| 00:01.10 | seanbright | putnopvut: that is what i had |
| 00:01.28 | seanbright | putnopvut: but the problem is when you pass OBJ_POINTER, it expects that argument to be hashable |
| 00:02.23 | russellb | and it is very much to your benefit for it to be hashable |
| 00:02.28 | russellb | otherwise, you kill your search performance |
| 00:02.41 | seanbright | putnopvut: another option might be to define a OBJ_EXTERNAL |
| 00:02.52 | seanbright | which is (1 << 16) or something |
| 00:03.06 | russellb | we'll probably have to change the API |
| 00:03.25 | russellb | it is clear that there is a need for another arg so you can pass something to compute the has one, and something else to pass as an arg to the callback |
| 00:03.29 | seanbright | right |
| 00:03.34 | russellb | and that would solve all of this |
| 00:03.34 | seanbright | will add |
| 00:04.27 | seanbright | do you recommend adding another function or just adding the arg to the existing one? |
| 00:04.35 | seanbright | i'm guessing the former |
| 00:05.30 | russellb | add it to ao2_callback |
| 00:05.38 | seanbright | ugh |
| 00:05.39 | seanbright | ok |
| 00:05.57 | russellb | not that you have to do it right now |
| 00:06.01 | seanbright | right |
| 00:08.22 | putnopvut | seanbright: ah, gotcha. Didn't think about the hashing aspect. |
| 00:08.36 | putnopvut | Of course... |
| 00:09.04 | putnopvut | since the peer's are hashed by name, if you make the first field of the new struct the name and size it at 80 bytes, the hash still might work. |
| 00:09.10 | putnopvut | s/peer's/peers/ |
| 00:09.27 | seanbright | doesn't it do the magic pointer subtraction stuff? |
| 00:09.31 | seanbright | astobj2, that is |
| 00:10.02 | russellb | that is terribly evil |
| 00:10.11 | russellb | please don't do that, heh |
| 00:10.13 | seanbright | but it makes it transparent |
| 00:10.16 | seanbright | oh |
| 00:10.17 | seanbright | right, no. |
| 00:10.49 | putnopvut | russellb: aw, why not ;) |
| 00:11.01 | *** join/#asterisk-dev eliel (n=eliel@151-202-114-200.fibertel.com.ar) |
| 00:11.01 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 00:13.24 | putnopvut | oh, and just for clarity, there is no internal magic done on the argument to ao2_callback. |
| 00:14.32 | seanbright | hmmm |
| 00:14.35 | seanbright | you are correct, sir. |
| 00:15.00 | seanbright | but i would have to change all the ao2_callback call sites to use the struct i defined, right? |
| 00:15.00 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 00:15.22 | putnopvut | seanbright: it's only used in find_peer, so just that one place. |
| 00:15.31 | putnopvut | The other ao2_callbacks use a different callback function. |
| 00:15.41 | seanbright | but the hash function is the important one |
| 00:15.49 | putnopvut | But still, like russellb said, it's a really evil idea anyway. |
| 00:15.56 | seanbright | oh right |
| 00:16.04 | seanbright | but it would work if the first member was 'name' |
| 00:16.05 | seanbright | i got you |
| 00:16.24 | seanbright | i'd call that "voodoo" not "evil" :) |
| 00:16.33 | putnopvut | heh |
| 00:16.39 | snuff-work | its polymorphisim :) |
| 00:16.47 | putnopvut | lol |
| 00:18.30 | seanbright | might be worth splitting up the ao2_callback_fn into ao2_callback_fn and ao2_compare_fn |
| 00:18.44 | russellb | :-/ |
| 00:18.56 | seanbright | i hate programming |
| 00:18.57 | seanbright | i quit. |
| 00:19.00 | russellb | lol |
| 00:19.06 | snuff-work | mm modifying core api.. i'm happy i'm not doing it ;) |
| 00:19.12 | russellb | honestly, i just need to focus on this before i make comments |
| 00:19.14 | russellb | ignore me for now |
| 00:19.24 | seanbright | done and done |
| 00:25.15 | *** join/#asterisk-dev SQLDarkly (n=nospam@p10-162.dsl.ecentral.com) |
| 00:35.28 | seanbright | hrmmm |
| 00:35.34 | seanbright | that is *exactly* what is done in main/manager.c |
| 00:35.49 | seanbright | line 3561 |
| 00:36.30 | russellb | vomits |
| 00:36.39 | russellb | This is why I like code reviews ^_^ |
| 00:36.50 | seanbright | yeah... |
| 00:37.03 | putnopvut | svn blame time! |
| 00:37.07 | russellb | chan_sip used to do that, actually, in its astobj1 code |
| 00:37.56 | russellb | i hated it there, too :-p |
| 00:38.45 | seanbright | 89537 tilghman /* Due to the simplicity of struct variable_count, it makes no difference |
| 00:38.49 | seanbright | wah wah waaahhhhhhhhhhhhh |
| 00:38.59 | seanbright | it's a neat trick though |
| 00:46.17 | seanbright | utils/astobj2.c == the unclean spawn of the beast with two backs |
| 00:48.25 | file | mumbles |
| 00:52.36 | russellb | pokes file |
| 00:52.43 | russellb | file: the package has been delivered |
| 00:52.48 | file | russellb: yay! |
| 00:53.04 | file | russellb: party tomorrow. |
| 00:54.42 | russellb | file: hawt |
| 00:54.46 | russellb | bbl, time to go home |
| 00:54.48 | *** part/#asterisk-dev russellb (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 01:04.47 | seanbright | annnnnnnnnnnnnnnnnnnnnnnd done |
| 01:04.59 | seanbright | wanders off |
| 01:28.03 | *** join/#asterisk-dev outtolunc (n=me@c-67-188-204-139.hsd1.ca.comcast.net) |
| 02:01.23 | seanbright | putnopvut: what are you trying to do in the makefile? |
| 02:02.20 | seanbright | just signify that timerfd is available? |
| 02:05.52 | *** join/#asterisk-dev russellb (n=russell@asterisk/digium-open-source-team-lead/russellb) |
