 | schmidts | Failed to open timing fd |
| 12:06.04 | schmidts | Command 'timing test ' failed. |
| 12:06.09 | schmidts | so timing test failes ;) |
| 12:08.13 | irroot | schmidts so it seems timing is a load killer |
| 12:08.27 | schmidts | no its a load blocker :D |
| 12:09.37 | irroot | ok new performance boost for RT queues it can now be cached for periodic look up or never / always |
| 12:10.27 | irroot | with never the queue will be loaded on demand and never rechecked a AMI/CLI reload is required this is consistant with static and desirable |
| 12:10.39 | schmidts | sounds great |
| 12:11.27 | schmidts | and bt musiconhold sounds fine, even with 5000 concurrent calls on the same moh file |
| 12:11.35 | irroot | mmm |
| 12:11.37 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 12:11.37 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 12:11.38 | schmidts | but i am not sure if sipp really act loads to moh |
| 12:11.45 | schmidts | s/act/add/ |
| 12:12.04 | irroot | kpfleming top o the morn |
| 12:13.15 | schmidts | irroot confbridge doesnt work anymore without timing but i can live with this ;) |
| 12:13.31 | kpfleming | irroot: yes, it is. stupid time change has me all messed up. |
| 12:13.35 | irroot | schmidts bit of a hack but streaming moh is a win save it as a raw slin file and cat it |
| 12:14.17 | irroot | kpfleming and a week apart from europe too confuse all |
| 12:14.50 | schmidts | irroot good idea ;) |
| 12:16.23 | irroot | http://pastebin.com/4jPe1RhJ schmidts quick and dirty way of playing raw slin files but timing will be required its read every 20ms from moh basically a wrapper for cat everyone will hear the same thing at the same time |
| 12:18.40 | schmidts | i am not sure if this would be a good idea for 1008 moh classes ;) |
| 12:18.59 | irroot | schmidts nah that is out of scope for sure |
| 12:19.18 | schmidts | :P |
| 12:19.34 | irroot | did say it was quick and dirty |
| 12:20.59 | schmidts | and btw i think asterisk 1.2 didnt have timing too so maybe thats why this machines perform so good |
| 12:21.36 | irroot | schmidts you onto something here indeed |
| 12:22.07 | irroot | the behavior you were having smelt like a timer problem |
| 12:22.25 | schmidts | it really looks like this |
| 12:22.26 | irroot | perhaps the way the hypervisor works ? |
| 12:22.41 | kpfleming | you can also stream the moh over a socket into asterisk |
| 12:22.51 | schmidts | i think its just too slow in doing what it should do and get stuck within some locks |
| 12:29.22 | *** join/#asterisk-dev black187 (5d67162a@gateway/web/freenode/ip.93.103.22.42) |
| 12:31.49 | black187 | Hi, I'm trying to compile Asterisk with hoard (configure, make, make install goes OK) - via https://wiki.asterisk.org/wiki/display/AST/Using+the+Hoard+Memory+Allocator+with, but are having problems with starting Asterisk - error: "/usr/sbin/asterisk: error while loading shared libraries: libhoard.so: cannot open shared object file: No such file or directory". I've checked the path to libhoard.so directory several times, and its |
| 12:39.43 | schmidts | ok now i have to wait again, if it really was this timing stuff then i hope the next hour everything will work fine. if not i am going to become a drawer or singer or something else which has nothing to do with computers |
| 12:42.18 | kpfleming | black187: your message was cut off, but most likely you need to set LD_LIBRARY_PATH to the directory where libhoard.so lives before starting Asterisk |
| 12:43.12 | black187 | Yeah, I did -> export LD_PRELOAD=/path/to/libhoard.so |
| 12:43.23 | black187 | But with my path offcourse. |
| 12:44.27 | black187 | Do I have to set the LD_LIBRARY_PATH instead of LD_PRELOAD? |
| 12:44.29 | irroot | schmidts striper go full monty |
| 12:45.00 | schmidts | irroot i dont want to starve so i guess striper wouldnt be the best choice :D |
| 12:45.46 | wdoekes2 | black187: probably.. but you shouldn't need either.. I got it working just fine when configured with --with-hoard=/path/to/hoard |
| 12:46.02 | black187 | With Asterisk 1.8.7.1 ? |
| 12:46.10 | wdoekes2 | no |
| 12:46.18 | irroot | members are now logged off /on or rather addqueuemember / removequeuemember is sent on load and unload this probbably needs to be optional |
| 12:46.58 | wdoekes2 | (but perhaps I put my hoard lib in a standard lib dir) |
| 12:47.00 | black187 | which version do you use? |
| 12:47.12 | wdoekes2 | past tense, but 1.6.2 |
| 12:47.21 | black187 | I also putted in /lib/ folder, but no success :( |
