| 00:00.00 | kpfleming | yeah... nothing happening on the serial console at all |
| 00:00.14 | Qwell | can you do a #. ? |
| 00:00.23 | Qwell | or ctrl-a,f |
| 00:00.26 | kpfleming | yes, at the sc> prompt now |
| 00:00.34 | Qwell | poweron |
| 00:00.36 | Qwell | console -f |
| 00:00.47 | kpfleming | it has reset |
| 00:00.51 | Qwell | excellent |
| 00:00.53 | kpfleming | will see what happens |
| 00:01.00 | kpfleming | it will only take 20 minutes to get linux running |
| 00:01.17 | seanbright | digium should buy me a sparc |
| 00:01.27 | seanbright | just cuz. |
| 00:02.02 | kpfleming | you don't want one |
| 00:02.03 | file | sexy party in the collab area... or sparc for seanbright |
| 00:02.03 | snuff-work | mm.. i use a v240 as my test box.. |
| 00:02.07 | kpfleming | at least not a niagara |
| 00:02.14 | file | I choose sexy party |
| 00:02.32 | Qwell | seanbright: http://www.sun.com/tryandbuy/ |
| 00:02.36 | snuff-work | gets to do training next week on the big M series server |
| 00:02.47 | Qwell | kpfleming: supposedly the newer Niagara's are much better with floating point |
| 00:03.05 | Qwell | at least, that's what Jonathan told me like...18 months ago |
| 00:03.06 | seanbright | file: but it's my birthday! |
| 00:03.07 | snuff-work | qwell they are 'less hopeless' |
| 00:03.16 | file | seanbright: pfft |
| 00:03.47 | snuff-work | will be interesting with their ROCK processor.. but that wont be out till q1 2009 if we're lucky |
| 00:03.51 | kpfleming | they are at least aware of what floating point is |
| 00:03.57 | seanbright | Qwell: i'm not shady like you |
| 00:04.06 | Qwell | they didn't take my calls! |
| 00:04.10 | Qwell | not my fault. |
| 00:06.45 | kpfleming | it is just about to load SILO |
| 00:06.58 | kpfleming | good thing i have not been holding my breath |
| 00:07.01 | Qwell | or is it going to reset again? |
| 00:07.20 | kpfleming | no SILO just loaded |
| 00:07.39 | Qwell | I wonder if it's smart enough to update the SILO configs |
| 00:08.53 | kpfleming | got an oops while the kernel was initializing |
| 00:09.11 | kpfleming | null pointer dereference |
| 00:09.19 | kpfleming | that is ungood |
| 00:10.26 | snuff-work | mm cant be too healthy.. |
| 00:10.58 | kpfleming | will try again with LinuxOLD |
| 00:11.17 | kpfleming | but i wonder if the 2006/03/02 firmware in this system should be updated :-) |
| 00:12.11 | snuff-work | mm couldn't hurt.. |
| 00:12.13 | file | crazy talk |
| 00:17.47 | kpfleming | ok, LinuxOLD attempting to boot now |
| 00:17.59 | kpfleming | is doing a firmware upgrade a 'burn ISO to CD and boot from it' sort of process? |
| 00:18.00 | *** part/#asterisk-dev Deeewayne (n=Deeewayn@nat/digium/x-6e72f799222c11d5) |
| 00:18.10 | Qwell | I have no idea.. |
| 00:19.15 | snuff-work | mm unsure in linux.. from my bad memory i was able to update firmware from within solaris.. |
| 00:19.21 | kpfleming | hmm |
| 00:19.24 | snuff-work | u might even able able to update it from the service console |
| 00:19.25 | kpfleming | i doubt it can be done from linux |
| 00:19.25 | Qwell | I imagine you do it from OBP |
| 00:19.33 | Qwell | http://sunsolve.sun.com/handbook_pub/validateUser.do?target=Systems/SunFireT2000/SunFireT2000 |
| 00:20.04 | Qwell | it might be the _R model actually |
| 00:20.20 | kpfleming | it's running with the 2.6.18 kernel |
