| 00:00.32 | *** join/#asterisk-dev docelmo (n=vircuser@c-76-99-153-242.hsd1.de.comcast.net) |
| 00:01.02 | docelmo | does anyone know what module has to be loaded to gain access to this function? ast_bridge_call |
| 00:01.13 | docelmo | I have been all over doxygen and I cant find anything on it. |
| 00:02.28 | elixer | res_features.so |
| 00:04.32 | elixer | egrep -ir "^[[:space:]]*int ast_bridge_call" * |
| 00:04.32 | *** join/#asterisk-dev denon (n=denon@208.122.43.201) |
| 00:34.43 | *** join/#asterisk-dev moy (n=moyhu@189.169.45.82) |
| 01:42.24 | *** join/#asterisk-dev outtolunc (n=me@c-67-174-216-60.hsd1.ca.comcast.net) |
| 02:07.21 | *** join/#asterisk-dev wunderkin (n=wunderki@ip72-223-86-126.ph.ph.cox.net) |
| 02:52.05 | *** part/#asterisk-dev twilson (n=twilson@CPE-65-30-30-20.kc.res.rr.com) |
| 03:18.43 | *** join/#asterisk-dev denon (n=denon@208.122.43.201) |
| 03:26.55 | *** join/#asterisk-dev netson99 (n=netson99@202.153.241.206) |
| 03:50.02 | *** join/#asterisk-dev outtolunc (n=me@c-67-174-216-60.hsd1.ca.comcast.net) |
| 04:44.19 | *** join/#asterisk-dev tuxd00d (n=tuxinato@128.187.174.112) |
| 05:17.00 | *** part/#asterisk-dev mog (n=mog@c-71-207-200-130.hsd1.al.comcast.net) |
| 05:46.58 | *** join/#asterisk-dev dlynes_laptop (n=dlynes@d154-20-9-152.bchsia.telus.net) |
| 05:51.25 | *** join/#asterisk-dev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 06:06.59 | *** join/#asterisk-dev Corydon76-home (i=six@pdpc/supporter/sustaining/Corydon76-home) |
| 06:06.59 | *** mode/#asterisk-dev [+o Corydon76-home] by ChanServ |
| 06:25.25 | *** join/#asterisk-dev pnlarsson (n=pnlarsso@c83-248-12-187.bredband.comhem.se) |
| 06:25.31 | jmls_sleeeeeeep | russellb: |
| 06:25.33 | jmls_sleeeeeeep | ping |
| 06:26.09 | jmls | Corydon76-home: ping as well. you may be interested |
| 06:35.13 | Corydon76-dig | Eh? |
| 06:35.27 | jmls | that problem I was having last night I can't reproduce any more. |
| 06:36.00 | jmls | with Progress |
| 06:36.12 | jmls | strange backtraces all over the place |
| 06:36.22 | jmls | you helped russellb diagnose the problem |
| 06:36.22 | Corydon76-dig | with the change I suggested or without? |
| 06:36.36 | jmls | without. |
| 06:36.37 | Corydon76-dig | and did you upgrade anything else? |
| 06:36.46 | jmls | no, but I did stop using views |
| 06:36.57 | Corydon76-dig | Ah |
| 06:37.05 | jmls | Ah ? |
| 06:37.07 | Corydon76-dig | Yeah, DEFINITELY a problem with the Progress driver, then |
| 06:37.40 | jmls | could the realtime engine be trying to update the view ? |
| 06:37.50 | Corydon76-dig | In fact, given that it showed up with views only, it may not even be the fault of the ODBC author, but one level above that |
| 06:38.13 | Corydon76-dig | i.e. the core Progress client library |
| 06:38.48 | Corydon76-dig | You should do them the courtesy of reporting this issue to them... I'm sure they'd like to know |
| 06:38.55 | jmls | I have done |
| 06:39.01 | jmls | well, I've managed to run nearly 1000 calls through the test system wthout it failing |
| 06:39.13 | jmls | previously, it would crash on or around 50 |
| 06:39.33 | jmls | I've also stopped using MixMonitor, so I'll add that back into the mix |
| 06:39.46 | jmls | ;) |
| 06:40.30 | Corydon76-dig | Cool |
| 06:41.48 | snuff-work | hehe |
| 06:41.51 | snuff-work | go mixmon |
| 06:42.05 | snuff-work | should almost use audiohooks now.. |
| 06:42.15 | snuff-work | or is that only a trunk feature i think |
| 06:42.29 | snuff-work | audiohooks is cleaner than mixmon correct? |
| 06:55.09 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@62.90.10.53) |
| 06:55.09 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 07:12.10 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 07:50.40 | *** join/#asterisk-dev nenadr (n=nenadr@rakac.absolutok.com) |
| 07:51.40 | *** part/#asterisk-dev nenadr (n=nenadr@rakac.absolutok.com) |
| 08:13.36 | *** join/#asterisk-dev dlynes_ (n=dlynes@d154-20-9-152.bchsia.telus.net) |
| 08:43.01 | *** join/#asterisk-dev nenadr (n=nenadr@rakac.absolutok.com) |
| 08:50.27 | *** join/#asterisk-dev key2 (n=Ritual@193.33.36.20) |
| 09:01.49 | *** join/#asterisk-dev hermuli (n=Eladamri@xdsl-83-145-207-63.nebulazone.fi) |
| 09:26.21 | *** join/#asterisk-dev qdk (n=qdk@213.237.44.34) |
| 09:48.21 | *** join/#asterisk-dev yannj_fr (n=yannj@chilli.esiee.fr) |
| 09:49.15 | *** join/#asterisk-dev michael-i (n=michael-@Ldb80.l.pppool.de) |
| 10:04.00 | *** join/#asterisk-dev jsmith (n=jsmith@000-181-995.area3.spcsdns.net) |
| 10:08.32 | *** part/#asterisk-dev nenadr (n=nenadr@rakac.absolutok.com) |
| 10:18.00 | *** join/#asterisk-dev nenadr (n=nenadr@rakac.absolutok.com) |
| 10:30.54 | *** part/#asterisk-dev nenadr (n=nenadr@rakac.absolutok.com) |
| 11:58.17 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 12:25.40 | *** join/#asterisk-dev Yourname`` (n=IM@unaffiliated/yourname/x-837320) |
| 12:25.40 | Yourname`` | Hi russellb, just wanted to thank you for your help from last night. About the new svn revision. It worked very well without a crash! :) |
| 12:25.45 | *** part/#asterisk-dev Yourname`` (n=IM@unaffiliated/yourname/x-837320) |
| 12:29.13 | *** join/#asterisk-dev yannj_fr (n=yannj@chilli.esiee.fr) |
| 13:25.13 | *** join/#asterisk-dev jsmith (n=jsmith@000-181-995.area3.spcsdns.net) |
| 13:26.39 | *** join/#asterisk-dev mog (n=mog@c-71-207-200-130.hsd1.al.comcast.net) |
| 13:26.39 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 13:54.37 | *** join/#asterisk-dev mbranca (n=matteo@host-210-mi.linuxserver.it) |
| 13:55.10 | *** join/#asterisk-dev dioedu (n=dioedu@c915230f.virtua.com.br) |
| 14:01.08 | *** join/#asterisk-dev moy (i=moyhu@nat/ibm/x-c9d206419eee3179) |
| 14:03.16 | *** join/#asterisk-dev Victor_Yure (n=aaaa@esp5.deibotoch.com.br) |
| 14:04.15 | *** join/#asterisk-dev anonymouz666 (n=anonymou@189.25.37.205) |
| 14:09.21 | *** join/#asterisk-dev jcmoore (n=jcmoore@unaffiliated/tgrman) |
| 14:18.14 | Juggie | russellb: ping |
| 14:19.46 | jsmith | Juggie: FWIW, the CDR stuff looks like it's coming along nicely |
| 14:21.00 | Juggie | ya... |
| 14:21.12 | Juggie | testing the res agi fix i posted now |
| 14:21.26 | Juggie | we have it up to 86 sip calls so 172 channels/agi connections |
| 14:22.11 | Juggie | russellb, i've been busy, but i've been lucky enough to get some help. |
| 14:22.32 | Juggie | there are a couple of issues we have to resolve, the odbc wrapper needs to provide more detailed information about what went wrong. |
| 14:23.10 | Juggie | but me and Corydon76 kind of bumped on that one and didnt come to a conclusion |
| 14:23.26 | Juggie | i need to know if the error is fatal or not, but all the odbc wrapper provides is if its an error, no additional informatino |
| 14:23.31 | Juggie | *information |
| 14:23.50 | Juggie | btw, i'm 27 today :) |
| 14:23.51 | jsmith | Juggie: It doesn't return an error number? |
| 14:23.56 | Juggie | jsmith, no. |
| 14:24.08 | jsmith | ~sing to Juggie |
| 14:24.16 | jsmith | ~birthday Juggie |
| 14:24.17 | jbot | ACTION Happy Birthday to you, Happy birthday to you!, HAPPY BIRTHDAY DEEEAARR Juggie, Happy Birthday toooo yoooou! |
| 14:24.33 | jsmith | Isn't jbot great? |
| 14:24.49 | Juggie | heh, 86x2 sip channels/ w/ 2 agi connections each is taking 5600kb/s total. |
| 14:25.12 | Juggie | silly sip, so much overhead. |
| 14:36.35 | Juggie | elixer, are you going to *-con? |
| 14:36.40 | elixer | Juggie: i am not |
| 14:36.45 | elixer | maybe next year |
| 14:36.57 | Juggie | bah, we just broke res_agi @ 88 concurrent calls (w/ agi) |
| 14:37.50 | jmls | whoa. Somethings _really_ wrong with realtime members and queues |
| 14:38.18 | jsmith | jmls: Yup... what's new? |
| 14:38.24 | jmls | lol |
| 14:38.33 | jmls | that's -not- funny .. |
| 14:38.47 | jsmith | No, I'm not trying to be the class clown, I'm being serious. |
| 14:39.01 | jmls | got a debug output from a dialplan that shows 2 calls being put through to the same agent at the same time |
| 14:39.03 | Juggie | voicemail and members/queues are the most broken things. |
| 14:39.08 | jsmith | There have been a number of issues with realtime members and queues for a while now |
| 14:39.28 | jsmith | jmls: Problems keeping track of which members are in-use? |
| 14:39.32 | jmls | heh. Chuck in Local channels as well |
| 14:39.57 | jsmith | Naw, local channels are mostly working now, except for the channel state stuff that file and I were discussing last week |
| 14:40.03 | jsmith | (but that should be hopefully fixed in trunk) |
| 14:40.37 | jmls | http://pastebin.ca/701940 |
| 14:41.51 | jmls | 2 different channels, same time, same member ... |
| 14:42.14 | jmls | this would explain a lot of things that are happening to me ;) |
| 14:42.25 | jsmith | Yeah, and you're using chan_agent? |
| 14:42.25 | elixer | so i've got this core file here |