| 02:05.52 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 02:20.42 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 02:20.43 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 02:28.33 | *** join/#asterisk-dev kamanashisroy (n=kamanash@119.30.34.3) |
| 02:39.17 | *** join/#asterisk-dev jtodd (n=jtodd@blob.fox-den.com) |
| 03:37.54 | *** join/#asterisk-dev jameswf (n=james@ip68-3-60-66.ph.ph.cox.net) |
| 04:02.35 | *** join/#asterisk-dev putnopvut (n=putnopvu@nat/digium/x-f7f48e86c9c89fef) [NETSPLIT VICTIM] |
| 04:02.35 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) [NETSPLIT VICTIM] |
| 04:02.35 | *** join/#asterisk-dev pgarcia (n=chatzill@yoda.kanatek.com) [NETSPLIT VICTIM] |
| 04:02.35 | *** join/#asterisk-dev mbranca (n=matteo@93-62-230-243.ip24.fastwebnet.it) [NETSPLIT VICTIM] |
| 04:02.35 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) [NETSPLIT VICTIM] |
| 04:02.35 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) [NETSPLIT VICTIM] |
| 04:02.35 | *** join/#asterisk-dev MuffinMan (n=muffinma@asterisk/issue-tracker-bot/muffinman) |
| 04:02.36 | *** join/#asterisk-dev corramor (n=corramor@nat/digium/x-d6c5048d458cbaea) [NETSPLIT VICTIM] |
| 04:02.36 | *** mode/#asterisk-dev [+ooo putnopvut kpfleming jpeeler] by irc.freenode.net |
| 04:44.28 | *** join/#asterisk-dev putnopvut (n=putnopvu@nat/digium/x-f7f48e86c9c89fef) [NETSPLIT VICTIM] |
| 04:44.28 | *** join/#asterisk-dev mbranca (n=matteo@93-62-230-243.ip24.fastwebnet.it) [NETSPLIT VICTIM] |
| 04:44.28 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) [NETSPLIT VICTIM] |
| 04:44.28 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) [NETSPLIT VICTIM] |
| 04:44.28 | *** join/#asterisk-dev MuffinMan (n=muffinma@asterisk/issue-tracker-bot/muffinman) |
| 04:44.28 | *** join/#asterisk-dev corramor (n=corramor@nat/digium/x-d6c5048d458cbaea) [NETSPLIT VICTIM] |
| 04:44.28 | *** mode/#asterisk-dev [+ooo putnopvut kpfleming jpeeler] by irc.freenode.net |
| 05:32.27 | *** join/#asterisk-dev CunningPike (n=arodgers@S010600095b33697f.vc.shawcable.net) |
| 06:08.38 | *** join/#asterisk-dev jeffspeff (n=jeff@c-69-245-31-86.hsd1.ky.comcast.net) |
| 06:53.58 | *** part/#asterisk-dev snuff-work (n=jsewell@203.176.96.250) |
| 07:07.07 | *** join/#asterisk-dev steliosk (n=Stelios@ipa107.2.tellas.gr) |
| 07:12.31 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 07:12.32 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 07:41.34 | *** join/#asterisk-dev mvanbaak (n=michiel@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 07:41.34 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 07:44.58 | *** join/#asterisk-dev grEvenX (n=even@apb9hb.ip.ssc.net) |
| 07:45.00 | grEvenX | hi guys! anyone have any tips on how to do performance profiling with Asterisk? on just one outgoing call I'm using 80% CPU... |
| 08:16.05 | tzafrir_laptop | grEvenX, which thread is it? |
| 08:16.12 | tzafrir_laptop | press 'H' in top |
| 08:16.25 | tzafrir_laptop | then strace this thread |
| 08:45.15 | grEvenX | tzafrir_laptop: thanks, I'll try that |
| 08:47.28 | grEvenX | hm |
| 08:47.45 | grEvenX | there seems to be a whole lot of threads |
| 08:47.53 | grEvenX | is that normal? |
| 08:48.17 | grEvenX | perhaps 30-40 threads when there's one call going through the dialplan |
| 09:56.20 | *** join/#asterisk-dev Pazzo (n=ugelt@reserved-225136.rol.raiffeisen.net) |
| 10:01.26 | *** join/#asterisk-dev jeffspeff2 (n=jeff@c-69-245-31-86.hsd1.ky.comcast.net) |
| 10:36.42 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 10:39.16 | *** join/#asterisk-dev eliel (n=eliel@200.61.172.61) |
| 10:39.16 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 10:55.42 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:03.17 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 11:03.17 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 11:04.58 | *** join/#asterisk-dev juanjoc (n=juanjoc@OL4-215.fibertel.com.ar) |
| 11:39.44 | *** join/#asterisk-dev hugorebelo (n=hugorebe@200-171-132-124.completo.com.br) |
| 12:06.04 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 12:06.04 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 12:09.34 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 12:09.35 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 13:11.23 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:11.24 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 13:19.03 | *** join/#asterisk-dev russellb (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 13:19.03 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 13:21.32 | *** join/#asterisk-dev shido6 (n=shido6@209.114.208.111) |
| 13:23.12 | snuff-home | mmm.. its friday.. or just gone sat here ;) |
| 13:23.41 | lmadsen | oh really |
| 13:23.46 | lmadsen | it is early friday morning here! |
| 13:23.48 | lmadsen | russellb: !!! |
| 13:24.18 | russellb | ohai!! |
| 13:24.29 | russellb | happy friday |
| 13:24.39 | lmadsen | happy happy friday! |
| 13:24.45 | russellb | does the Friday dance |
| 13:24.55 | lmadsen | .dance |
| 13:24.56 | lmadsen | :) |
| 13:25.06 | lmadsen | imagines a kirby like figure dancing |
| 13:25.26 | snuff-home | would consider dancing.. if pastry was back ;) |
| 13:25.59 | *** mode/#asterisk-dev [-r] by russellb |
| 13:26.31 | *** join/#asterisk-dev DonutPuncher (n=DonutPun@user-24-214-112-81.knology.net) |
| 13:26.34 | russellb | .dance |
| 13:26.34 | DonutPuncher | <PROTECTED> |
| 13:26.34 | DonutPuncher | <PROTECTED> |
| 13:26.34 | DonutPuncher | <PROTECTED> |
| 13:26.36 | DonutPuncher | <PROTECTED> |