| 00:20.32 | kpfleming | don't know what else to try to get 2.6.26 to boot, should probably start with firmware updates |
| 00:20.39 | file | hammer. |
| 00:21.27 | Qwell | 6.6.7 is Oct 29, 2008 |
| 00:24.07 | Qwell | hmm... do I even have a sunsolve account? |
| 00:24.19 | snuff-work | if u want i can dl for u.. |
| 00:26.10 | Qwell | nope, I got it |
| 00:26.13 | kpfleming | ok |
| 00:26.18 | Qwell | java archive...wtf |
| 00:26.53 | Qwell | "Please refer to the Install.info file for instructions on updating [...]" |
| 00:27.01 | Qwell | if I can get it... stupid java |
| 00:28.04 | Qwell | <PROTECTED> |
| 00:29.35 | Qwell | looks easy enough. requires a local ftp server, but that's doable |
| 00:34.50 | *** join/#asterisk-dev bbryant (n=brett@c-68-59-20-153.hsd1.sc.comcast.net) |
| 00:34.50 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 00:45.34 | *** join/#asterisk-dev eliel (n=eliel@151-202-114-200.fibertel.com.ar) |
| 00:45.34 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 00:57.53 | kpfleming | seanbright: inotify! neato |
| 00:59.48 | file | kpfleming: shhhhhhhh! private |
| 00:59.58 | seanbright | damn it |
| 00:59.59 | seanbright | heh |
| 01:00.14 | seanbright | i forgot you guys could see that stuff |
| 01:00.17 | seanbright | sigh. |
| 01:00.34 | file | we see all. |
| 01:05.09 | kpfleming | oh yea, sorry |
| 01:05.17 | seanbright | no problem :) |
| 01:05.25 | seanbright | only now i actually have to finish it |
| 01:05.28 | file | do remember to autoconf that up |
| 01:05.44 | seanbright | yeah, and going to make it a runtime option as well |
| 01:06.12 | seanbright | apparently it is problematic with some filesystems |
| 01:08.35 | seanbright | another thing i would like to do... or at least discuss at some point... is standardized filename and path handling |
| 01:09.20 | seanbright | right now we do PATH_MAX, FILENAME_MAX, and sometimes we just do 'char fn[256];' |
| 01:11.01 | seanbright | but everything i have found on t3h interwebs (like the glibc docs) say to always dynamically allocate buffers for filenames |
| 01:11.08 | seanbright | but... that is for another day |
| 01:15.34 | *** join/#asterisk-dev tacubo (n=tacvbo@189.146.192.181) |
| 01:22.50 | *** join/#asterisk-dev russellb (n=russell@asterisk/digium-open-source-team-lead/russellb) |
| 01:22.50 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 01:30.14 | eliel | russellb: i am uploading review requests but maybe this is not the way you would like me to handle this stuff, you prefer to continue submitting issues on the bug tracker or use the review board for this type of thing related to documentation that needs a second opinion |
| 01:30.40 | eliel | s/you prefer/you prefer me/ |
| 01:31.19 | *** join/#asterisk-dev stevie_ramjet (n=putnopvu@c-71-228-178-34.hsd1.al.comcast.net) |
| 01:31.19 | *** mode/#asterisk-dev [+o stevie_ramjet] by ChanServ |
| 01:31.45 | seanbright | if only putnopvut was here... |
| 01:31.50 | russellb | eliel: I don't mind you creating requests |
| 01:32.08 | eliel | k |
| 01:32.11 | russellb | I have not yet decided on the workflow i want to use for when to use this vs. mantis etc ... |
| 01:32.23 | russellb | and also, I have ways to prioritize mantis issues |
| 01:32.34 | russellb | but how to prioritize stuff on reviewboard along side mantis ... haven't figured that out yet |
| 01:32.38 | russellb | so anyway, feel free to use it for now |
| 01:32.42 | russellb | there are no rules yet :) |