| 14:42.33 | jmls | jsmith: nope |
| 14:42.39 | jmls | certainly not :) |
| 14:42.44 | elixer | in gdb, how do i get to where the segfault actually occurred? |
| 14:43.19 | jsmith | elixer: "bt" , then look for line 0 |
| 14:43.29 | jmls | jsmith: AddQueueMember and Local/1234@Agentq/n for the interface name |
| 14:43.45 | jmls | then 1234 dials SIP/1234 |
| 14:44.12 | jsmith | Ahh... then you should be pretty good. |
| 14:44.33 | jsmith | Your extension 1234@Agentq uses GROUP() to see if it's already busy, right? |
| 14:45.12 | Juggie | elixer, the problem is usually at the top |
| 14:45.14 | jmls | in the stress test scenario, I don't actually do the dial, just answer the call and play moh |
| 14:45.49 | elixer | well i've made changes to main/cdr.c and cdr/cdr_custom.c and the segfault looks to be happening in logger_print_normal |
| 14:45.52 | elixer | i don't really get it. |
| 14:46.30 | jmls | so i get agent-34 playing music on 2 different channels to 2 different calls. Not a good thing if I *were* to pass the call on. |
| 14:46.38 | Juggie | elixer, did you try a make clean |
| 14:46.52 | Juggie | some times it just gets unhappy after so many changes |
| 14:47.03 | elixer | yeah i did. will try again. |
| 14:47.04 | Juggie | i've had that happen before, chased a bug for a while, then a make clean, make install fixed it. |
| 14:47.24 | Juggie | elixer, you could be passing something weird to the logger |
| 14:47.25 | Juggie | ? |
| 14:47.39 | Juggie | make sure you are compiling in debug mode if you want gdb to make any sense |
| 14:47.52 | jmls | jsmith: I don't understand the GROUP thingie : "Gets or Sets the channel group" |
| 14:47.59 | elixer | yeah, already doing that. DONT_OPTIMIZE and MALLOC_DEBUG are enabled. |
| 14:48.47 | jmls | however, as you can see from the pastebin, I have 2 different channels for my Local/Agent-34 --> Local/Agent-34@agentq-2e2e and Local/Agent-34@agentq-1c2f |
| 14:49.12 | Juggie | Local channels and agents/members dont get along too too well. |
| 14:49.57 | jmls | Juggie: but it's the only way to get to the dialplan before the call is passed to the member |
| 14:50.03 | file | sure they do |
| 14:50.11 | jsmith | jmls: The whole idea with GROUP() is that you set the current channel as part of a group when you actually Dial() into the SIP device, |
| 14:50.31 | jsmith | jmls: Then, you can also check the GROUP() to see if the channel is already in use |
| 14:50.57 | jmls | so, you add the channel -to- the group |
| 14:51.12 | jmls | and then can check the number of channels -in- the group |
| 14:52.11 | *** part/#asterisk-dev mog (n=mog@c-71-207-200-130.hsd1.al.comcast.net) |
| 14:52.12 | jmls | and when the channel hangs up, it is automatically removed from the group ? |
| 14:53.05 | jsmith | You've got it. |
| 14:53.16 | jsmith | Here, let me see if I can find something to pastebin that'll make more sense |
| 14:53.31 | Juggie | ot question: how can i see the speed my linux network card is connected at, eg 100mbit, 1000mbit, duplex, and so on |
| 14:53.50 | jsmith | Juggie: mii-tool will show it on *some* cards |
| 14:54.28 | jsmith | Juggie: Or try "ethtool" |
| 14:55.32 | *** join/#asterisk-dev miguel3239 (n=elguerom@ns1.nashuacs.com) |
| 14:56.09 | *** join/#asterisk-dev denon (n=denon@208.122.43.201) |
| 14:57.11 | *** join/#asterisk-dev mog (i=mog@nat/digium/x-054a1723a9e9839c) |
| 14:57.11 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 14:59.03 | jsmith | jmls: Try this: http://asterisk.pastebin.ca/701963 |
| 14:59.18 | jsmith | jmls: It's not guaranteed to be perfect, but it should hopefully help you on your way |
| 14:59.32 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 14:59.39 | jsmith | jmls: And it's fairly well commented (with lots of calls to Verbose) |
| 15:00.53 | jmls | many thanks. Will look into it. |
| 15:06.52 | *** join/#asterisk-dev anthm (n=anthm@CPE-72-131-113-50.wi.res.rr.com) |
| 15:06.52 | *** mode/#asterisk-dev [+o anthm] by ChanServ |
| 15:07.24 | jmls | jsmith: I can't see where you are setting the group for the channel .. |
| 15:09.29 | jsmith | jmls: No problem, I'm open :-) |
| 15:09.49 | jsmith | jmls: It's the magic of the OUTBOUND_GROUP channel variable (or OUTBOUND_GROUP_ONCE) |
| 15:10.01 | jsmith | jmls: That automatically sets the group at the time that Dial() is called |
| 15:10.07 | jsmith | Sorry, I should have explained that piece better |
| 15:10.23 | *** join/#asterisk-dev putnopvut (i=putnopvu@nat/digium/x-76a70b6438e4bc37) |
| 15:10.24 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 15:11.29 | Juggie | elixer: ping |
| 15:11.33 | elixer | pong |
| 15:11.37 | elixer | sup?/ |
| 15:11.44 | Juggie | did you figure out your crash? |
| 15:11.52 | elixer | well, it stopped crashing. |
| 15:11.54 | elixer | not 100% why |
| 15:11.56 | elixer | err |
| 15:11.58 | elixer | not 100% sure why |
| 15:12.05 | Juggie | after a clean? |
| 15:12.16 | elixer | yeah. but i did the same thing earlier. distclean even. |
| 15:12.21 | Juggie | weird. |
| 15:12.23 | elixer | bizarre. i do have a question though. |
| 15:12.30 | elixer | and this might demonstrate my ignorance |
| 15:12.32 | Juggie | good cuz i have one too, you go first. |
| 15:12.43 | elixer | should this line bother me: if (!(i = ast_calloc(1, sizeof(*i)))) |
| 15:13.07 | elixer | the '*i' bothers me, and i am not sure why |
| 15:13.20 | Juggie | where is that? |
| 15:13.25 | elixer | main/cdr.c |
| 15:13.31 | *** join/#asterisk-dev miguel3239 (n=elguerom@ns1.nashuacs.com) |
| 15:14.00 | elixer | sizeof is evaluated at runtime, yeah? |
| 15:14.38 | russellb | no it is not |
| 15:14.46 | elixer | ah, ok |
| 15:14.56 | elixer | then *i shouldn't bother me |
| 15:15.30 | russellb | it's just the lazy way, instead of sizeof(typeof(i)) |
| 15:15.32 | Juggie | russellb, that fgets w/ fastagi & 1.4 seem fine here after some testing. festr posted that it worked fine for him as well. |
| 15:15.44 | russellb | Juggie: cool |
| 15:16.30 | elixer | Juggie: you had a ? for me? |
| 15:16.34 | Juggie | elixer, as for ast_odbc_prepare_and_execute |
| 15:16.38 | elixer | aye |
| 15:17.11 | Juggie | did you come up with a solution to that little wonder |
| 15:17.39 | elixer | i did not |
| 15:17.44 | elixer | that is next on my list |
| 15:18.08 | elixer | i have a patch headed your way to allow multiple cdr_custom sinks |
| 15:18.34 | Juggie | we could modify it to allow to pass a detailed information variable by reference |
| 15:18.47 | elixer | thats along the lines of what i was thinking |
| 15:19.40 | Juggie | we cant really use stmt because thats a 'SQLHSTMT' return type |
| 15:19.51 | elixer | which is just a long |
| 15:19.54 | Juggie | oh ok. |
| 15:20.33 | Juggie | well in theory we *COULD* modify the return, but that probally isnt appropriate. |
| 15:21.05 | Juggie | passing back detailed information by ref is probally best, then anyone who doesnt care passes in a null and no code change required. |
| 15:22.07 | elixer | yeah, i'm down with that |
| 15:24.22 | *** join/#asterisk-dev mog (i=mog@nat/digium/x-1f2831cb2fa46107) |
| 15:24.22 | *** mode/#asterisk-dev [+o mog] by ChanServ |
| 15:24.28 | Juggie | ok, i guess we need to define some statuses then in res_odbc.h? |
| 15:25.56 | elixer | yeah, but i can't work on it right now |
| 15:26.09 | elixer | i have been pretty much 100% neglecting my actual job for the last few days |
| 15:26.14 | elixer | i'd like to stay employed |
| 15:26.15 | elixer | :) |
| 15:26.15 | Juggie | k, there look to be 2-3 functions that would need that extension. |
| 15:26.21 | Juggie | hehe... me too kinda... |
| 15:26.41 | Juggie | but i work for the gov so i'd neeed to do no work for like 6 months |
| 15:26.49 | Juggie | and then i might get a written warning |
| 15:27.21 | elixer | there's no "kinda" about it. i could paste the code i have written for work in the last week into this channel and wouldn't even get kicked for flooding :-) |
| 15:27.52 | elixer | sorry.. no "kinda" about it for /me/. |
| 15:28.21 | file | Asterisk is addictive |
| 15:28.22 | file | be afraid. |
| 15:28.30 | file | (yes, we do have counselling classes available) |
| 15:29.04 | Juggie | elixer: pm |
| 15:33.36 | elixer | file: yeah, its sucking my will to live. i might have to uninstall the ssh client from my pc entirely. |
| 15:34.03 | elixer | otherwise i will code all damn day |
| 15:34.14 | elixer | s/code/code on asterisk/ |
| 15:34.31 | russellb | elixer: that's fine, you can do that :) |
| 15:34.49 | russellb | elixer: but be careful, most people that do that end up getting hired :) |
| 15:35.41 | elixer | russellb: i wouldn't be able to survive in alabama... can i work at the remote office in baltimore (aka my apartment)? ;-) |
| 15:35.58 | Juggie | ask file how he does it |