| 13:26.38 | DonutPuncher | <PROTECTED> |
| 13:26.40 | DonutPuncher | <PROTECTED> |
| 13:26.42 | *** mode/#asterisk-dev [+r] by russellb |
| 13:26.56 | snuff-home | lol |
| 13:27.21 | lmadsen | ummm... w00t! |
| 13:27.32 | russellb | .acronym lmadsen |
| 13:27.35 | DonutPuncher | LIQUID MIGHTST ARMOUR DAMES SURVEYORS ENGROSSED NEATLY |
| 13:27.45 | lmadsen | obviously |
| 13:27.48 | russellb | clearly. |
| 13:27.51 | russellb | ok, enough, heh |
| 13:27.54 | *** kick/#asterisk-dev [DonutPuncher!n=russellb@asterisk/digium-open-source-team-lead/russellb] by russellb (bye!) |
| 13:28.04 | lmadsen | turns on his 8" monitor to watch his development server boot up to make sure the network connects |
| 13:28.26 | lmadsen | .lyrics richard_berry louie_louie |
| 13:28.26 | eliel | hello |
| 13:28.37 | snuff-home | disappears to get some well needed zzz's and think about packing for trip |
| 13:28.38 | russellb | eliel: hey there |
| 13:28.40 | eliel | russellb: Pastrycheff is dead.... |
| 13:28.43 | lmadsen | doh! :) |
| 13:28.44 | russellb | nods |
| 13:28.52 | russellb | I guess I should fix that. |
| 13:30.46 | eliel | when you can |
| 13:31.52 | russellb | shouldn't take long ... *works on it now* |
| 13:38.12 | russellb | bots are back |
| 13:38.14 | russellb | now, do they work? |
| 13:38.19 | russellb | we shall see |
| 13:39.03 | russellb | now it should |
| 13:39.25 | eliel | I was looking at astmm.c and I think we have an inconsistency with the API, if we have ast_debug_malloc turned on, and we have an allocation failure, functions like asprintf() will set the pointer to NULL, but if we don't have debug_malloc, that pointer will have an undefinied content... |
| 13:43.05 | *** join/#asterisk-dev sasargen (n=chatzill@173.100.55.16) |
| 13:48.04 | lmadsen | so question... is moving away from nested functions possibly going to help with random crash issues? |
| 13:48.44 | russellb | it would have nothing to do with it ... |
| 13:48.47 | jsmith | lmadsen: Not as I understand it |
| 13:48.54 | lmadsen | oh ok, was just curious |
| 13:49.30 | jsmith | lmadsen: Any luck with your MySQL problem? |
| 13:49.30 | russellb | it's just removing a construct that is convenient, but has portability and security issues that come with it |
| 13:49.39 | russellb | so, backing it back out... |
| 13:49.39 | lmadsen | jsmith: none what-so-ever :) |
| 13:50.01 | lmadsen | russellb: ah ok, that's the answer I was looking for, thx! |
| 13:54.15 | jsmith | lmadsen: I wish mytop was working for you -- that'd most likely solve your problem |
| 13:54.27 | jsmith | lmadsen: (or, I mean, *help* you solve it) |
| 13:54.31 | lmadsen | that'd be nice |
| 13:54.40 | lmadsen | I'm 100% out of ideas now |
| 13:55.28 | lmadsen | just did 'yum install mytop -y', but I guess there is something wrong with the RPM or something |
| 13:55.38 | lmadsen | seems to depend a lot on CPAN modules, and I'm just not going there |
| 13:57.36 | jsmith | Why do you think it's a problem with RPM? |
| 13:58.10 | jsmith | It seems to me that RPM did it's job perfectly fine, but that there's a problem with mytop itself |
| 13:58.22 | jsmith | Did you specify a user/password/host in ~/.mytop? |
| 14:05.22 | jsmith | lmadsen: The other thing I'd look at is the query log. It's supposed to log the database connections as well |
| 14:13.04 | lmadsen | jsmith: sorry, I meant with what was packaged in the RPM, not the RPM itself |
| 14:13.08 | lmadsen | jsmith: I passed arguments to mytop to allow it to connect |
| 14:13.29 | jsmith | lmadsen: Hmmmn... do you have that error again? |
| 14:14.51 | lmadsen | sure! let me run it |
| 14:16.11 | lmadsen | Use of uninitialized value in substitution (s///) at /usr/bin/mytop line 955. |
| 14:28.09 | lmadsen | brb |
| 14:28.29 | *** join/#asterisk-dev Deeewayne (n=Deeewayn@nat/digium/x-f4d503413c27c8c0) |
| 14:28.29 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 14:28.31 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 14:28.32 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 14:54.27 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 15:02.34 | *** join/#asterisk-dev agx (n=AGX@host63-216-static.34-88-b.business.telecomitalia.it) |
| 15:03.11 | agx | is the Russo's stun work available for 1.4.x too in some svn branches or patches? |
| 15:04.19 | russellb | i do not know of any 1.4 patches for that |
| 15:05.34 | agx | russellb, ok thanks |
| 15:05.50 | russellb | np |
| 15:07.51 | *** join/#asterisk-dev telecos (n=sergio@84.166.219.87.dynamic.jazztel.es) |
| 15:19.06 | *** join/#asterisk-dev jameswf (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 15:21.37 | seanbright | good morning, chaps. |
| 15:24.08 | jsmith | Morning sean |
| 15:26.28 | lmadsen | greetings! |
| 15:26.30 | lmadsen | salutations! |
| 15:26.33 | lmadsen | good tidings! |
| 15:28.31 | seanbright | reiterates his hatred of utils/ |
| 15:41.49 | putnopvut | seanbright: I think everyone hates utils/ |
| 15:47.13 | tzafrir_laptop | utils introduces a strage notion of code reuse |
| 15:50.00 | seanbright | code re-use by duplication |
| 15:50.05 | seanbright | i should write a whitepaper |
| 15:51.13 | jsmith | And duplicate it a bunch of times? |
| 15:51.36 | seanbright | yes, with subtle differences each time |
| 15:51.37 | seanbright | :) |
| 15:51.41 | tzafrir_laptop | so how can I read an asterisk config file, |
| 15:51.48 | seanbright | extconf.c |
| 15:51.55 | seanbright | in... utils/ |