| 01:32.48 | stevie_ramjet | seanbright, :) |
| 01:36.02 | eliel | i was thinking about the review board as a second step, we submit an issue, we (bug marshals) have a link near the submitted patch, and pressing that could post a review (for the kind of patches that needs those reviews), but always first submitting the issue in the bug tracker |
| 01:36.32 | russellb | yeah |
| 01:39.05 | eliel | also the ship it could be posted to the bug tracker with all the reviews and the last patch submitted to the review board :) (ok, stop dreaming) |
| 01:41.28 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.168) |
| 01:42.36 | russellb | the problem is, all of this takes work |
| 01:42.36 | russellb | heh |
| 01:42.40 | russellb | and i don't know PHP :( |
| 01:42.52 | russellb | but I know python basics, so i'm working on the reviewboard side of things right now |
| 01:44.35 | eliel | hmmm |
| 01:46.44 | russellb | However, I'm going to work on documenting the plans for this integration |
| 01:46.53 | russellb | and if I can spell it out enough, I can probably find some help :) |
| 01:47.28 | eliel | ok, I will checkout repotools and see if i can help with the php (ugly) code |
| 01:48.05 | eliel | on the weekend |
| 01:48.15 | russellb | you don't have to! |
| 01:48.24 | russellb | you've taken on enough work :) |
| 01:48.28 | russellb | you need sleep, too, you know |
| 01:49.33 | eliel | i could implement that integration in my job |
| 01:50.06 | eliel | s/implement/use/ |
| 01:51.38 | russellb | luckily they both have a web api |
| 01:51.53 | russellb | so it should be relatively straight forward to submut review requests from mantis ... |
| 01:52.31 | russellb | the post-review script might be a helpful starting point to know what to do .. |
| 01:53.31 | eliel | yes, the only "problem" will be the authentication |
| 01:53.47 | russellb | yeah ... |
| 01:53.51 | russellb | that's what i'm working on right now |
| 01:54.01 | russellb | a new auth backend for reviewboard that will use the mantis users database |
| 01:55.33 | jsmith | Oooh, fun stuff |
| 01:55.45 | eliel | that will simplify things a lot |
| 01:56.56 | *** join/#asterisk-dev Deeewayn1 (n=dwayne@76.29.245.9) |
| 01:56.56 | *** mode/#asterisk-dev [+o Deeewayn1] by ChanServ |
| 02:07.24 | *** join/#asterisk-dev outtolunc (n=me@c-67-188-204-139.hsd1.ca.comcast.net) |
| 02:11.22 | eliel | see ya, good luck with your new president |
| 02:11.34 | russellb | thanks :) |
| 02:15.40 | Juggie | oh ya who's winning so far |
| 02:16.06 | Juggie | heh |
| 02:16.12 | snuff-work | who u think.. |
| 02:16.14 | Juggie | how does mccain even get any votes |
| 02:16.26 | snuff-work | its called the guy he's up against is black |
| 02:16.35 | Juggie | the crazy thing is look @ the numbers, they are @ like 50/50 vote percentage |
| 02:16.50 | stevie_ramjet | Juggie, it's almost always like that. |
| 02:17.16 | stevie_ramjet | Based on the early results, I'd have to say Obama has a nice advantage right now. |
| 02:17.25 | stevie_ramjet | (I'm assuming he'll get California) |
| 02:17.45 | Juggie | the problem is we let old people vote |
| 02:17.55 | outtolunc | haha |
| 02:18.02 | Juggie | look @ the exit stats though |
| 02:18.13 | Juggie | the old people vote is like 30/70 obama/mccain |
| 02:18.16 | Juggie | if your an old fart |
| 02:19.02 | Juggie | and on the other end, young people voted for obama |