| 15:35.58 | file | good morning baltimore! |
| 15:36.10 | Juggie | somehow he works from his apartment in canada |
| 15:36.24 | jsmith | elixer: I work from the Digium offices in Virginia... is that close enough? |
| 15:36.32 | file | it involves walking to my office, sitting down, and working |
| 15:37.16 | Juggie | digium has offices in virginia? |
| 15:37.30 | putnopvut | Yeah, at jsmith's house! |
| 15:37.35 | Corydon76-dig | Juggie: and here, in Tennessee |
| 15:37.35 | Juggie | heh |
| 15:37.40 | jmls | putnopvut: got another queue problem for you |
| 15:37.47 | putnopvut | jmls: YES!! |
| 15:37.49 | jmls | http://pastebin.ca/701940 |
| 15:38.01 | jmls | 2 calls to the same queue member at the same time ! |
| 15:38.15 | Corydon76-dig | Juggie: remember, no income tax in Tennessee |
| 15:38.18 | putnopvut | How did you manage that? |
| 15:38.25 | denon | nothing else in TN either :) |
| 15:38.26 | jmls | *I* didn't ! |
| 15:38.30 | jmls | app_queue did ! |
| 15:38.43 | jsmith | putnopvut: Because app_queue doesn't mind doing that :-) |
| 15:38.54 | jsmith | (especially for Local channels) |
| 15:38.55 | russellb | elixer: we certainly prefer relocation, but for people that have proven their ability to contribute, we consider remote work |
| 15:38.57 | jmls | is this because of autofill=yes |
| 15:39.04 | jsmith | jmls: Nope |
| 15:39.10 | Juggie | russellb, you sound like a recruiter :) |
| 15:39.12 | russellb | elixer: we have 3 full time swdev employees remote |
| 15:39.19 | putnopvut | russellb: 4 |
| 15:39.20 | russellb | Juggie: i am in part :) |
| 15:39.25 | jsmith | russellb is a recruiter! |
| 15:39.31 | denon | truth's out :) |
| 15:39.33 | russellb | we need more people, so yeah, i try to find them |
| 15:39.36 | denon | now nobody will trust him :) |
| 15:39.38 | Juggie | i only wish part time work was avail. |
| 15:39.51 | jcmoore | Uncle Asterisk wants YOU! |
| 15:40.27 | putnopvut | jmls: This happened when running your stress test? |
| 15:40.33 | putnopvut | And what happened as a result? |
| 15:40.48 | jmls | my virtual agent got one call, and then another |
| 15:41.00 | jmls | (when I say call, I mean jabber message) |
| 15:41.08 | jmls | because they are virtual ;) |
| 15:41.15 | putnopvut | Well, I can't say for sure, but this sounds like a textbook race condition. |
| 15:41.44 | jmls | 10 agents on an inbound queue, up to 20 calls waiting |
| 15:41.50 | putnopvut | jmls: How reproduceable is this? |
| 15:41.58 | jmls | every time |
| 15:42.06 | jmls | perhaps 5 out of 200 calls |
| 15:42.11 | elixer | russellb: well let me prove myself and get fired from this job first. then we'll talk :-) |
| 15:42.32 | russellb | ;) |
| 15:42.42 | russellb | http://www.vote756.com/marcecko/ --- everyone vote to brand it with an asterisk!! |
| 15:42.58 | Juggie | elixer, its not vs6 i hope. |
| 15:43.01 | putnopvut | jmls: If you don't mind, could you open a bug on this? |
| 15:43.07 | elixer | Juggie: c# |
| 15:43.14 | Juggie | elixer, eugh! |
| 15:43.23 | elixer | hater |
| 15:43.37 | putnopvut | jmls: Also, if you reproduce it, could you attach a full console output on the bug report as well? |
| 15:43.38 | elixer | :) |
| 15:43.47 | jmls | no problem. what do you want me to say ? |
| 15:43.50 | jmls | putting 2 calls to the same member at the same time is bad, bad, bad ! BAD asterisk |
| 15:43.51 | elixer | come on, man. all the cool kids are doing it! |
| 15:43.52 | putnopvut | jmls: Thanks for finding these problems. It's really helpful. |
| 15:44.10 | jsmith | putnopvut: I'm not sure it's a bug... |
| 15:44.18 | jmls | WHAT ?? |
| 15:44.42 | putnopvut | jsmith: I'm all ears. |
| 15:45.05 | jsmith | putnopvut: See, to be honest, I think it was done that way on purpose... |
| 15:45.44 | jsmith | ... so that if you added Local channels as members of a queue, that they'd get called every time, and your dialplan could decide whether to return busy or dial a device, etc.. |
| 15:46.06 | jsmith | Shoot... I've been using it that way for a year and a half now |
| 15:46.20 | putnopvut | I see. A local channel could potentially call one of many extensions. |
| 15:46.40 | jsmith | Exactly... a Local channel can do just about anything |
| 15:46.41 | putnopvut | So if you "exploit" this, you could have multiple people be a single queue member. |
| 15:46.47 | jmls | well, it could, but what's the point ? |
| 15:46.50 | jsmith | putnopvut: Exactly! |
| 15:46.54 | jmls | why have a queue at all ? |
| 15:47.04 | jsmith | jmls: The point is that there's a very good way of getting around the problem, which I've tried to show you |
| 15:47.25 | jmls | jsmith: but you still called it "a problem" |
| 15:47.30 | jsmith | jmls: Some people like Asterisk's *queueing*, just not the *distribution* piece |
| 15:47.45 | jsmith | jmls: Sure, it's a problem, just not a bug |
| 15:47.58 | jmls | lol |
| 15:48.40 | jmls | I would say then that the queue should have an option that allows for "single-use" members ;) |
| 15:48.58 | jmls | then we have both have what we want ! |
| 15:48.59 | jsmith | jmls: In this case, it's not a problem with queues at all... it's the device state stuff for Local channels. |
| 15:49.22 | jsmith | jmls: It's up to the device state to say "Hey, this channel is busy... move on and try the next member" |
| 15:49.27 | putnopvut | I was about to say, I don't believe there are any direct references to local channels in the queue code. |
| 15:49.41 | jsmith | In the case of chan_local, it doesn't say "I'm busy" if there's already a local channel of the same name |
| 15:49.56 | jsmith | Now, I think that may have been changed in trunk... we should really ask file, as he was looking at it last week |
| 15:49.59 | jsmith | file: Care to comment? |
| 15:50.05 | file | it has changed in trunk, aye |
| 15:51.18 | jsmith | So, how does it now work in trunk? |
| 15:51.43 | file | it looks at all the Local channels up to determine whether it is in use or not |
| 15:51.58 | file | but if the Local channels get optimized away, then that test is useless |
| 15:54.29 | jsmith | Right... let's assume that we're using Local/12345@blah/n (note the /n on the end) |
| 15:54.35 | jsmith | In that case, what would it do? |
| 15:55.21 | file | return inuse if inuse I think, but I have not actually tested it |
| 15:56.04 | jsmith | Is that change documented in UPGRADE.txt? |
| 15:56.12 | jsmith | Because it's gonna break a lot of my systems if that's the case. |
| 15:56.34 | jsmith | Do we need another flag on the end of the local channel to specify "only one at a time"? |
| 15:56.55 | jsmith | Locall/12345@blah/nm (where the "m" stands for "multiple")? |
| 15:57.14 | jsmith | Because I can see it being useful sometimes |
| 15:57.31 | jsmith | (especially if we're going and changing the current behavior) |
| 16:02.28 | russellb | heh, i broke the build of chan_zap in trunk yesterday |
| 16:02.31 | russellb | and nobody has noticed yet |
| 16:03.20 | file | russellb: the murf noticed |
| 16:03.44 | russellb | but did not yell at me? |
| 16:03.47 | russellb | that's just too nice |
| 16:04.41 | codefreeze | I did see that you added those refs... thought, well, he must have missed checking in the main/ file... |
| 16:04.58 | codefreeze | he'll get around to it |
| 16:05.16 | russellb | :) |
| 16:05.18 | putnopvut | codefreeze: you have to ride his tail to make sure that little screw-up does things right ;) |
| 16:05.32 | Juggie | codefreeze, have a minute? |
| 16:05.44 | codefreeze | Juggie: for you, anytime! |
| 16:05.45 | russellb | putnopvut: orly! |
| 16:05.56 | *** mode/#asterisk-dev [-o putnopvut] by russellb |
| 16:06.09 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 16:06.23 | *** topic/#asterisk-dev by russellb -> Asterisk Development Discussion -=- http://www.asterisk.org/developers -=- IAX2/guest@misery.digium.com/asterisk-dev -=- Asterisk trunk: res_telepathy release expected October 2007 -=- Tier 2 and Tier 3.14159 support is in #asterisk -=- ast_atomic_fetchadd_int(&developers_engaged, +1); -=- This is still not tier 2 support. Actually, this is not a support channel at all! -=- Happy Birthday Juggie! |
| 16:06.38 | codefreeze | putnopvut: note that russellb is still the fastest kick-banner in the west! |
| 16:06.40 | putnopvut | Happy birthday! |
| 16:06.54 | Juggie | codefreeze, ast_prepare_and_execute doesnt tell me enough information aboout why a sql statement failed, only that it did. |
| 16:07.06 | *** mode/#asterisk-dev [+o Juggie] by russellb |
| 16:07.10 | russellb | that's your present, hehe |
| 16:07.25 | Juggie | i need to extend it, mine and elixir's initial thought was to pass in a variable by reference, and fill it with the full detailed status. |
| 16:07.25 | file | the op count is unbalanced now! |