| 15:52.09 | tzafrir_laptop | without duplicating asterisk's code, and without assuming asterisk is running? |
| 15:52.27 | seanbright | shrugs |
| 15:52.28 | seanbright | perl! |
| 15:53.03 | seanbright | russellb: minor point -> "Using OBJ_MULTIME with OBJ_NODATA |
| 15:53.23 | seanbright | i shall call him... multi-me. |
| 15:53.37 | jsmith | tzafrir_laptop: Not very easily... as a total hack you could use ast_live and AMI on that instance of Asterisk to modify the other Asterisk configs :-/ |
| 15:54.45 | russellb | seanbright: ha ... spelling fail |
| 16:01.36 | mvanbaak | reviewboard is cool |
| 16:05.07 | russellb | :) |
| 16:05.22 | russellb | msg ChipX86 on freenode and tell him how much you <3 it |
| 16:05.23 | russellb | hehe |
| 16:05.34 | seanbright | stalker |
| 16:05.37 | russellb | yes. |
| 16:06.50 | mvanbaak | I dont know what kevin means with the 'allocate after the structure' trick |
| 16:07.00 | russellb | i saw that |
| 16:07.03 | russellb | i was about to write up an explanation |
| 16:07.05 | russellb | it's tricky |
| 16:07.07 | mvanbaak | cool |
| 16:07.10 | mvanbaak | thank you |
| 16:07.29 | mvanbaak | <--- is learning everyday when touching the asterisk code |
| 16:07.41 | russellb | cool :) |
| 16:07.51 | russellb | i wish we had a whiteboard, that would make it easier |
| 16:08.00 | mvanbaak | I like it you guys actually spend time to teach ppl |
| 16:08.01 | lmadsen | www.skrbl.com |
| 16:08.03 | russellb | I'll just explain here |
| 16:08.08 | lmadsen | russellb: ^^ |
| 16:08.15 | russellb | lmadsen: ooh |
| 16:08.20 | lmadsen | oh snap eh? :) |
| 16:08.32 | russellb | heh |
| 16:08.45 | russellb | i'll try to explain in text first, and if we have trouble, i'll draw. |
| 16:08.58 | lmadsen | is a big fan of The Draw |
| 16:09.07 | russellb | ok, so let's make a struct ... struct foo { char *str1; char *str2; }; |
| 16:09.50 | russellb | Now, we want to allocate a block of memory big enough such that we'll have this struct, and have str1 = "ohai" and str2 = "kthxbye" |
| 16:10.07 | russellb | Step 1) How big should the allocation be? |
| 16:10.32 | russellb | Answer: sizeof(struct foo) + strlen("ohai") + strlen("kthxbye") + 2; |
| 16:10.40 | mvanbaak | indeed |
| 16:10.45 | russellb | that's enough space for the struct, both strings, and null char terminators for each |
| 16:10.49 | russellb | with me so far? |
| 16:10.52 | mvanbaak | yup |
| 16:10.54 | jsmith | follows |
| 16:11.15 | russellb | Ok. Next step is how to initialize the string to point to the buffers that are big enough to hold the strings we want. |
| 16:11.22 | russellb | Pointer magic time. |
| 16:11.29 | mvanbaak | \o/ |
| 16:11.50 | russellb | struct foo *myfoo = ast_calloc(1, sizeof(*myfoo) + strlen("ohai") + strlen("kthxbye") + 2); |
| 16:12.17 | russellb | myfoo->str1 = myfoo + 1; |
| 16:12.32 | russellb | Now, where does myfoo point? |
| 16:12.40 | russellb | errr, myfoo->str1 |
| 16:13.10 | russellb | Answer: the first byte _after_ sizeof(struct foo) |
| 16:13.10 | mvanbaak | erm |
| 16:13.17 | mvanbaak | aaaaaaaaaaaah |
| 16:13.24 | mvanbaak | was running around in circles |
| 16:13.32 | russellb | hehe, does it make sense why? |
| 16:13.45 | russellb | It's just how pointer math works. |
| 16:13.45 | russellb | It's just how pointer math works |
| 16:13.47 | russellb | erm. |
| 16:14.03 | *** join/#asterisk-dev hlsroddy (n=hlsroddy@213-156-55-137.fastres.net) |
| 16:14.07 | mvanbaak | I think I get it |
| 16:14.11 | russellb | +1 in this case is + (sizeof the object pointed to by the variable) |
| 16:14.16 | russellb | Here is another way you could set str1: |
| 16:14.31 | russellb | myfoo->str1 = ((char *) myfoo) + sizeof(*myfoo); |
| 16:14.31 | russellb | same result. |
| 16:14.38 | hlsroddy | hi all |
| 16:14.51 | russellb | But the pointer math might be easier to understand there, since we've forced it to do the math in bytes |
| 16:14.59 | mvanbaak | yeah |
| 16:15.09 | russellb | So, does that make sense? |
| 16:15.23 | russellb | (we'll do str2 next) |
| 16:15.25 | mvanbaak | a little. it's not senseless |
| 16:15.29 | russellb | heh ... |
| 16:15.37 | seanbright | you're basically saving yourself 2 allocations |
| 16:15.44 | russellb | myfoo->str2 = ((char *) myfoo) + strlen("ohai") + 1; |
| 16:15.58 | russellb | now, we're setting str2 to be after the struct, and then after the buffer we're using for the first string. |
| 16:16.03 | russellb | seanbright: yep. |
| 16:16.16 | mvanbaak | yeah, I get that |
| 16:16.24 | mvanbaak | *pling* |
| 16:16.27 | mvanbaak | light goes on |
| 16:16.35 | russellb | finally ... strcpy(myfoo->str1, "ohai"); |
| 16:16.42 | russellb | strcpy(myfoo->str2, "kthxbye"); |
| 16:16.50 | russellb | since you _know_ there is enough space there, no need for bounds checking |
| 16:17.00 | mvanbaak | indeed |
| 16:17.02 | russellb | ta-da |
| 16:17.04 | mvanbaak | makes sense now |
| 16:17.30 | russellb | So, you use a single allocation that allocates just enough for your struct, plus the dynamically sized strings that you want to put in there |
| 16:17.42 | russellb | (this works as long as these strings never change after allocation!!) |
| 16:18.09 | mvanbaak | which is the case with the cli alias |
| 16:18.18 | russellb | correct. |
| 16:18.24 | mvanbaak | because you read it from configfile, and use that string without modification |
| 16:19.42 | *** join/#asterisk-dev bbryant (n=brett@68.208.65.50) |