| 02:19.23 | snuff-work | yer.. old ppl.. |
| 02:19.41 | snuff-work | much greater agnst against a brother.. |
| 02:20.12 | Juggie | ya, alot of old people are much more discriminating |
| 02:21.07 | snuff-work | like i know a lot of older ones.. are very 'set' in what they want |
| 02:40.35 | codefreeze-lap | is trying to figure why the "alertpipe" is clogging and giving thousands of error messages in __ast_read.... hmmm |
| 02:40.58 | russellb | clogging? |
| 02:41.52 | codefreeze-lap | russellb: well, the message comes out: read() failed: Resource temporarily unavailable... both 1.4 and trunk... |
| 02:42.40 | jsmith-away | hands codefreeze-lap a bottle of Drano and a big snake |
| 02:42.56 | codefreeze-lap | channel.c: 2041 in 1.4 |
| 02:43.42 | codefreeze-lap | 64 bit machine. Problem started very recently |
| 02:58.32 | codefreeze-lap | drinks the Drano and starts chewing on the snake |
| 02:58.40 | *** join/#asterisk-dev DarkRift (n=dark@bas1-sthubert21-1096612081.dsl.bell.ca) |
| 03:38.18 | codefreeze-lap | russellb: ahhhh, I think I see: this was one of Kevin's fixes where the results of read() were being ignored; in this case, tho, apparently, an error is common and no big deal (apparently???) |
| 03:39.09 | russellb | Yes, we should not be printing an error if errno is EINTR or EAGAIN |
| 03:59.21 | stevie_ramjet | dangit |
| 03:59.32 | stevie_ramjet | I can't svnmerge init on the new branch I just created. |
| 04:00.03 | stevie_ramjet | GnuTLS internal error |
| 04:09.34 | lmadsen | does anyone know why asterisk might clear registrations out of the database even when I have ignoreregexpire=yes and rtautoclear=3600, even when the phones register every 30 seconds? |
| 04:10.17 | lmadsen | I get a query such as: UPDATE sipfriends SET fullcontact='', ipaddr='', port='', regseconds='0', username='', regserver='' WHERE name |
| 04:10.19 | lmadsen | ='0004f216ae8d' |
| 04:21.18 | *** join/#asterisk-dev jameswf (n=james@ip68-3-60-66.ph.ph.cox.net) |
| 04:24.07 | *** join/#asterisk-dev CunningPike (n=arodgers@S010600095b33697f.vc.shawcable.net) |
| 05:56.07 | *** join/#asterisk-dev obnauticus (n=lol@about/windows/regular/obnauticus) |
| 06:36.23 | *** join/#asterisk-dev kamanashisroy (n=kamanash@202.56.7.182) |
| 06:57.49 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@local.xorcom.com) |
| 06:57.49 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 07:13.40 | *** join/#asterisk-dev philipp64 (n=chatzill@63.224.43.239) |
| 07:54.47 | *** join/#asterisk-dev mvanbaak (i=vanbaak@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 07:54.47 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 07:57.13 | wedhorn | mvanbaak: ping |
| 07:58.55 | wedhorn | M13806 |
| 07:58.56 | MuffinMan | [assigned] [Asterisk] Channels/chan_skinny 0013806: codecs settings does work only in device specific section reported by pj (Karma: +9.00) http://bugs.digium.com/view.php?id=13806 |
| 07:59.44 | wedhorn | I've been playing with this one and the codec pref stuff is a bit ordinary in that the prefs are actually overwritten by what the device can do |
| 08:00.03 | wedhorn | and the line codecs are overwritten by the device codecs |
| 08:00.58 | wedhorn | Sort of stumped as to whether I should fix, or just remove the line specific stuff and attach codec choice to devices only. Any thoughts, when you get a chance. |