| 16:07.30 | *** join/#asterisk-dev CunningPike (n=arodgers@204.239.12.183) |
| 16:07.32 | Juggie | just wondering what you thought. |
| 16:07.42 | codefreeze | Juggie: what should we do? |
| 16:07.53 | Juggie | codefreeze, see my previous line |
| 16:08.05 | Qwell | codefreeze: hey, can you briefly explain the difference between cdr->channel and cdr->dstchannel? |
| 16:08.09 | Qwell | (when you have a minute) |
| 16:08.15 | Corydon76-dig | Juggie: I already told you what you should do |
| 16:08.39 | jsmith | file: And it's not prime |
| 16:08.40 | codefreeze | Juggie: missed that. |
| 16:08.40 | Juggie | Corydon76-dig, i am aware, but that doesnt fit within he cdr model me and codefreeze have already hashed out. |
| 16:08.59 | codefreeze | Qwell: From and to channels involved in a bridge, basically |
| 16:09.04 | *** join/#asterisk-dev tsurko (n=tsurko@78.90.100.214) |
| 16:09.06 | Juggie | *the |
| 16:09.21 | Corydon76-dig | You're going to send us down a road of continually adding more and more criteria to check, for each possible temporary failure. Don't do that. |
| 16:09.48 | Qwell | codefreeze: so, there is a patch on the bug tracker to update the cdr in queues.. agents already has this ability.. agents set cdr->channel, but the queues patch sets cdr->dstchannel. Is there any obvious reason you could think of that he might have done that? |
| 16:09.51 | Corydon76-dig | It's a failure. Treat it ALWAYS as temporary, unless it continues to fail over a certain defined time period |
| 16:09.58 | Juggie | it is not acceptable to keep scheduling things in the future, nor is it acceptable to keep hammering a system with a command that will never work. |
| 16:10.14 | putnopvut | Qwell: the perspective is different |
| 16:10.21 | Corydon76-dig | Juggie: there is simply NO way for you to know that in advance |
| 16:10.31 | putnopvut | app_queue views the caller as the channel and the queue member as the peer. |
| 16:10.37 | Juggie | Corydon76-dig, what if i simply want to know if it was a connection error or not |
| 16:10.39 | Corydon76-dig | Juggie: not in the general case |
| 16:10.46 | putnopvut | chan_agent is the opposite. |
| 16:10.47 | Juggie | i cannot even know that information, thats not enough |
| 16:10.52 | Corydon76-dig | Juggie: that's not a good solution |
| 16:10.54 | Qwell | putnopvut: okay, makes sense |
| 16:11.19 | Corydon76-dig | There are lots of transitory errors, and people are going to demand that we handle ALL of them. |
| 16:11.25 | Juggie | Corydon76-dig, theres no reason i cant know why something failed. |
| 16:11.56 | Corydon76-dig | Juggie: absolutely, but the reason is not important programatically |
| 16:12.07 | Corydon76-dig | Juggie: the reason is only useful to the admin |
| 16:12.43 | Corydon76-dig | So you print an error and let the admin fix it, if he feels like he shouldn't lose any CDRs |
| 16:12.56 | codefreeze | Juggie: Corydon76-dig: Do we at least log the reason? |
| 16:13.10 | Corydon76-dig | "Failing because of ERROR, will retry for the next 20 minutes" |
| 16:13.17 | Corydon76-dig | codefreeze: yes, we do |
| 16:13.36 | Juggie | Corydon76-home, and you think its acceptable to hammer a db over and over and over if the statement may never ever work |
| 16:13.52 | Qwell | putnopvut: sure enough, right after you answered that, the reporter answered it too |
| 16:14.00 | codefreeze | Is it good enough for the admin to fix it, whatever it was, the "ERROR" explanation? |
| 16:14.04 | Corydon76-dig | Juggie: 40 times, once every 30 seconds is hammering a database? |
| 16:14.18 | Juggie | Corydon76-dig, it doesnt seem like much with one CDR |
| 16:14.41 | Juggie | but it would be exponential based on the number of cdr records affected |
| 16:14.50 | Corydon76-dig | codefreeze: any error is going to need admin intervention, no matter if it's a connection issue or a bad index |
| 16:15.22 | Corydon76-dig | That's the whole point of it being an error. Asterisk cannot self-correct, admin intervention required. |
| 16:15.39 | Juggie | Corydon76-dig, your not getting it, i dont care WHY the insert failed |
| 16:15.48 | Juggie | a failed insert is fatal |
| 16:15.55 | Juggie | a connection problem is temprary |
| 16:16.02 | Corydon76-dig | Why oh why should we get into the cesspool of "this error I'm going to retry, but this other error I won't."? |
| 16:16.20 | Corydon76-dig | Juggie: a connection error causes a failed insert |
| 16:16.50 | Corydon76-dig | A connection problem is permanent |
| 16:16.59 | Juggie | yes, but cant we tell if the problem is a connection error, or if the actual insert is bad. |
| 16:17.02 | Corydon76-dig | unless you count admin intervention |
| 16:17.20 | Corydon76-dig | but if you count admin intervention, then ALL errors are temporary |
| 16:17.46 | Corydon76-dig | Juggie: it does not matter |
| 16:18.13 | Corydon76-dig | Juggie: there are going to be countless edge cases, and it's a cesspool to try to catch them all |
| 16:18.21 | Corydon76-dig | Juggie: handle it generically |
| 16:18.37 | Juggie | well, its just as bad to throw away cdr records |
| 16:19.09 | Corydon76-dig | At a certain point, you have no choice |
| 16:19.30 | Corydon76-dig | If you retry N times, and it fails, throwing away the CDR is your only choice at that point |
| 16:19.52 | Juggie | right, but why define a time limit |
| 16:20.05 | Juggie | maybe a switch burned out and it will take 2 hours to get a new one in place. |
| 16:20.09 | mvanbaak | so the sysadmin can correct shit |
| 16:20.17 | Corydon76-dig | Because if you don't define a time limit, you're going to fill memory with useless crap that will never work |
| 16:20.23 | Juggie | its better i think to set a max records |
| 16:20.43 | Juggie | because 20minutes on a high volume system means something totally different, then 20minutes on a system that gets few calls |
| 16:20.46 | Corydon76-dig | Max records is going to be completely different on different systems |
| 16:21.03 | *** join/#asterisk-dev ManxPower (n=manxpowe@139.sub-75-202-162.myvzw.com) |
| 16:21.06 | Juggie | how so, isnt the cdr struct a fairly consitant size? |
| 16:21.26 | Corydon76-dig | Sure, a max records might be great over all, but I'd set it so high that it won't activate unless you're in danger of eating all memory |
| 16:21.50 | Corydon76-dig | The CDR struct itself is a constant size, but arbitrary variables can be attached to it |
| 16:21.51 | jmls | putnopvut: you do not want me to raise a bug in the multiple member assignment then ? |
| 16:22.12 | Juggie | Corydon76-dig, right i realize that. |
| 16:22.24 | *** join/#asterisk-dev imesper (n=ian@201-43-159-125.dsl.telesp.net.br) |
| 16:23.50 | Corydon76-dig | codefreeze: and I agree. Every 30 seconds, maximum 40 retries, gives the admin 20 minutes to fix the problem or forget it |
| 16:24.13 | codefreeze | why not provide the use with a multi-level sort of system, where if we can't store this in the db, then ... X, and the user can fill in the blank... |
| 16:24.21 | Corydon76-dig | codefreeze: and if you really want to be anal, log the failed query to a log file |
| 16:24.45 | mvanbaak | in mysql binlog format |
| 16:24.47 | mvanbaak | ;) |
| 16:24.57 | codefreeze | and if the log is full, then Y, and so on...? |
| 16:25.25 | codefreeze | And if all of X, Y, and Z fail, then give up. |
| 16:25.26 | Corydon76-dig | codefreeze: I'm talking about the logger |
| 16:25.43 | Corydon76-dig | the only way for the logger to fail is if you run out of disk space |
| 16:25.51 | imesper | Hi, I am working on meetme and tried to to something like ztc.confmode ^= (ZT_CONF_TALKER | ZT_CONF_LISTENER) but I can't get it to work right, Am I missing something? |
| 16:25.57 | Corydon76-dig | and bad things happen when you run out of disk space |
| 16:26.40 | Corydon76-dig | imesper: I think you mean "|=" |
| 16:26.40 | caio1982 | which is pretty common for embedded asterisk system... |
| 16:26.43 | codefreeze | Ah, but does it have to? Can't we then, if the user is smart, log to some other partition or system? |
| 16:27.19 | imesper | I tried to put the preson inside a conference "on hold", not listening nor talking |
| 16:27.52 | Corydon76-dig | Oh, then you want &= ~(ZT_CONF_TALKER|ZT_CONF_LISTENER) |
| 16:28.31 | imesper | Thanks a lot |
| 16:32.53 | putnopvut | jmls: no, jsmith is right in his analysis of local channel device state. |
| 16:46.33 | codefreeze | Corydon76-dig: I'm not grokking the objection to Juggie's assertion: wouldn't it be OK for him to hold some user-specified number of CDR's in a buffer, just in case a temp error gets fixed? I think that would be a cool feature, personally. |
| 16:47.00 | russellb | agreed |
| 16:48.26 | imesper | Corydon76-dig, it's the same thing the user stop listen, but keep talking |