| 16:19.42 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 16:19.48 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 16:20.12 | atis_work | Corydon76-dig: ping |
| 16:20.47 | mvanbaak | hhmm, what's the benefit over the code we have now in res_clialiases.c |
| 16:20.52 | mvanbaak | that's not really clear to me |
| 16:21.06 | russellb | you go from 3 allocations and 3 calls to free to 1 |
| 16:21.15 | russellb | that's about it. |
| 16:21.49 | russellb | we generally try to minimize the number of allocations because 1) it's more efficient, and 2) it's less error prone |
| 16:24.32 | seanbright | 1 down, 2 to go. |
| 16:29.54 | *** join/#asterisk-dev jeffspeff2 (n=jeff@c-69-245-31-86.hsd1.ky.comcast.net) |
| 16:29.58 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 16:29.58 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 16:30.31 | mvanbaak | if (!(alias = ao2_alloc(sizeof(*alias), alias_destroy))) { |
| 16:30.41 | mvanbaak | so that line should be changed |
| 16:31.00 | mvanbaak | and the setting of alias and real_cmd should not be done with ast_strdup but with strcpy |
| 16:31.23 | mvanbaak | and in between I have to set the alias and real_cmd to the correct place of course |
| 16:32.57 | russellb | correct |
| 16:33.05 | russellb | and the last thing is just removing the ast_free calls of those 2 fields |
| 16:33.28 | mvanbaak | course. we dont want asterisk to go *KABOOM* |
| 16:33.33 | russellb | :) |
| 16:33.50 | mvanbaak | let me see if I can fix it |
| 16:36.27 | mvanbaak | ao2_alloc can be used instead of ast_calloc ? |
| 16:36.39 | russellb | correct |
| 16:42.26 | mvanbaak | http://pastebin.ca/1247967 |
| 16:42.28 | mvanbaak | like that ? |
| 16:43.28 | russellb | yup |
| 16:43.34 | russellb | except for a couple missing spaces after "+" |
| 16:43.34 | russellb | :) |
| 16:43.51 | mvanbaak | yup, saw that as well |
| 16:43.59 | mvanbaak | hhmm, tricky but neat :) |
| 16:46.19 | *** join/#asterisk-dev CunningPike (n=arodgers@204.239.8.157) |
| 16:48.00 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 16:48.00 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 16:48.01 | mvanbaak | hurray ! |
| 16:48.05 | mvanbaak | ok, bbl |
| 16:48.05 | *** join/#asterisk-dev outtolunc (n=me@c-67-188-204-139.hsd1.ca.comcast.net) |
| 16:49.56 | *** join/#asterisk-dev fred-tmft (n=fred-tea@c-98-243-174-131.hsd1.mi.comcast.net) |
| 16:50.15 | lmadsen | mvanbaak: ping when you return! |
| 16:50.16 | *** part/#asterisk-dev fred-tmft (n=fred-tea@c-98-243-174-131.hsd1.mi.comcast.net) |
| 17:23.43 | *** join/#asterisk-dev sruffell (n=sruffell@nat/digium/x-89dd2e3d5d31de7b) |
| 17:23.43 | *** mode/#asterisk-dev [+o sruffell] by ChanServ |
| 17:31.08 | jsmith | putnopvut: Ping? |
| 17:32.15 | putnopvut | jsmith: Pong |
| 17:33.39 | jsmith | putnopvut: Can I get a status update on where we are on bug 8824? Just curious, not trying to be pushy... |
| 17:33.51 | putnopvut | jsmith: sure, I'll pm you |
| 17:54.32 | eliel | uuh, we loose the redirection bugs.digium.com/ISSUEID in the migration |
| 17:54.55 | eliel | rewrite actually |
| 17:55.31 | kpfleming | let me check the apache config on the old server |
| 17:55.36 | russellb | ah ... yeah, was probably config |
| 17:55.46 | russellb | actually, it may have been that 404.php |
| 17:55.53 | russellb | that didn't appear to make it to the new server |
| 17:55.56 | russellb | guess it wasn't in svn ... |
| 17:56.47 | kpfleming | i think that's it, yes |
| 18:00.05 | jtodd | It's sounds like a 60hz hum, so probably A/C bleed. |
| 18:00.39 | kpfleming | wrong channel? |
| 18:00.46 | jtodd | So is this dinner going to be FedEx'ed to me here? |
| 18:00.55 | russellb | O.O |
| 18:02.05 | jtodd | I think that the first wideband codecs on mobile will be transported via IP (g.722) |
| 18:02.16 | file | jtodd: you can come out for dinner with me if you'd like! |
| 18:02.20 | eliel | huh? |
| 18:02.28 | jtodd | Supposedly IMS will solve all those problems. I sceptically await those changes. |
| 18:03.33 | jtodd | eliel: was that directed at me? My point was that (per the audio discussion) mobile networks will upgrade to "wideband" when hell freezes over, or they convert to IMS. |
| 18:03.45 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-179-75-202.static.bezeqint.net) |
| 18:03.45 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 18:04.32 | jtodd | eliel: Or, clients running over the WiFi (or other wide/local area IP networking) portion of the device will use G.722 or other wideband codecs first without the cooperation of the carrier that uses the regulated portion of the mobile spectrum. |
| 18:04.49 | file | jtodd: do you mean for these to go to the users conf chatroom? |
| 18:05.21 | jtodd | Shoot. Too many windows. Sorry- this was going to voip-users-conference, where we're talking about wideband audio. Gotta keep these windows straight. :-) |
| 18:06.02 | jtodd | Too many of the same people on the channels - hard to use the "participant list" as the key that I use to distingush between windows. |
| 18:06.11 | eliel | jtodd: my huh was for that :-) |
| 18:06.23 | jtodd | Gotcha. |
| 18:11.43 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 18:14.17 | *** join/#asterisk-dev Schmooze (n=Schmooze@rrcs-67-53-189-178.west.biz.rr.com) |
| 18:54.03 | Corydon76-dig | jtodd: I find keeping windows crooked works better |
| 19:05.05 | jtodd | Corydon76-dig: I'll tilt my monitor. |
| 19:06.14 | *** join/#asterisk-dev harry_v (n=pcsuppor@S0106001d7e52cc78.vs.shawcable.net) |