| 12:47.32 | wdoekes2 | did you do that *before* configure? |
| 12:47.52 | black187 | yeah, before and after, once also in the middle... :) |
| 12:49.25 | black187 | hmm, I tried with other machine and compiled horde OK. Strage, the only difference is, that I use 64 bit Centos on non-horde-working machine. |
| 12:50.43 | wdoekes2 | did you compile the right (x86-64) hoard version? |
| 12:52.27 | black187 | Yeah, I did make linux-gcc-x86-64. |
| 12:53.09 | wdoekes2 | then I'm out of ideas (you did try LD_LIBRARY_PATH?) |
| 12:53.24 | black187 | yeah, I will try with older Asterisk version |
| 13:13.46 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 13:13.46 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 13:15.39 | *** part/#asterisk-dev mjordan (~mjordan@user-69-1-6-49.knology.net) |
| 13:24.39 | leifmadsen | if I made a change to the 10 branch, then merged to trunk, and forgot to make it in 1.8 as well, do I use record810 or block810 after I make the same change in 1.8? I'm thinking record810? |
| 13:25.32 | irroot | record810 |
| 13:25.45 | leifmadsen | thx |
| 13:26.04 | irroot | dat what wiki say |
| 13:26.29 | leifmadsen | makes sense |
| 13:26.59 | leifmadsen | there we go |
| 13:27.07 | irroot | oh no !!!! |
| 13:27.10 | irroot | lol |
| 13:29.25 | *** join/#asterisk-dev elred_ (~elred_@unaffiliated/elred-/x-5010831) |
| 13:48.38 | irroot | its more consistant to use time.h / timeval than time_t and time ?? |
| 13:48.45 | *** topic/#asterisk-dev by kpfleming -> Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- Tier 2 and 3.14159265 support is in #asterisk -=- Tier 4 support is in #2,000 |
| 13:49.06 | kpfleming | irroot: they serve different purposes |
| 13:49.29 | kpfleming | but yes, we have much more usage of timeval (and lots of helper macros/functions) |
| 13:53.56 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-zuqvjhalrfuyrhlk) |
| 13:53.56 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:26.43 | *** join/#asterisk-dev bklang (~bklang@2001:470:8:5de:225:ff:fe49:f8cd) |
| 14:47.01 | schmidts | is there any tool for commit messages pointing to a review? |
| 14:48.25 | wdoekes2 | how do you mean? |
| 14:49.22 | schmidts | in mantis was a commit message generator if you close an issue and i was wondering if there is something similar just for commits which point to a reviewboard page |
| 14:51.50 | mjordan | wdoekes2: voicemail config is weird. |
| 14:54.16 | leifmadsen | schmidts: unfortunately not -- it's all manual now :( I just look up on the wiki the format for commit messages |
| 14:55.05 | schmidts | leifmadsen thank you ;) then i will just copy the review link from another commit |
| 15:36.43 | *** join/#asterisk-dev putnopvut (~putnopvut@asterisk/master-of-queues/mmichelson) |
| 15:36.43 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 15:42.12 | kaii | 7j #2,000 |
| 15:42.56 | kaii | #2 and #000 are boring |
| 15:44.37 | schmidts | \o/ OEJ i got you :D |
| 15:44.38 | schmidts | pbx-03*CLI> core show channels count |
| 15:44.38 | schmidts | 12000 active channels |
| 15:44.38 | schmidts | 12000 active calls |
| 15:44.38 | schmidts | 24243 calls processed |
| 15:45.18 | Qwell | schmidts: is that with audio? |
| 15:45.42 | schmidts | theoretical yes but its only with sipp ;) |
| 15:47.35 | wdoekes2 | are tests in the test suite supposed to clean up after themselves? |
| 15:48.32 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 15:48.32 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:48.49 | wdoekes2 | (an app_voicemail test creates a mailbox and doesn't remove it..) |
| 15:55.47 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 15:55.47 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 15:58.32 | *** join/#asterisk-dev otherwiseguy (~otherwise@69.76.221.79) |
| 15:59.15 | *** join/#asterisk-dev The_Boy_Wonder (~vossel@asterisk/batman-developer/dvossel) |
| 15:59.15 | *** mode/#asterisk-dev [+o The_Boy_Wonder] by ChanServ |
| 16:00.26 | otherwiseguy | wdoekes2: In looking at one of your issues, I came across the fact that in 1.8+ when parsing a uri with an ip:port, we return both as the "domain". This makes domain matching fail for things like domain=192.168.7.7. |
| 16:04.21 | *** part/#asterisk-dev pnlarsson (~niklas@fw1.gml.g.icnet.infracom.se) |
| 16:06.47 | *** join/#asterisk-dev kpfleming (~kpfleming@asterisk/digium-director-of-software-tech/kpfleming) |
| 16:06.47 | *** mode/#asterisk-dev [+o kpfleming] by ChanServ |