| 16:51.52 | ManxPower | Did you know there is not a single example of using CALLERID in extensions.conf in 1.2? |
| 16:52.09 | Juggie | also a feature that was requested was the ability to pause cdrs |
| 16:52.15 | outtolunc | i never noticed before that local channel pairs aren't mated with 'a' and 'b' +common random |
| 16:52.17 | Juggie | eg, lets say you have to take a db down for mainmtenance |
| 16:52.31 | Juggie | pause the backend for sometime, do your work, then start it up when you are done |
| 16:53.45 | russellb | M10553 |
| 16:53.46 | MuffinMan | [feedback] [Asterisk] Resources/res_agi 0010553: [patch] the fgets in res_agi can sometimes get interupted, but we dont handle it. reported by juggie http://bugs.digium.com/view.php?id=10553 |
| 16:53.52 | russellb | Juggie: i dont' see how that patch worked |
| 16:54.04 | russellb | you define AGI_RESULT_SUCCESS_FAST, but no code ever returns that |
| 16:54.10 | russellb | so it's not possible for the fast flag to ever get set |
| 16:57.49 | russellb | file: chan_refcount doesn't crash anymore ) |
| 16:57.51 | russellb | er, :) |
| 16:57.56 | file | russellb: lame, that's no fun |
| 16:59.45 | russellb | doesn't appear to be leaking any memory in the general case, either. |
| 17:00.38 | russellb | that's hot |
| 17:04.57 | codefreeze | russellb: any good news on the ast_obj2 slowdown/speedup? |
| 17:05.03 | Corydon76-dig | codefreeze: I'm not opposed to that. What I'm opposed to is the idea that we're going to magically know which errors can be corrected by an admin and which can't |
| 17:05.48 | codefreeze | Ahhh, OK. |
| 17:07.45 | Corydon76-dig | codefreeze: the problem is that there are all sorts of cases that an admin, somewhere, thinks he can correct, and he will be pissed that Asterisk doesn't retry those CDRs |
| 17:08.54 | Corydon76-dig | Rather than trying to fix every edge case and satisfy every admin, if we go for the generic case (retry everything), the logic stays simple and it works. |
| 17:11.15 | Corydon76-dig | (and we have an out on every admin complaint: either you set the limit too high or too low) |
| 17:19.48 | JerJer | has anyone ever come up with a 'good' way to give a very much non-technical person access to debug information from asterisk ? |
| 17:19.54 | JerJer | say like level 1 support |
| 17:21.06 | Corydon76-dig | JerJer: yeah, I've given web access to the logfile |
| 17:21.49 | ManxPower | It might be helpful if Asterisk complained when you try to reference a dialplan variable that does not exist. |
| 17:22.12 | JerJer | yeah, however when there are 300 simultaneous calls on a box, thats a stupid amount of data for a level 1 tech to filter thru |
| 17:22.18 | JerJer | o |
| 17:25.46 | ManxPower | Not complaining seems to cause many issues with debugging. |
| 17:25.46 | JerJer | i'm thinking like "my sip phone doesn't work" |
| 17:25.46 | Corydon76-dig | JerJer: you can filter on threadid |
| 17:25.46 | JerJer | or my DID from you doesn't work |
| 17:25.46 | Corydon76-dig | It's just a matter of creating the right filter for them to access |
| 17:25.46 | JerJer | and since sip phones don't register with our asterisk box debugging has always been a serious problem |
| 17:25.47 | JerJer | *boxes |
| 17:25.47 | Corydon76-dig | Yeah, that's not usually something I've tried to solve for users, since we're in a different position... we provision all the phones |
| 17:25.47 | Corydon76-dig | or if we don't, it's level 2 that is provisioning phones |
| 17:27.52 | jmls | MacroExclusive is meant to be that, no ? |
| 17:28.17 | Corydon76-dig | Yes |
| 17:29.59 | tzafrir | JerJer, I like the front page of trixbox. Very informative |
| 17:30.18 | tzafrir | It's easy to provide information |
| 17:30.31 | tzafrir | You just have to figure out which information |
| 17:30.42 | jmls | Corydon76-dig: http://asterisk.pastebin.ca/702125 |
| 17:31.21 | jmls | I run the macroexclusive, which gets the next available agent (where paused = 0), sets paused = 2 and returns |
| 17:31.58 | jmls | in the exact same microsecond I have 2 calls to the macro that got the same agent record because paused was 0. |
| 17:32.14 | Corydon76-dig | jmls: are you using 1.4 or trunk? |
| 17:32.21 | jmls | 1.4 |
| 17:32.32 | jmls | latest svn |
| 17:32.33 | Corydon76-dig | You need to backslash your comma, not quote it |
| 17:37.33 | *** join/#asterisk-dev JerJer (n=PhatJ@pdpc/supporter/bronze/jerjer) [NETSPLIT VICTIM] |
| 17:37.33 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) [NETSPLIT VICTIM] |
| 17:37.33 | *** join/#asterisk-dev key2 (n=Ritual@193.33.36.20) |
| 17:37.33 | *** join/#asterisk-dev jmls (n=jmls@62.49.235.130) [NETSPLIT VICTIM] |
| 17:37.33 | *** join/#asterisk-dev ido (n=ido@unaffiliated/ido) [NETSPLIT VICTIM] |
| 17:37.33 | *** mode/#asterisk-dev [+o JerJer] by irc.freenode.net |
| 17:37.52 | jmls | who |
| 17:37.56 | jmls | where ? |
| 17:37.58 | jmls | why ? |
| 17:38.17 | jmls | you guys are back, then ... |
| 17:38.17 | outtolunc | janitor, in the colo, with a vacuum |
| 17:38.22 | Corydon76-dig | jmls: You need to backslash your comma, not quote it |
| 17:38.26 | jmls | heh |
| 17:38.49 | jmls | Eh ? |
| 17:39.01 | jmls | what comma ? Where ? |
| 17:39.56 | Corydon76-dig | in the CUT |
| 17:40.33 | jmls | er, I use that throughout the dialplan and it works just fine ... |
| 17:40.51 | jmls | the data is returned "a,b,c" |
| 17:41.00 | Corydon76-dig | Does it look fine in 'dialplan show'? |
| 17:41.41 | jmls | 2. Set(USERCODE=${CUT(RESULT|,|1)}) [pbx_config] |
| 17:42.12 | Corydon76-dig | You know you could be using ARRAY(), right? |
| 17:42.37 | jmls | yeah, that was pointed out to me last time I raised my head above the trenches ;) |
| 17:42.48 | jmls | just haven't got round to replacing it all |
| 17:44.50 | jmls | doesn't explain how the same agent was found at the same time by the exclusive macro .. |
| 17:45.50 | Corydon76-dig | I'd need to see your database table contents to know for sure |
| 17:45.53 | jsmith | jmls: By the exclusive macro this time, eh? |
| 17:45.54 | Corydon76-dig | and your query |
| 17:46.30 | Corydon76-dig | jmls: show me your queries |
| 17:46.56 | Corydon76-dig | GETNEXTAGENT and AGENTPAUSED |
| 17:48.47 | jmls | Corydon76-dig: http://asterisk.pastebin.ca/702144 |
| 17:52.33 | Corydon76-dig | jmls: can you confirm that the user you're connecting as has update privilege on that table? |
| 17:53.01 | jmls | yes. absolutely. |
| 17:53.16 | Corydon76-dig | Odd |
| 17:53.27 | jsmith | jmls: On your GETNEXTAGENT query, how are you ordering the results? |
| 17:53.31 | jmls | it happened once out of 1000 calls |
| 17:54.02 | jmls | jsmith: there's an index defined on queuename, then paused status then datetime of last update to the record |
| 17:54.18 | jmls | iow it gets the oldest non paused member |
| 17:54.24 | Corydon76-dig | Any possibility of some other process getting in there to clear that flag? |
| 17:54.28 | jsmith | jmls: Cool... |
| 17:55.26 | Corydon76-dig | jmls: because that's the only thing I can think of |
| 17:55.57 | Corydon76-dig | i.e. do you have some process that goes around and clears flags? |
| 17:56.04 | jmls | Corydon76-dig: I really don't think so, as they are all virtual agents "waiting" for a cal (paused = 0)l. the only time they go not ready (paused > 0) is when the receive a call. After call has hung up, they wait for 10 seconds and then go ready again |
| 17:56.49 | jmls | if something resets the flag, then the datetime puts them to the bottom of the queue |
| 17:57.44 | jmls | the agents do not reject the call, they answer it. So they would instantly go paused > 0 |
| 17:57.47 | Corydon76-dig | bottom of the queue is shallow if there's only one available agent |
| 17:57.52 | jmls | true |
| 17:58.19 | jmls | put they have a 10 second pause after they are busy before they go ready again |
| 17:58.33 | jmls | so i can't see how the same record could have been grabbed in the same second |
| 17:59.14 | Corydon76-dig | I can't, either |
| 17:59.34 | Corydon76-dig | unless you've got two macros that do the same thing |
| 18:00.16 | Corydon76-dig | MacroExclusive is only exclusive to the name, not to the whole process |
| 18:02.41 | jmls | only one macro .. |
| 18:05.15 | *** join/#asterisk-dev Juggie (n=Juggie@CPE001601df17fb-CM000a73a18a20.cpe.net.cable.rogers.com) |
| 18:09.24 | *** join/#asterisk-dev dlynes_ (n=dlynes@d154-20-9-152.bchsia.telus.net) |
| 18:10.04 | *** join/#asterisk-dev lrizzo (n=luigi@88-149-143-175.dynamic.ngi.it) |
| 18:10.42 | *** join/#asterisk-dev elixer (i=elixer@65.207.74.18) |
| 18:14.52 | *** join/#asterisk-dev qdk (n=qdk@0x50c627be.bynxx11.adsl-dhcp.tele.dk) |
| 18:18.04 | *** part/#asterisk-dev Victor_Yure (n=aaaa@esp5.deibotoch.com.br) |
| 18:22.28 | *** join/#asterisk-dev ctooley (n=ctooley@209.33.109.249) |