| 19:10.40 | harry_v | http://www.pastebin.ca/1248060 is this a smsq.o or limits.h issue? make install for 1.4 stops when doing a make install after make clean. |
| 19:11.18 | seanbright | run 'make NOISY_BUILD=yes' |
| 19:11.24 | harry_v | k |
| 19:12.36 | harry_v | looks like the same output error |
| 19:13.01 | seanbright | it is, but you should see the full compile line now |
| 19:13.06 | seanbright | and not the [cc] -> stuff |
| 19:13.36 | harry_v | it is |
| 19:13.40 | seanbright | pastebin it |
| 19:14.18 | harry_v | http://www.pastebin.ca/1248089 |
| 19:14.46 | seanbright | hmmm |
| 19:14.56 | seanbright | looks like your build environment is screwed up |
| 19:15.07 | harry_v | mmm |
| 19:15.21 | seanbright | have you upgraded gcc or glibc recently? |
| 19:15.31 | harry_v | not that I know of |
| 19:16.00 | seanbright | what is the output of "gcc --version" |
| 19:16.07 | kpfleming | 42 |
| 19:16.19 | seanbright | ~thwack kpfleming |
| 19:16.20 | jbot | ACTION beats kpfleming on the finger with a UNIX Manual |
| 19:16.21 | harry_v | 3.4.6-10 |
| 19:17.01 | seanbright | ls -al /usr/include/linux/limits.h |
| 19:17.48 | seanbright | (i'm just taking wild stabs in the dark at this point) |
| 19:18.58 | harry_v | <PROTECTED> |
| 19:19.03 | harry_v | its there |
| 19:19.16 | seanbright | hmm. |
| 19:19.22 | seanbright | shrugs |
| 19:19.31 | seanbright | something broke your build environment |
| 19:19.55 | harry_v | so basicly what the error is saying that smsq.c is looking for limits.h but there is no inlcude in the file? |
| 19:19.57 | seanbright | i.e. not an asterisk bug |
| 19:20.01 | harry_v | yea |
| 19:20.05 | harry_v | i suspected so |
| 19:20.29 | seanbright | brb |
| 19:20.34 | harry_v | but how. 1.2 was working fine and just wanted to upgrade. |
| 19:29.12 | seanbright | i'm not sure. |
| 19:29.32 | seanbright | someone else here might know, but i don't think it's an asterisk thing |
| 19:30.40 | *** join/#asterisk-dev kpfleming|laptop (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) |
| 19:30.40 | *** mode/#asterisk-dev [+o kpfleming|laptop] by ChanServ |
| 19:31.59 | *** join/#asterisk-dev bkw_ (n=brian@freeswitch/developer/bkw) |
| 19:33.48 | *** join/#asterisk-dev harry_v (n=pcsuppor@S0106001d7e52cc78.vs.shawcable.net) |
| 19:47.08 | putnopvut | harry_v: My suggestion when any sort of build problem happens is to try a make distclean and then try rebuilding. I agree with seanbright that the problem appears to not be an Asterisk problem though. |
| 19:47.39 | seanbright | likes it when putnopvut agrees with him |
| 20:00.15 | harry_v | okay |
| 20:00.28 | harry_v | never done this with linux but will give it a shot. |
| 20:00.40 | seanbright | done what with linux? |
| 20:01.00 | harry_v | distclean |
| 20:01.13 | seanbright | he is suggesting doing that with asterisk, not linux |
| 20:01.18 | seanbright | cd /path/to/asterisk/src |
| 20:01.20 | seanbright | make distclean |
| 20:01.26 | seanbright | ./configure |
| 20:01.28 | seanbright | etc, etc |
| 20:01.31 | harry_v | k |
| 20:02.00 | seanbright | 'make distclean' = 'start over from scratch' |
| 20:11.17 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 20:29.30 | *** join/#asterisk-dev harry_v (n=pcsuppor@S010600a0c93f6f7e.vs.shawcable.net) |
| 20:47.12 | seanbright | putnopvut: what are the chances of making the IMAP and ODBC support for voicemail loadable at runtime? |
| 20:47.31 | seanbright | without resorting them app_voicemail_imap.c and app_voicemail_odbc.c, etc. |
| 20:47.35 | russellb | It is theoretically possible. |
| 20:47.48 | seanbright | it would certainly clean that code up |
| 20:48.01 | russellb | There was an effort (ast_storage) to abstract it out |
| 20:48.02 | putnopvut | I'm clueless as far as ODBC goes, but I don't think it can be done with IMAP. |
| 20:48.15 | russellb | but that effort aimed to abstract it out a lot more that would be absolutely necessary ... |
| 20:48.17 | putnopvut | Because the IMAP library is statically linked. |
| 20:48.26 | seanbright | i've kinda given up on my inotify to replace pbx_spool thing... i don't think there is really a win there. so i'm looking for something else :0 |
| 20:48.32 | russellb | putnopvut: you would statically link it into the loadable module that provides the IMAP storage .. |
| 20:48.41 | seanbright | right |
| 20:48.52 | russellb | app_voicemail would accept storage provider registrations at load time |
| 20:48.57 | seanbright | OR roll your own imap library |
| 20:48.59 | seanbright | kidding |
| 20:49.01 | russellb | or perhaps only a single storage backend registration ... |
| 20:49.02 | putnopvut | russellb: ah, gotcha. I thought he just meant keeping app_voicemail the way it is but dynamically linking libcclient.a |
| 20:49.02 | russellb | shrugs |
| 20:49.09 | russellb | nah |
| 20:49.19 | putnopvut | seanbright: I actually considered it a couple of times. |
| 20:49.22 | putnopvut | But it's not worth it. |
| 20:49.25 | seanbright | ast_storage... is that a group/ proj? |
| 20:49.30 | Qwell | seanbright: yeah |
| 20:49.30 | putnopvut | group branch |
| 20:49.37 | Qwell | very outdated I'm sure |
| 20:49.37 | putnopvut | it's insanely out of date |
| 20:49.48 | seanbright | did mvanbaak fix svnmerge yet? |
| 20:50.02 | russellb | I think it's doable with something that doesn't go quite as far as ast_storage ... |
| 20:50.20 | seanbright | i started working on it a while back |
| 20:50.44 | russellb | devil's in the details, probably |