| 08:04.22 | mvanbaak | wedhorn: pong |
| 08:04.25 | mvanbaak | interesting :) |
| 08:04.37 | mvanbaak | we should fix it if it's different then before the patch |
| 08:05.47 | *** join/#asterisk-dev steliosk (n=Stelios@ipa107.2.tellas.gr) |
| 08:09.59 | steliosk | tzafrir_laptop : Morning ! How stable you think zaphfc would be if we use the kernel clock for timing, instead of the build in 8Khz clock ? |
| 08:10.27 | tzafrir_laptop | steliosk, hmm.. no idea |
| 08:10.32 | steliosk | something similar to the ztdummy does to get timing but actually run an isr |
| 08:10.46 | tzafrir_laptop | I know many people have a paroblem with using the system clock at all on some kernels |
| 08:11.33 | tzafrir_laptop | and I also suspect that the system clock and the card's clock (or the provider's clock) would drift a bit |
| 08:13.13 | steliosk | that's sure but even in this setup we endup loosing packages anyway |
| 08:13.13 | steliosk | i.e this=existing |
| 08:16.31 | steliosk | BTW is there any example of how to make an app or codec execute only one instance at a time ? |
| 08:16.49 | wedhorn | mvanbaak: we sort of caused it because we split up lines and devices |
| 08:17.12 | steliosk | i am doing some experiments using nidia cuda and it *seriously* dislikes multithreding |
| 08:19.28 | wedhorn | mvanbaak: I think the question really is: do we want to be able to specify different codecs for each line on a device? |
| 08:24.35 | tzafrir_laptop | steliosk, BTW: did you have some time to look at the new digium bri dahdi driver? |
| 08:26.39 | steliosk | tzafrir_laptop : not actually. i am working on an assembly speedup for oslec for powerpc |
| 08:27.19 | steliosk | tzafrir_laptop : need to get it down to same mips as x86 |
| 08:27.32 | tzafrir_laptop | I gather that on powerpc the floating-point unit is not as badly-desgined as the one in x86? |
| 08:28.17 | tzafrir_laptop | so you can actually use it in the kernel? |
| 08:28.27 | steliosk | well the one i got has no fpu :( |
| 08:29.10 | steliosk | low power comes at a cost.... |
| 08:33.09 | mvanbaak | wedhorn: hhmm, maybe |
| 08:33.22 | mvanbaak | wedhorn: that way one can test different codecs with one device |
| 08:33.44 | mvanbaak | wedhorn: I use it sometimes with the SIP channel to see why a box has trouble with say gsm codec |
| 08:35.04 | tzafrir_laptop | lowpowerpc |
| 09:25.36 | wedhorn | mvanbaak: ok, I'll fix it up, will just need to add another var to lines to keep the prefs. Maybe I should add it to devices as well for realtime |
| 09:33.18 | mvanbaak | yup |
| 09:33.21 | mvanbaak | that sounds good |
| 09:41.34 | *** join/#asterisk-dev ccesario (n=ccesario@189-19-9-100.dsl.telesp.net.br) |
| 10:00.06 | *** join/#asterisk-dev steliosk (n=Stelios@ipa107.2.tellas.gr) |
| 10:19.17 | *** join/#asterisk-dev gmaruzz (n=gmaruzz@213-140-6-122.ip.fastwebnet.it) |
| 10:19.57 | *** part/#asterisk-dev gmaruzz (n=gmaruzz@213-140-6-122.ip.fastwebnet.it) |
| 10:23.57 | *** join/#asterisk-dev DarKnesS_WolF (n=sherif@unaffiliated/sherif) |
| 10:24.18 | DarKnesS_WolF | mmm noone will review this :-) ? http://bugs.digium.com/view.php?id=13667 |
| 10:24.56 | DarKnesS_WolF | tzafrir_laptop: hey how are u ? |
| 10:25.02 | tzafrir_laptop | hi |
| 10:31.48 | *** join/#asterisk-dev eliel (n=eliel@200.61.172.61) |
| 10:31.48 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 11:21.54 | *** join/#asterisk-dev caio1982 (i=caio1982@CAcert-br/caio1982) |