| 16:11.18 | *** join/#asterisk-dev moy (~moy@216.172.42.74) |
| 16:12.31 | wdoekes2 | otherwiseguy: that might not be wrong. maybe you want domain=1.2.3.4:5678 to include the port |
| 16:19.11 | *** join/#asterisk-dev hardwire (~spencersr@cl-36.anc-01.us.sixxs.net) |
| 16:20.35 | mjordan | wdoekes2: my opinion only - yes, they should clean up after themselves. However, the tests are only run in dev mode and are not expected to execute on a production system, so there's less emphasis on their impact on the system as a whole. For example - if you wanted to test if a global setting overrode a user setting, you'd have to change the global setting - which, since they are global statics - means if anyone calls in while the t |
| 16:21.17 | mjordan | that being said, they should attempt to have as little impact as possible. Good design can mitigate the impact of the unit tests on a production system as well. |
| 16:36.47 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 16:36.48 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 16:37.49 | *** join/#asterisk-dev irroot (~irroot@197.175.50.109) |
| 16:39.00 | otherwiseguy | wdoekes2: it is certainly an undocumented change in behavior. It seems like the port would be redundant in that case. You already don't listen on ports you don't want to accept traffic from. :-) |
| 16:39.22 | irroot | congrats folks looks like a 10-rc1 tag has been cut |
| 16:40.36 | kpfleming | it has |
| 16:44.40 | leifmadsen | but it is not released |
| 16:45.07 | kpfleming | released, shmeleased... the tag in subversion is all that matters :-) |
| 16:45.24 | leifmadsen | kpfleming: until something goes wrong and it needs to be deleted for whatever reason |
| 16:45.38 | leifmadsen | I'd prefer to do it all in private until it was ready for release |
| 16:45.39 | kpfleming | when you run the mkunrelease script |
| 16:45.49 | leifmadsen | kpfleming: OF COURSE! |
| 16:45.53 | kpfleming | that could probably be arranged |
| 16:46.43 | leifmadsen | kpfleming: I like your commitment on that answer :) |
| 16:47.37 | kpfleming | maybe that's something you could bring up during our upcoming sprint to work on tools and processes :-) |
| 16:49.31 | leifmadsen | kpfleming: that sounds like an idea! |
| 16:49.50 | leifmadsen | kpfleming: any idea when we're shooting to have the meeting to discuss what we're going to work on? |
| 16:50.06 | leifmadsen | because I want to make sure I compile my Wish List to Santa |
| 16:50.39 | kpfleming | it is currently targeted for the 28th and 29th, assuming that the asterisk 10 release doesn't hold it up |
| 16:59.21 | *** join/#asterisk-dev sruffell (~Adium@asterisk/the-kernel-guy/sruffell) |
| 16:59.21 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 16:59.49 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 16:59.49 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 17:10.42 | *** join/#asterisk-dev Devon_ (~chatzilla@63.214.236.169) |
| 17:33.37 | *** join/#asterisk-dev leifmadsen (~Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 17:33.37 | *** mode/#asterisk-dev [+o leifmadsen] by ChanServ |
| 18:17.01 | *** join/#asterisk-dev otherwiseguy (~otherwise@rrcs-173-197-0-67.west.biz.rr.com) |
| 18:46.20 | *** join/#asterisk-dev moy (~moy@216.172.42.74) |
| 18:47.10 | *** join/#asterisk-dev moy (~moy@173.239.155.74) |
| 19:06.53 | irroot | 10.0.0-rc1 in production @ office and in repository for customers |
| 19:26.21 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-azvmcmkpsmrinukc) |
| 19:26.21 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 19:27.58 | leifmadsen | irroot: apologies if things break because dvossel just told me to hold off on that release because of a regression |
| 19:28.38 | irroot | leifmadsen its all good i watch the bits carefully |
| 19:28.44 | leifmadsen | :) |
| 19:29.18 | irroot | easy to test and fix ... been running 10 SVN at office for long time have some early adopters too |
| 19:32.54 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-azvmcmkpsmrinukc) |
| 19:38.03 | *** join/#asterisk-dev snuff-home (~snuffy@203-219-29-182.tpgi.com.au) |
| 19:38.03 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 19:38.47 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-azvmcmkpsmrinukc) |
| 19:38.48 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 20:04.25 | leifmadsen | irroot: tell them we like them :) |
| 20:05.20 | irroot | leifmadsen will do ... so far so good but only smaller sites |