| 20:50.44 | seanbright | and also a way to abstract the queue strategies out |
| 20:50.44 | seanbright | but then... well... i lost interest |
| 20:54.50 | *** join/#asterisk-dev sivadnz (n=sivad@202-78-149-14.cable.telstraclear.net) |
| 21:00.52 | *** join/#asterisk-dev harry_v (n=pcsuppor@S010600a0c93f6f7e.vs.shawcable.net) |
| 21:01.09 | *** part/#asterisk-dev harry_v (n=pcsuppor@S010600a0c93f6f7e.vs.shawcable.net) |
| 21:11.33 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 21:11.59 | seanbright | Qwell: those 64 bit centos package dependency issues |
| 21:12.06 | Qwell | kill them. |
| 21:12.07 | seanbright | do you know if they are resolved in 5.2? |
| 21:12.13 | Qwell | no, they're caused by 5.2. |
| 21:12.13 | sasargen | putnopvut: just wanted to mention, line 319 of latest voicemail.conf.sample in branch/1.6.0 still has one instance of per-context option setting that probably doesn't work. could you remove that one too when you get a chance? |
| 21:12.19 | seanbright | oh great |
| 21:12.34 | putnopvut | sasargen: I'll give that a look. Thanks! |
| 21:12.48 | sasargen | putnopvut: np |
| 21:13.00 | putnopvut | Ah yes, the tz thing. |
| 21:14.31 | seanbright | Qwell: short of "don't install 5.2" is there a safe way? |
| 21:14.49 | Qwell | seanbright: I just do a `yum remove *.i386` after install, before doing anything... |
| 21:15.03 | seanbright | serious? |
| 21:15.10 | Qwell | maybe *.i686 too.. I forget if .i386 satisfies those or not |
| 21:15.17 | Qwell | OR |
| 21:15.26 | Qwell | just *always* add .x86_64 to your yum installs |
| 21:15.30 | Qwell | yum install curl-devel.x86_64 |
| 21:15.35 | seanbright | right |
| 21:15.37 | seanbright | ok |
| 21:17.11 | seanbright | thank you sir |
| 21:27.20 | *** part/#asterisk-dev sivadnz (n=sivad@202-78-149-14.cable.telstraclear.net) |
| 21:33.20 | *** join/#asterisk-dev Marquis42 (n=root@65-127-126-34.dia.static.qwest.net) |
| 21:47.40 | seanbright | "In the new yum (3.2) not specifying the arch will cause it to take all available archs it can find for the package. The old yum (the one in 5.0 and 5.1) in that case will default to the arch of the running kernel (only x86_64)." |
| 21:47.50 | seanbright | "This it not a real bug" |
| 21:47.54 | seanbright | pffffffffffft! |
| 21:47.58 | seanbright | how is that not a bug? |
| 21:48.18 | Qwell | that comment is incorrect |
| 21:48.25 | *** join/#asterisk-dev snuff-home (n=snuffy@60-240-117-7-nsw-pppoe.tpgi.com.au) |
| 21:48.25 | *** mode/#asterisk-dev [+o snuff-home] by ChanServ |
| 21:48.29 | Qwell | it doesn't "take all available arches" |
| 21:48.38 | Qwell | it takes "whatever the hell it wants" |
| 21:48.51 | seanbright | that makes *no* sense |
| 21:48.56 | Qwell | the problem is that i386 packages satisfy deps for x86_64 packages |
| 21:48.58 | Qwell | which is stupid |
| 21:49.13 | Qwell | so, you install curl-devel, it grabs x86_64 and i386, but ONLY i386 of curl |
| 21:49.15 | seanbright | so maybe i just install 5.1 to be safe from all of those |
| 21:49.29 | Qwell | but, if you always specify arch, it'll always get only that arch (which is good) |
| 21:50.00 | seanbright | is that the behavior going forward? |
| 21:50.06 | seanbright | they are not going to change it? |
| 21:50.07 | Qwell | got me |
| 21:50.13 | Qwell | I think they fixed something in F9 |
| 21:50.26 | seanbright | so lame. |
| 21:50.52 | Qwell | very |
| 21:51.36 | seanbright | i might give up computers. |
| 21:52.00 | seanbright | i think we'd all be better off. |
| 21:52.01 | Qwell | could go live in a monastery |
| 21:52.20 | putnopvut | Go to one of those monasteries where they brew beer |
| 21:52.42 | seanbright | do they have one where they brew beer *and* you can have sex with ladies/ |
| 21:52.45 | seanbright | if so, i'm in. |
| 21:53.03 | putnopvut | seanbright: we can start our own sect |
| 21:53.30 | file | O.O |
| 21:53.32 | seanbright | we'll need a name... that could take years to come up with |
| 21:53.36 | Qwell | putnopvut: I can't see anything at all that could possibly ever go wrong. |
| 21:54.04 | putnopvut | neither could i |
| 21:54.38 | seanbright | putnopvut: so should i consider your "ship it" as gospel? |
| 21:54.43 | seanbright | you'll take the blame if something is wrong? |
| 21:55.07 | putnopvut | seanbright: Of course I think you should take my ship it as gospel. |
| 21:55.16 | putnopvut | If it means taking the blame, then sure, I'll take it. |
| 21:55.20 | seanbright | sweet |
| 21:56.06 | seanbright | my *only* concern would be if we ever wanted to add yet another argument |
| 21:56.15 | putnopvut | Why would we want to do that? |
| 21:56.19 | seanbright | shrugs |
| 21:56.26 | putnopvut | You can always cram it in a struct and put it in the one argument we already added. |
| 21:56.33 | seanbright | good point |
| 21:56.48 | Marquis42 | So... I'm trying to fix a bug in func_odbc. Anybody here with some knowledge of thread storage and/or ast_str have a free moment to talk about it? |
| 21:57.01 | seanbright | Corydon76-dig: ping |
| 21:57.13 | Corydon76-dig | pong |
| 21:57.19 | seanbright | Corydon76-dig: Marquis42 wants to talk to you |
| 21:57.51 | Corydon76-dig | ~ask |
| 21:57.52 | jbot | well, ask is Questions in the channel should be specific, informative, complete, concise, and on-topic. Don't ask if you can ask a question first. Don't ask if a person is there; just ask what you intended to ask them. Better questions more frequently yield better answers. We are all here voluntarily or against our will. |