| 11:30.15 | *** join/#asterisk-dev dioedu (n=dioedu@201.7.117.114) |
| 11:39.07 | *** join/#asterisk-dev mvanbaak (i=vanbaak@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 11:39.07 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 12:05.31 | *** join/#asterisk-dev boolean12 (n=boolean1@tandem.uplinktel.com) |
| 12:28.17 | *** join/#asterisk-dev DarkRift (n=dark@65.92.165.55) |
| 12:40.45 | *** join/#asterisk-dev IgorG (n=IgorG@host-92-126-202-223.pppoe.omsknet.ru) |
| 12:43.07 | *** join/#asterisk-dev random_wolf (n=boolean1@tandem.uplinktel.com) |
| 13:05.02 | *** join/#asterisk-dev sasargen (n=chatzill@174-146-187-150.area3.spcsdns.net) |
| 13:26.57 | *** join/#asterisk-dev anonymouz666 (n=anonymou@201.19.199.65) |
| 13:32.55 | *** join/#asterisk-dev lmadsen (n=Leif@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:32.55 | *** mode/#asterisk-dev [+o lmadsen] by ChanServ |
| 13:40.48 | *** join/#asterisk-dev Deeewayn1 (n=dwayne@76.29.245.9) |
| 13:40.49 | *** join/#asterisk-dev pgarcia (n=chatzill@yoda.kanatek.com) [NETSPLIT VICTIM] |
| 13:40.49 | *** join/#asterisk-dev mbranca (n=matteo@93-62-230-243.ip24.fastwebnet.it) [NETSPLIT VICTIM] |
| 13:40.49 | *** join/#asterisk-dev write__erase (n=write@royale.aixmarseille.com) [NETSPLIT VICTIM] |
| 13:40.49 | *** join/#asterisk-dev kpfleming (n=kpflemin@asterisk/digium-director-of-software-tech/kpfleming) [NETSPLIT VICTIM] |
| 13:40.49 | *** join/#asterisk-dev jpeeler (n=jpeeler@asterisk/digium-software-dev/jpeeler) |
| 13:40.49 | *** join/#asterisk-dev MuffinMan (n=muffinma@asterisk/issue-tracker-bot/muffinman) |
| 13:40.49 | *** join/#asterisk-dev corramor (n=corramor@nat/digium/x-d6c5048d458cbaea) [NETSPLIT VICTIM] |
| 13:40.49 | *** mode/#asterisk-dev [+ooo Deeewayn1 kpfleming jpeeler] by irc.freenode.net |
| 13:51.49 | *** join/#asterisk-dev Pazzo (n=ugelt@reserved-225136.rol.raiffeisen.net) |
| 13:57.18 | *** join/#asterisk-dev rbd (n=rbd@adsl-074-229-183-112.sip.rmo.bellsouth.net) |
| 14:10.32 | *** join/#asterisk-dev ctooley_ (n=ctooley@doc-24-32-196-69.concordia.ks.cebridge.net) |
| 14:27.55 | *** join/#asterisk-dev dbailey (n=doug@nat/digium/x-d1f20775ba6e33b8) |
| 14:30.18 | *** join/#asterisk-dev sergee (n=serg@voip1.west-call.com) |
| 14:50.18 | sergee | Corydon76-dig: hi there! if you have some free time, could you please review my little patch? M13775 Thank you! |
| 14:51.09 | *** join/#asterisk-dev smurf (n=smurf@debian/developer/smurf) |
| 14:51.48 | *** part/#asterisk-dev smurf (n=smurf@debian/developer/smurf) |
| 14:52.45 | *** mode/#asterisk-dev [+o ctooley] by ChanServ |
| 14:57.04 | *** join/#asterisk-dev smurfix (n=smurf@debian/developer/smurf) |
| 14:59.28 | *** join/#asterisk-dev anthm (n=anthm@freeswitch/developer/anthm) |
| 15:03.32 | jsmith | ~buy developers donut |
| 15:03.33 | jbot | ACTION goes to S-Mart and gets a dozen donuts for developers |
| 15:05.37 | *** join/#asterisk-dev putnopvut (n=putnopvu@nat/digium/x-2ffa7b3e8f8fcffc) |
| 15:05.37 | *** mode/#asterisk-dev [+o putnopvut] by ChanServ |
| 15:08.41 | *** join/#asterisk-dev russellb (n=russellb@asterisk/digium-open-source-team-lead/russellb) |
| 15:08.41 | *** mode/#asterisk-dev [+o russellb] by ChanServ |
| 15:21.02 | *** join/#asterisk-dev Deeewayne (n=Deeewayn@nat/digium/x-fa972e9e7e3881d2) |
| 15:21.02 | *** mode/#asterisk-dev [+o Deeewayne] by ChanServ |