| 18:36.45 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-88-154-177-39.red.bezeqint.net) |
| 18:36.45 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 18:40.56 | *** join/#asterisk-dev bkruse (i=bkruse@nat/digium/x-1c04bfa2b599c2c1) |
| 18:45.53 | russellb | codefreeze: well, try the test again and see what results you get. i cut it down to 25% of the time on my box with a small change to hashtest2 |
| 19:08.15 | *** join/#asterisk-dev mitcheloc (n=mitchel@adsl-68-120-230-95.dsl.irvnca.pacbell.net) |
| 19:22.37 | Juggie | russellb, hah, good question, it should of been returned from launch_netscript |
| 19:23.21 | Juggie | but then, why would it have even worked for festr |
| 19:23.33 | Juggie | gr |
| 19:24.22 | Juggie | i know i did that, i must of lost it at some point when i reapplied the patch to a clean svn before doing the diff |
| 19:25.10 | russellb | worked fine for him because without that flag set, the code is no different |
| 19:26.02 | Juggie | ya i know... |
| 19:26.18 | Juggie | without the flag set, nothing has changed |
| 19:26.28 | russellb | yup |
| 19:26.41 | Juggie | but it fixed it for me too |
| 19:26.51 | Juggie | i was able to reproduce his problem |
| 19:27.22 | russellb | cooool |
| 19:27.28 | Juggie | i shall have to test it again, i know i modified launch_netscript |
| 19:27.39 | russellb | yeah, i figured it was probably one line or something |
| 19:27.50 | Juggie | i must of lost it somehow, not sure... |
| 19:27.58 | Juggie | ok, i'll post a couple of new patches in a bit, sorry about that |
| 19:28.03 | russellb | no problem :) |
| 19:37.06 | jsmith | ~tickle elmo |
| 19:37.07 | jbot | ACTION jumps on elmo, yelling "TICKLE FIGHT!!!!" |
| 19:40.05 | putnopvut | Can someone explain the purpose of the while loop in this code snippet? http://pastebin.ca/702296 |
| 19:41.23 | russellb | putnopvut: the equivalent of ast_skip_blanks() |
| 19:41.28 | russellb | skipping over whitespace |
| 19:42.38 | putnopvut | So it only increments on non-whitespace. |
| 19:42.46 | russellb | correct |
| 19:43.32 | *** join/#asterisk-dev bweschke (n=BJ_Wesch@c-24-0-209-5.hsd1.nj.comcast.net) |
| 19:43.32 | *** mode/#asterisk-dev [+o bweschke] by ChanServ |
| 19:44.04 | *** join/#asterisk-dev twilson (n=twilson@CPE-65-30-30-20.kc.res.rr.com) |
| 19:46.02 | Corydon76-dig | Welcome welcome |
| 19:46.17 | twilson | Thank you, thank you |
| 19:46.36 | twilson | It's good to be here. |
| 19:48.11 | file | is it REALLY? |
| 19:48.31 | Qwell | file: shh, don't tell him! |
| 19:48.35 | Qwell | err, umm, hi |
| 19:48.41 | twilson | Well, apparently it would be better to be somewhere with a Cheeburger. :-) |
| 19:48.50 | Qwell | twilson: there is one in St Louis |
| 19:48.56 | putnopvut | And one in Little Rock |
| 19:48.57 | Qwell | it said so on the menu |
| 19:49.01 | bweschke | twilson: did he mention that he hates you for that? |
| 19:49.04 | bweschke | if not, it's coming |
| 19:49.08 | bweschke | wait for it......... |
| 19:49.39 | twilson | But alas, I am in KC which is about a 4-5 hour drive. |
| 19:49.46 | twilson | to STL |
| 19:50.04 | Qwell | it'll be worth it |
| 19:50.30 | twilson | I have friends there, maybe I'll go visit them. |
| 19:51.55 | jsmith | twilson! |
| 19:52.01 | jsmith | ~welcome twilson |
| 19:52.01 | jbot | Howdy, twilson! |
| 19:52.17 | Qwell | jsmith: that was anti-climatic... |
| 19:52.18 | twilson | hello all! |
| 19:52.20 | bweschke | jbot? |
| 20:00.12 | Corydon76-dig | or anticlimactic, even? |
| 20:00.38 | Qwell | whichever |
| 20:00.48 | Corydon76-dig | global warming is anticlimatic |
| 20:00.57 | Corydon76-dig | So is a nuclear winter |
| 20:01.08 | Qwell | oh, well the greeting was not nuclear winter |
| 20:01.12 | jsmith | bweschke: jbot is our friendly IRC bot |
| 20:01.17 | jsmith | ~zap bweschke |
| 20:01.18 | jbot | ACTION takes out a cattle prod and gives bweschke a good jolt. |
| 20:01.23 | file | oh, bweschke snuck in before twilson |
| 20:01.34 | putnopvut | ~putnopvut |
| 20:01.35 | jbot | you are probably a can of diet coke |
| 20:01.39 | Corydon76-dig | bj's been here for awhile |
| 20:01.47 | file | jbot: file? |
| 20:01.48 | jbot | rumour has it, file is Joshua Colp, jcolp@digium.com, and is the best looking software developer at Digium |
| 20:01.59 | file | jbot: botsnack |
| 20:01.59 | jbot | thanks, file |
| 20:02.01 | putnopvut | That needs to be updated. |
| 20:02.15 | bweschke | jbot: liar! |
| 20:02.16 | jbot | hey, I just work here; I don't have to take that kind of abuse from you |
| 20:02.16 | Corydon76-dig | Yeah, you were on DDP yesterday |
| 20:02.16 | Qwell | ~qwell |
| 20:02.17 | jbot | you are probably a patented liquid formula that contains three plant-based bio-active agents that work together in a perfectly balanced combination. These agents act synergistically to boost your good cholesterol and slash the bad. |
| 20:02.25 | jsmith | jbot: No, file is Joshua Colp, jcolp@digium.com |
| 20:02.26 | jbot | jsmith: okay |
| 20:02.29 | jsmith | ~file |
| 20:02.44 | jsmith | ~wobble Qwell |
| 20:02.44 | jbot | ACTION pushes Qwell so that he wobbles like file |
| 20:03.06 | Corydon76-dig | lol, that one's great |
| 20:03.18 | bweschke | does jbot do Astricon slide decks too? then he'd be really useful |
| 20:03.21 | Corydon76-dig | ~jsmith |
| 20:03.22 | jbot | jsmith is probably perpetually hungry, or the author of Asterisk: The Future of Telephony |
| 20:03.38 | jsmith | ~powerpoint |
| 20:06.17 | russellb | jbot: powerpoint is lame, use openoffice, instead |
| 20:06.18 | jbot | russellb: okay |
| 20:06.27 | russellb | ~powerpoint |
| 20:06.28 | jbot | hmm... powerpoint is lame, use openoffice, instead |
| 20:06.31 | Corydon76-dig | ~corydon76 |
| 20:06.32 | jbot | i guess corydon76 is the master of asstricks, or an Asterisk developer, or a stalker of file |
| 20:06.48 | putnopvut | ~corydon |
| 20:06.49 | jbot | it has been said that corydon is one of the most l33t Asterisk developers around. He has been around longer than you have. |
| 20:06.56 | russellb | ~russell |
| 20:06.56 | jbot | rumour has it, russell is the coolest person, ever. |
| 20:06.58 | *** join/#asterisk-dev elixer (i=elixer@65.207.74.18) |
| 20:07.02 | bweschke | jbot: lol |
| 20:07.02 | jbot | it has been said that lol is stands for Laughs Out Loud. It is grammatically incorrect to use LOL in the first person; use 'heh' or 'haha' instead. If you want to use LOL, do '/me lol' instead. |
| 20:07.05 | file | jbot is totally right |
| 20:07.07 | putnopvut | ~russellb |
| 20:07.08 | jbot | it has been said that russellb is Russell Bryant <russell@digium.com>. He is also kind of a big deal, people know him. ...NOT!! |
| 20:07.31 | russellb | ~drumkilla |
| 20:07.31 | jbot | drumkilla is, like, Russell Bryant, the Asterisk release branch maintainer <russelb@clemson.edu>, or someone who should be ph33r3d |
| 20:07.39 | jmls | Corydon76-dig: about my duplicate agent grabber ... |
| 20:07.49 | Corydon76-dig | jmls: what'd you find? |
| 20:08.11 | jmls | how about this: first call into the exclusive macro grabs the agent |
| 20:08.32 | jmls | before it can update the record, the call is hung up. *but* the variables are set |
| 20:08.49 | jmls | what happens at this point ? Does the macro return or hang up ? |
| 20:09.04 | Corydon76-dig | It returns |
| 20:09.12 | *** join/#asterisk-dev tsurko (n=tsurko@213.91.216.130) |
| 20:09.23 | jmls | right - so the variables are updated, but not the db record |
| 20:09.30 | jmls | so the next call in gets the same agent |
| 20:09.32 | Corydon76-dig | On a hangup, a macro always exits immediately (unless it's already in a hangup routine) |
| 20:10.04 | bweschke | muhahahaha |
| 20:10.07 | bweschke | drumkilla |
| 20:10.24 | jmls | so the jabber messages is sent to the agent, and the second call _updates_ the agent record and then sends the message to the agent |
| 20:10.50 | Corydon76-dig | file: I haven't stalked you, in like, forever |
| 20:11.01 | jmls | is there anyway of forcing the macro to complete before returning ? |
| 20:11.11 | Qwell | Corydon76-dig: I'll be sure to stalk him next week |
| 20:11.15 | bweschke | Corydon76-dig: phoenix is only next week |
| 20:11.17 | Corydon76-dig | file: I'll need to get back in the habit... |
| 20:11.18 | bweschke | LOL |
| 20:11.22 | jmls | hmm. I may update the database before I set the variables. I'll let you know. |
| 20:11.24 | Corydon76-dig | I won't be in Phoenix |
| 20:11.29 | bweschke | doh! |
| 20:11.41 | bweschke | why in the heck not ? |
| 20:11.48 | Corydon76-dig | I'm saving the company money by not going |
| 20:12.13 | Corydon76-dig | (Also, I find it really difficult to eat when I'm away from home) |
| 20:13.22 | *** join/#asterisk-dev shido6 (n=shido6@74-130-56-203.dhcp.insightbb.com) |