| 21:59.24 | seanbright | Marquis42: ask your question |
| 21:59.29 | Corydon76-dig | I love when people say "I have a question" and then don't ever type out their question |
| 21:59.30 | Marquis42 | In acf_odbc_read, there is a bug where it builds the comma-delimited list of column names. Actually, two... First, ast_str doesn't get reset and so the function just appends the column names to whatever is already in that buffer. Second, it |
| 22:00.50 | Marquis42 | The second bug is that it seems to copy the \0 into the buffer and so when that is copied into the chanell variable you only ever get the first column name rather than all of them. |
| 22:00.50 | seanbright | putnopvut: do you think there is *any* benefit to breaking ao2_callback_fn into ao2_callback_fn and ao2_compare_fn? |
| 22:01.14 | russellb | does not |
| 22:01.31 | seanbright | asked putnopvut |
| 22:01.35 | seanbright | :P |
| 22:01.36 | russellb | :-p |
| 22:01.47 | seanbright | ok, that's good enough for me |
| 22:01.53 | putnopvut | seanbright: well, not really, to be honest. |
| 22:02.24 | Marquis42 | Sorry, the variable is call colnames (ast_str is obviously the type of structure...). |
| 22:03.53 | russellb | seanbright: putnopvut: especially since the way cmp is used, it's just another ao2_callback use in main/astobj2.c ... |
| 22:05.59 | putnopvut | Marquis42: it looks like Corydon76-dig just fixed those bugs. |
| 22:06.03 | Corydon76-dig | Marquis42: thank you; fixed in SVN trunk and 1.6.1 |
| 22:06.06 | Marquis42 | Yeah, I saw the commit. Thank you. |
| 22:06.18 | seanbright | ah. it will just be an unused arg for all the *_cmp_* funcs, then. |
| 22:06.50 | russellb | unless you want to provide custom data to control how your compare func compares |
| 22:06.53 | Corydon76-dig | loves unused args |
| 22:06.53 | Marquis42 | FWIW, you also answered teh *actual* question I had about howt o fix the bug. That was pretty much what I had done, so good to see I wasn't off in the weeds. |
| 22:07.15 | *** join/#asterisk-dev harry_v (n=pcsuppor@S010600a0c93f6f7e.vs.shawcable.net) |
| 22:07.44 | Corydon76-dig | Marquis42: good job tracking down the actual issue. :-) |
| 22:07.52 | Marquis42 | Thanks |
| 22:08.52 | seanbright | russellb: but you can't pass anything to it |
| 22:08.54 | Corydon76-dig | I wish all bug reports were like that |
| 22:09.10 | russellb | seanbright: sure you can, if you use ao2_callback directly |
| 22:09.16 | Marquis42 | :) I try to be fairly specific if I can be. |
| 22:09.16 | seanbright | right |
| 22:09.23 | russellb | (or if we add an arg to ao2_find, which is something i put on your review) |
| 22:09.33 | seanbright | that has already been added |
| 22:09.39 | russellb | so there you go, you can do it |
| 22:09.45 | seanbright | ok. |
| 22:10.04 | harry_v | seanbright well, no go from your last instructions. |
| 22:10.08 | harry_v | same issue. |
| 22:10.15 | seanbright | harry_v: i'm fired |
| 22:10.55 | harry_v | I guess in this case from a time wise standpoint just replace linux all over again but that would seem kind of a waste. |
| 22:11.04 | harry_v | no one seems to understand the issue |
| 22:11.13 | *** join/#asterisk-dev atis_work (n=atis_wor@193.238.212.171) |
| 22:11.27 | putnopvut | harry_v: yeah, that error is really weird. |
| 22:11.34 | harry_v | yours was the first set of instructions that could have made a difference. |
| 22:11.59 | Marquis42 | Wait, I don't think that func_odbc fix is quite right. Now I get every character in the field name twice. |
| 22:12.20 | Marquis42 | Like this: ~ODBCFIELDS~=mmeessssaaggee__iidd,mmaaiillbbooxx,ccoonntteexxtt |
| 22:13.29 | snuff-home | heh.. stuttering odbc.. |
| 22:13.29 | harry_v | now I can resinstall linux on top of linux or fresh install? I dont feel like going to the next room and installing though the display but rather ssh in and do it. |
| 22:13.29 | Marquis42 | What I had done, specifically, is flip the check for \0 with the assignment of the character. That seemed to work. |
| 22:14.20 | Marquis42 | Wait, nevermind... :/ My fault. |
| 22:14.21 | jsmith | harry_v: That question would be better asked in another channel, but in short, you can start linux installing in the other room, them watch it/finish it over VNC |
| 22:16.10 | wedhorn | mvanbaak: ping |
| 22:20.21 | Corydon76-dig | Marquis42: yeah, I just tried it and didn't get that |
| 22:20.45 | Marquis42 | Yup, it was unquestionably user error. |
| 22:21.28 | Marquis42 | In my defense, I haven't slept in 38 hours. ;) |
| 22:29.37 | *** join/#asterisk-dev bkw_ (n=brian@freeswitch/developer/bkw) |
| 22:30.53 | seanbright | "It can only be set in the general |
| 22:30.54 | seanbright | section or per-mailbox." |
| 22:31.09 | seanbright | we might just want to go ahead and make that possible |
| 22:34.29 | putnopvut | seanbright: feel free |
| 22:34.32 | seanbright | heh |
| 22:34.43 | seanbright | well by "we" i meant... someone that wasn't me |
| 22:34.44 | seanbright | :) |
| 22:34.52 | seanbright | i'm lazy |
| 22:35.02 | putnopvut | seanbright: yeah, any chance I get to not touch app_voicemail.c is an opportunity I am glad to take. |
| 22:35.07 | seanbright | heh |
| 22:35.09 | seanbright | fair enough |
| 22:40.10 | seanbright | 2 down, 1 to go |
| 23:35.09 | *** join/#asterisk-dev sasargen (n=chatzill@174-146-69-119.pools.spcsdns.net) |
| 23:48.37 | *** join/#asterisk-dev outtolunc (n=me@c-67-188-204-139.hsd1.ca.comcast.net) |