| 20:05.35 | leifmadsen | ya, still good to get the small sites stable as a base |
| 20:07.25 | irroot | leifmadsen indeed and most the time they love the extra attention we give them when rolling out new bits |
| 20:07.47 | leifmadsen | who doesn't?! |
| 20:07.57 | leifmadsen | ok the coffee is kicking in |
| 20:07.58 | irroot | the guys are always checking for core dumps and will call them for feedback |
| 20:08.22 | irroot | leifmadsen epinefirin has bigger kick :P |
| 20:08.29 | leifmadsen | I would love to know how projects like Fedora have a crash analyzer and can associate backtraces with open issues automatically |
| 20:08.50 | leifmadsen | maybe some day I'll google and find out |
| 20:11.01 | irroot | leifmadsen yeah that is some cool shit .... |
| 20:11.55 | irroot | id guess they link it to a version and create a pat as they binary |
| 20:12.03 | irroot | there not much change |
| 20:12.33 | irroot | problem is my build of asterisk from today may differ tommorow if i tweak compiler options |
| 20:13.17 | irroot | another evil i strip the bin files before shiping so i need to download the raw core and then get a BT on the dev box |
| 20:18.36 | irroot | there is a pesky problem with task processor if a module unloads with tasks pending it does cause mischief |
| 20:19.08 | kpfleming | don't unload modules |
| 20:19.35 | irroot | kpfleming indeed not ideal to do so but its one of the stress tests i perform |
| 20:19.42 | kpfleming | seriously... there are probably dozens of places in asterisk where module refs aren't handled well enough to make a module unload be deferred until it is safe |
| 20:20.04 | kpfleming | until you know the code has been written to take than into account, it's not a stress test, it's a functional test |
| 20:20.16 | irroot | ill try get app_queue little better in this regard |
| 20:21.41 | *** join/#asterisk-dev Netgeeks (~chris@gw1.netgeeks.net) |
| 20:21.56 | irroot | with the messing arround with containers and objects i have been doing there is a order to what must be unloaded but the tps API is where its breaking |
| 20:24.58 | irroot | intresting this time it was AMI that bombed it |
| 20:30.10 | *** join/#asterisk-dev oej (~oej@ns.webway.se) |
| 20:30.10 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 20:42.31 | *** join/#asterisk-dev oej (~oej@ns.webway.se) |
| 20:42.31 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 22:18.34 | *** join/#asterisk-dev tzafrir (~tzafrir@212.179.75.202) |
| 22:18.34 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 22:20.11 | *** join/#asterisk-dev irroot (~irroot@197.172.248.230) |
| 22:30.56 | *** join/#asterisk-dev mnicholson (~mnicholso@nat/digium/x-mwxqzxbfdmyrvsxk) |
| 22:30.56 | *** mode/#asterisk-dev [+o mnicholson] by ChanServ |
| 22:31.12 | irroot | mnicholson evening |
| 22:31.27 | mnicholson | irroot, brb |
| 22:32.01 | *** join/#asterisk-dev mnicholson (~mnicholso@nat/digium/x-jlhmqhwkgixbgnss) |
| 22:32.01 | *** mode/#asterisk-dev [+o mnicholson] by ChanServ |
| 22:35.20 | mnicholson | irroot: ok, back |
| 22:35.38 | irroot | hi there just greeting |
| 22:35.50 | mnicholson | irroot: oh, greetings :) |
| 22:36.32 | irroot | still working on that issue with app_queue and found a novel way of "bubble" sort using ao2_callback_data to find rules now in containers |
| 22:36.56 | mnicholson | hmm ok |
| 22:36.59 | mnicholson | cool |
| 22:37.59 | irroot | http://pastebin.com/97nxvZDj if you want to look at it |
| 22:38.29 | irroot | returns best match with ref held |
| 22:39.03 | irroot | ao2_callback_data(qe->rules->rules, OBJ_NODATA | OBJ_MULTIPLE, get_best_rule_cb, &time, &qe->pr); |
| 22:44.24 | *** join/#asterisk-dev jrose_atDigium (~jon@nat/digium/x-wnprsiohpjmqjzfs) |
| 22:44.49 | irroot | jrose_atDigium o/ |
| 22:48.09 | jrose_atDigium | o/? |
| 22:48.54 | irroot | o/ <- wave |
| 22:49.07 | jrose_atDigium | Oh, wasn't familiar with that one. |
| 22:50.44 | irroot | shame |
| 22:50.49 | irroot | oops wrong window |
| 22:54.32 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-218-155-145.cablep.bezeqint.net) |
| 22:54.32 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 23:02.28 | *** join/#asterisk-dev mjordan (~mjordan@nat/digium/x-catefqwfgqdqvpfj) |
| 23:02.28 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 23:16.57 | *** part/#asterisk-dev mjordan (~mjordan@nat/digium/x-catefqwfgqdqvpfj) |
| 23:59.00 | *** join/#asterisk-dev michael-i (~anonymous@204.11.230.58.static.etheric.net) |