| 20:13.45 | bweschke | so who's going then from camp Digium? |
| 20:13.57 | file | robots |
| 20:13.59 | Qwell | like 6 or 8 of us |
| 20:14.10 | russellb | meeeee |
| 20:14.15 | Qwell | russellb, file, kpfleming, me, Doug, umm...other people |
| 20:14.20 | Qwell | putnopvut: |
| 20:14.23 | Qwell | (sorry) |
| 20:14.26 | file | all the cool people |
| 20:14.32 | bweschke | oh nice - I don't think I've met putnopvut yet |
| 20:14.45 | Qwell | bweschke: he was at astridevcon |
| 20:15.00 | bweschke | seems like eons ago |
| 20:15.03 | Qwell | you were at astridevcon, right? |
| 20:15.11 | bweschke | I was, yes .. |
| 20:15.11 | Qwell | yeah, of course you were |
| 20:15.40 | bweschke | davevg-btwtech almost didn't notice I was gone from the office until you all snitched on me. :) |
| 20:15.43 | putnopvut | bweschke: I probably talked with you at some point at the devcon. I tried to talk to everyone a bit. |
| 20:16.10 | Qwell | bweschke stands out in a crowd. bweschke rocks. |
| 20:16.13 | bweschke | putnopvut: quite possible |
| 20:16.23 | file | I also heard that bweschke sometimes rolls as well |
| 20:16.37 | Qwell | file: rumor, I saw it on snopes |
| 20:16.46 | bweschke | look for the glare coming off my head, and there I am... |
| 20:16.48 | russellb | Deewayne: you are super cool! |
| 20:16.58 | russellb | file: apologize to Deeewayne ! |
| 20:17.12 | Qwell | Dwayne is so cool, that if he went to Phoenix, he'd melt :( |
| 20:17.17 | file | :( |
| 20:17.47 | bweschke | no. |
| 20:17.49 | bweschke | no I am not |
| 20:18.28 | Deeewayne | Qwell: thanks! |
| 20:18.33 | Deeewayne | I was cool once |
| 20:21.01 | Deeewayne | that was a swift tackle hug file |
| 20:26.10 | Corydon76-dig | Deeewayne: oh, are you as blind as I am without glasses? |
| 20:27.04 | Corydon76-dig | Be happy that you don't have those infernal video billboards around Huntsville... they're impossible to focus on with glasses |
| 20:27.27 | putnopvut | We have some. If you mean the ones that switch pictures every few seconds. |
| 20:27.38 | putnopvut | Or do you mean actual video video? |
| 20:27.55 | Corydon76-dig | putnopvut: yeah, those things |
| 20:28.20 | bweschke | corydon76-dig: ya the resolution on those guys isn't quite 1080i.. |
| 20:28.29 | bweschke | or 480i even for that matter. |
| 20:28.47 | putnopvut | The only thing I like about those is that there's one that occasionally will put the weather forecast up, but aside from that I don't like them. |
| 20:28.58 | Corydon76-dig | If they go below a certain font size, I can't read them |
| 20:30.09 | Corydon76-dig | or it could be my astigmatism |
| 20:33.33 | Corydon76-dig | Someone remind me how to turn off [CC] so I can see the gcc line? |
| 20:34.40 | russellb | make NOISY_BUILD=omgyesplzkthx |
| 20:35.30 | russellb | M9238 |
| 20:35.31 | MuffinMan | [feedback] [Asterisk] Core/ManagerInterface 0009238: [branch] manager seems to not flush it's eventq like it should reported by bweschke http://bugs.digium.com/view.php?id=9238 |
| 20:35.32 | kpfleming | lol |
| 20:35.46 | russellb | bweschke: looks like the latest patches there have found the root cause of the issue, i'm going to apply a fix shortly |
| 20:36.10 | bweschke | russelb: good deal |
| 20:36.27 | elixer | getting lots of patches queued up on the NEW_CLI bug... |
| 20:36.41 | bweschke | it seemed like that guy Ivan was on the right track |
| 20:36.41 | elixer | tracking bug, that is. |
| 20:36.47 | bweschke | even with his original patch |
| 20:37.14 | jmls | heh |
| 20:37.23 | jmls | [Sep 18 21:36:54] ERROR[14140]: astobj2.c:169 ao2_ref: refcount -1 on object 0xb7348b78 |
| 20:37.54 | russellb | crapx0r .. |
| 20:38.05 | snuff-away | nice |
| 20:38.11 | jmls | :) :) |
| 20:39.02 | elixer | M10724 |
| 20:39.03 | MuffinMan | [new] [Asterisk] Resources/res_features 0010724: [patch] use NEW_CLI for CLI commands. reported by eliel http://bugs.digium.com/view.php?id=10724 |
| 20:39.09 | elixer | 17 patches |
| 20:39.19 | bweschke | whoa |
| 20:39.24 | bweschke | that's enough to keep someone busy |
| 20:39.38 | bweschke | jmls: which one are you testing ? |
| 20:44.18 | *** part/#asterisk-dev shido6 (n=shido6@74-130-56-203.dhcp.insightbb.com) |
| 20:45.49 | jmls | 1.4 |
| 20:45.52 | jmls | svn |
| 20:47.56 | bweschke | app_queue ? |
| 20:49.14 | *** join/#asterisk-dev VoicePulse (n=contact@unaffiliated/voicepulse) |
| 20:56.44 | russellb | do you guys ever type so much that the tips of your fingers start hurting? :( |
| 20:57.16 | Qwell | no, but I get mouse-finger occasionally |
| 20:57.49 | file | russellb: it's a sign of impending doom |
| 20:58.18 | jmls | bweschke: yes |
| 20:58.38 | russellb | Qwell: what is that? |
| 20:58.50 | Qwell | same as keyboard-finger, except with a mouse |
| 20:58.59 | russellb | oh , heh |
| 20:59.01 | Qwell | or playstation-thumb |
| 20:59.05 | russellb | i spend most of my time in a terminal ... |
| 20:59.07 | Qwell | I hate playstation-thumb |
| 20:59.18 | russellb | wii ... arm? |
| 20:59.22 | Qwell | heh |
| 20:59.35 | russellb | i get wii arm playing trauma center |
| 20:59.37 | russellb | that game pwns |
| 21:02.01 | russellb | ooh, moy is from ibm? |
| 21:07.17 | snuff-away | get my email russellb ;) |
| 21:12.50 | russellb | yes |
| 21:15.37 | snuff-away | cool |
| 21:17.34 | snuff-away | worked in 1.4.6 then didn't work in 1.4.9/1.4.11 |
| 21:18.00 | russellb | :( |
| 21:18.26 | Qwell | SCO blames Linux |
| 21:18.36 | snuff-away | there is no sco almost anymore :) |
| 21:18.37 | Qwell | http://yro.slashdot.org/article.pl?sid=07/09/18/175259 |
| 21:18.54 | russellb | lol |
| 21:19.28 | russellb | ... d'uh ? |
| 21:22.03 | russellb | re: slashdot link |
| 21:22.59 | Qwell | yeah, you know, maybe they blamed...themselves |
| 21:23.05 | Qwell | that wouldn't happen though, heh |
| 21:27.52 | jmls_away | goodnight all |
| 21:37.49 | *** part/#asterisk-dev pnlarsson (n=pnlarsso@c83-248-12-187.bredband.comhem.se) |
| 21:58.04 | Qwell | hey putnopvut |
| 21:58.52 | Qwell | I think there are places in add_to_queue where membername is checked, where it can be NULL |
| 21:59.56 | Qwell | maybe not..hold that thought |
| 22:00.29 | elixer | who is terry wilson? he here? |
| 22:00.33 | Qwell | yeah, nevermind |
| 22:00.34 | Qwell | elixer: twilson |
| 22:00.37 | elixer | heh |
| 22:00.43 | elixer | twilson: congrats |
| 22:01.07 | twilson | elixer:thanks |
| 22:04.50 | twilson | I'm still mapping real people to handles. :-) |
| 22:05.45 | elixer | ah, well i'm sean |
| 22:05.55 | elixer | but you can call me sean |
| 22:05.57 | elixer | heh |
| 22:07.10 | elixer | well terry and i don't count |
| 22:07.14 | elixer | since we just went over that |
| 22:08.02 | file | Qwell: you can't name me! |
| 22:08.09 | Qwell | ~file |
| 22:08.11 | Qwell | I don't have to |
| 22:08.16 | Qwell | erm |
| 22:10.50 | elixer | hmmm |
| 22:10.52 | elixer | ~elixer |
| 22:11.02 | elixer | :( |
| 22:11.10 | elixer | never like you anyway, stupid bot. |
| 22:11.17 | elixer | s/like/liked/ |
| 22:15.05 | file | Qwell: like I said, can't name me |
| 22:18.49 | snuff-work | mm.. anyone testing/using M10590 |
| 22:18.54 | snuff-work | M10590 |
| 22:18.55 | MuffinMan | [new] [Asterisk] Core/RTP 0010590: RTP statistics returned by ast_rtp_get_quality reflects the last RTCP packet not a call overall reported by gasparz http://bugs.digium.com/view.php?id=10590 |
| 22:47.14 | Qwell | A82930 |
| 22:47.17 | MuffinMan | [82930] Committed by qwell on 2007-09-18 19:43:45: (issue #10724) |
| 22:47.17 | MuffinMan | p.c.patch uploaded by moy (license 222) |
| 22:47.17 | MuffinMan | 64) |
| 22:47.32 | Qwell | yeah, I'm gonna go ahead and put the message first from now on |
| 22:48.06 | russellb | heh |
| 22:48.32 | Qwell | oh, and the "Include reporter" checkbox doesn't do anything :D |
| 22:48.40 | Qwell | it seems to always include the reporter |
| 22:48.53 | Qwell | (commit template) |
| 22:49.02 | russellb | heh |
| 22:55.45 | Qwell | russellb: any chance we could change the commit template thing to say to put the actual message before all that other stuff? |
| 22:56.29 | russellb | only if ... |
| 22:56.30 | russellb | um ... |
| 22:56.42 | russellb | you have 50 cents |
| 22:56.55 | Qwell | I have like...37 |
| 22:56.59 | russellb | hm |
| 22:57.03 | putnopvut | I have 10 |
| 22:57.09 | putnopvut | With our powers combined!! |
| 22:57.21 | Qwell | we could buy ... I don't know... half a soda |
| 22:57.22 | putnopvut | we'd have 47 cents... |
| 22:57.51 | file | math is good. |
| 22:57.51 | Qwell | I want to make a vending machine that will dispense exactly as much beverage as you have money for |
| 22:58.12 | russellb | i have 30 |
| 22:58.15 | russellb | and some pennies |
| 22:58.25 | putnopvut | We could buy a soda and split it 3 ways. |
| 22:58.28 | russellb | i really want more caffiene |
| 22:58.38 | putnopvut | I'd get 10/75ths of it. |
| 22:59.21 | russellb | i'm game |
| 22:59.53 | putnopvut | russellb: should've had soda at lunch, man |
| 23:00.28 | Qwell | I wish cheeburger would use Styrofoam (xchat said to use a cap S) cups or something |
| 23:00.37 | Qwell | ((apparently Styrofoam is a TM)) |
| 23:00.49 | Qwell | (((I did not know that))) |
| 23:01.00 | russellb | Qwell: i'm looking at mantis to change what you asked for ... |
| 23:01.03 | russellb | but i'm scared of it. |
| 23:01.06 | Qwell | heh |
| 23:02.35 | Qwell | the reporter thing isn't a big deal |
| 23:03.05 | russellb | Qwell: it didn't work |
| 23:03.10 | russellb | now i'm scared i broke something |
| 23:03.45 | Qwell | http://uuoc.com/1904 |
| 23:03.45 | Qwell | :D |
| 23:04.34 | Qwell | ...in theory |
| 23:05.04 | Qwell | or maybe <br /> twice on the [commit message] line |
| 23:05.21 | russellb | i was looking at the wrong file |
| 23:05.34 | Qwell | yeah, 2x <br /> |
| 23:05.52 | russellb | you know, you have commit to this ... |
| 23:06.00 | Qwell | so I do |
| 23:06.24 | russellb | failed! |
| 23:07.03 | Qwell | committed! |
| 23:07.07 | Qwell | how does it get updated? |
| 23:07.26 | russellb | i do it. |
| 23:07.28 | russellb | and it's done |
| 23:07.39 | Qwell | ooo, much better |
| 23:07.41 | russellb | ssh carolina ; cd aldskjflaskdf ; svn up |
| 23:07.42 | russellb | heh |
| 23:10.20 | Qwell | you still sshed in? |
| 23:10.38 | russellb | yes |
| 23:10.51 | Qwell | try that? :D |
| 23:11.00 | Qwell | first stab at the reporter checkbox thingie |
| 23:11.06 | russellb | updated |
| 23:11.21 | Qwell | rockin! |
| 23:11.25 | Qwell | it worked |
| 23:11.37 | Qwell | no more reporter if the box isn't checked |
| 23:12.57 | russellb | hmm |
| 23:13.01 | russellb | what else should we hack :-p |
| 23:14.42 | file | russellb: hack on... CCM |
| 23:15.03 | russellb | ... |
| 23:20.17 | *** join/#asterisk-dev amarzouk (n=chatzill@217.54.201.107) |
| 23:21.59 | amarzouk | Hi, I am having problems compiling asterisk 1.4.11 on centos, my problem is with the chan_vpb has anybody faced similar problems? |
| 23:23.16 | russellb | amarzouk: disable building chan_vpb |
| 23:23.28 | Qwell | what is vpb for anyhow? |
| 23:23.30 | russellb | amarzouk: run make menuselect, go to channels, chan_vpb, hit space to turn it off, 'x' to save and quit |
| 23:23.35 | russellb | Qwell: voicetronix boards |
| 23:23.39 | Qwell | ahh |
| 23:23.43 | russellb | but the driver is obviously not being maintained |
| 23:23.48 | Qwell | right |
| 23:24.15 | amarzouk | russellb: but I have a voicetronix board, that's why I need chan_vpb |
| 23:24.23 | russellb | amarzouk: ah. |
| 23:24.44 | russellb | amarzouk: in that case, if it has compiler errors, put it up on bugs.digium.com |
| 23:25.03 | russellb | if it's a problem getting the right stuff installed, try the asterisk-users list, or cotnact voicetronix for help |
| 23:25.29 | russellb | but like i said, they haven't been maintaining chan_vpb for a long time now |
| 23:25.34 | russellb | so ... beware ... |
| 23:27.34 | amarzouk | russellb: I have already installed the board, the tests run without problems. |
| 23:27.36 | amarzouk | russellb: so do you think if I install an older version of asterisk it would work? |
| 23:28.49 | file | naughty putnopvut |
| 23:31.11 | *** join/#asterisk-dev elixer (n=seanbrig@c-68-55-114-113.hsd1.md.comcast.net) |
| 23:31.25 | elixer | Qwell: was the rollup patch a waste? |
| 23:32.01 | elixer | in other words, should i bother in the future? |
| 23:32.15 | elixer | in other words, hey how's it going? |
| 23:38.46 | putnopvut | file: What? |
| 23:39.37 | file | putnopvut: made buildbot all red and unhappy |
| 23:39.43 | putnopvut | Oh dear... |
| 23:39.58 | putnopvut | When did I manage that? |
| 23:40.04 | file | app_queue stuffz |
| 23:40.11 | putnopvut | From this morning? |
| 23:40.18 | russellb | lawlerskatez |
| 23:40.22 | putnopvut | I could have sworn I test compiled it. |
| 23:40.43 | file | silly world |
| 23:40.56 | putnopvut | In 1.4 or trunk? |
| 23:41.00 | file | both |
| 23:41.14 | file | merging up the fix :D |
| 23:41.24 | file | very minor |
| 23:42.09 | putnopvut | It compiles here... |
| 23:42.53 | putnopvut | What gcc version are you using? |
| 23:43.26 | file | I would have to go and look at the builders to answer that |
| 23:43.41 | *** join/#asterisk-dev outtolunc (n=me@c-67-174-216-60.hsd1.ca.comcast.net) |
| 23:43.45 | putnopvut | What was the build error? |
| 23:43.56 | file | uninitialized variable |
| 23:44.11 | file | it's all happy again |
| 23:44.20 | putnopvut | That's weird... |
| 23:46.04 | russellb | oh good lord ... chan_refcount is never going to be finished |
| 23:46.32 | elixer | just reading the commit log is exhausting for me |
| 23:46.33 | elixer | :) |
| 23:46.39 | russellb | there are so many places channel pointers get stored is not even funny |
| 23:46.48 | putnopvut | I find it hilarious. |
| 23:47.19 | russellb | i started looking at chan_sip and it is killing my soul |
| 23:51.15 | elixer | sweet sweet mother of god |
| 23:52.05 | Skyelar | russellb: ahh, just rewrite it in C++ and have the pointers refcount themselves |
| 23:52.30 | elixer | smart pointers! |
| 23:52.32 | elixer | weeeeee! |
| 23:52.42 | russellb | if only |
| 23:53.00 | file | Skyelar: don't make me refcount you |
| 23:53.41 | russellb | you won't hear me argue against C++ |
| 23:53.59 | Skyelar | I rather liked the tirade from Linus T on it recently though |
| 23:54.16 | elixer | i still have that patch, russellb :-) |
| 23:54.18 | russellb | heh, haven't read that |
| 23:54.29 | russellb | elixer: the one to convert asterisk to C++? heh |
| 23:54.31 | russellb | awesome |
| 23:54.34 | russellb | let's put it in a branch. |
| 23:54.45 | elixer | its a one liner |
| 23:54.50 | putnopvut | Let's have a Digium-maintained C++ fork of Asterisk!! :) |
| 23:54.50 | russellb | oh, nice |
| 23:55.00 | elixer | sed -i -e 's/gcc/g++/g' *.c |
| 23:55.09 | elixer | err |
| 23:55.10 | elixer | damniot |
| 23:55.15 | putnopvut | elixer: that won't compile. |
| 23:55.15 | russellb | asterisk won't even come close to building in C++ mode |
| 23:55.20 | elixer | nevermind, i am no longer funny |
| 23:55.27 | Skyelar | his main point appeared to be not so much that C++ was bad, but that it attracted so many lackluster "lets use object brokers for everything" type programmers |
| 23:55.37 | russellb | heh |
| 23:55.40 | putnopvut | elixer: unfortunately we use many reserved words from C++ as variable names and such. |
| 23:55.45 | russellb | Skyelar: fair point |
| 23:55.50 | russellb | it's not even that |
| 23:55.58 | elixer | putnopvut: yeah, i've seen them scattered around |
| 23:56.03 | russellb | but implicit casting of void * _everywhere_ |
| 23:56.12 | russellb | initializer list syntax which isn't allowed in C++ _everywhere_ |
| 23:56.15 | elixer | true, true. |
| 23:56.35 | elixer | initializer syntax is a little fragile if you ask me, anyway. |
| 23:56.41 | elixer | but you didn't... so... |
| 23:56.42 | elixer | :) |
| 23:56.48 | putnopvut | russellb: yeah but what I said will cause compilation errors nonetheless. |
| 23:56.53 | russellb | elixer: in what, C++? |
| 23:57.03 | russellb | putnopvut: right, definitely |
| 23:57.15 | elixer | russellb: you mean like 'struct foo = { 0, 1, "Something" };' |
| 23:57.15 | elixer | ? |
| 23:57.21 | russellb | nope |
| 23:57.27 | elixer | what do you mean then |
| 23:57.28 | Skyelar | (oops, I didn't actually mean to reignite the whole C++ thing :-) |
| 23:57.34 | russellb | struct foo = { .a = 1, .b = 2, .c = 3, }; |
| 23:57.38 | elixer | ahhhh |
| 23:57.52 | russellb | i would agree that the latter syntax is fragile |
| 23:57.52 | putnopvut | The comma at the end of the initializer always strikes me as odd. |
| 23:57.54 | russellb | to put it nicely |
| 23:58.02 | elixer | yeah, i didn't know that even existed until i saw asterisk source |
| 23:58.05 | elixer | is that C99? |
| 23:58.09 | russellb | C99, yeah |
| 23:58.26 | elixer | the inner functions you were committing the other day had me scratching my head |
| 23:58.30 | russellb | previous to C99 picking that, there was also another one, which we have in some place ... |
| 23:58.49 | russellb | struct foo foo = { a: 1, b: 2, c: 3 }; |
| 23:58.58 | russellb | oh, embedded functions? |
| 23:59.01 | elixer | yeah |
| 23:59.02 | russellb | heh, gcc extension ... |
| 23:59.34 | russellb | though some other compilers support it, too |
| 23:59.40 | elixer | its interesting. can you store pointers to them? |
| 23:59.47 | russellb | yeah |
| 23:59.47 | elixer | or are they dead outside of the function |
| 23:59.56 | russellb | but they're only valid within the scope where the function is defined |