| 15:22.22 | jets | Oh look... Donuts! |
| 15:22.42 | jsmith | ~buy jets donut |
| 15:22.43 | jbot | ACTION goes to S-Mart and gets a dozen donuts for jets |
| 15:37.38 | *** join/#asterisk-dev jameswf (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 15:39.14 | *** join/#asterisk-dev bartpbx (n=bartpbx@217.24.210.202) |
| 15:40.19 | bartpbx | hello, we updated to 1.4.22 last night and see some coredumps during the day related to the realtime iax2 prune. |
| 15:40.44 | bartpbx | I search the bugtracker but it looks like nobody else has seen it before |
| 15:40.51 | bartpbx | can anyone confirm this? |
| 15:42.11 | lmadsen | not sure I've seen anything like that, but you may want to verify you're still seeing it in the 1.4 branch to see if it was already resolved |
| 15:49.04 | jsmith | Whoa! That was weird... an email I sent more than a week ago just finally went through to the -users list |
| 15:49.07 | jsmith | blinks |
| 15:53.06 | putnopvut | weird |
| 15:53.11 | jameswf | jsmith: there was a change in the matrix |
| 15:53.33 | jsmith | jameswf: I guess so... |
| 15:55.03 | *** join/#asterisk-dev fred-tmft (n=fred-tea@c-98-243-174-131.hsd1.mi.comcast.net) |
| 15:55.16 | *** part/#asterisk-dev fred-tmft (n=fred-tea@c-98-243-174-131.hsd1.mi.comcast.net) |
| 16:16.38 | kpfleming | for those who watch svn-commits, yes, i just posted a bunch of old commit messages from the menuselect repo, it hadn't been sending email for quite some time |
| 16:17.44 | *** join/#asterisk-dev bweschke (n=BJ_Wesch@74.94.3.214) |
| 16:17.44 | *** mode/#asterisk-dev [+o bweschke] by ChanServ |
| 16:19.28 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 16:19.28 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 16:19.35 | *** part/#asterisk-dev bartpbx (n=bartpbx@217.24.210.202) |
| 16:19.41 | jsmith | Good afternoon oej |
| 16:20.05 | oej | Good afternoon, happy Smith! |
| 16:20.28 | lmadsen | afternoon! |
| 16:20.49 | jsmith | oej: Well, not *that* happy... but hello... |
| 16:21.32 | lmadsen | happy happy joy joy |
| 16:21.57 | oej | Tag wars in svn... |
| 16:22.08 | jsmith | oej: Naw... |
| 16:22.17 | jsmith | oej: "kpfleming> for those who watch svn-commits, yes, i just posted a bunch of old commit messages from the menuselect repo, it hadn't been sending email for quite some time" |
| 16:22.52 | oej | Aha |
| 16:23.19 | oej | But all the autotag stuff wasn't menuselect, or was it? |
| 16:23.43 | russellb | it was |
| 16:24.01 | oej | OK then |
| 16:24.56 | kpfleming | yeah, the svnmailer config was slightly messed up, and i wanted to ensure that at least the last couple of months of commits got sent out |
| 16:25.27 | seanbright | and based on those e-mails, that seanbright guy is *awesome* |
| 16:25.36 | seanbright | seanbright <3 |
| 16:28.38 | seanbright | no? no takers? that was a softball, people! |
| 16:29.03 | seanbright | wanders off to a meeting |
| 16:29.24 | codefreeze-lap | I seek someone who knows how to use the Park() app usefully.... any such person present? |
| 16:30.41 | russellb | ~ask |
| 16:30.41 | jbot | extra, extra, read all about it, 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. |
| 16:30.43 | russellb | :-p |
| 16:30.50 | lmadsen | hits seanbright with a baseball bat |
| 16:31.51 | Corydon76-dig | seanbright: sorry, were you looking for some hot male-on-male loving? |
| 16:32.05 | russellb | lols |
| 16:36.54 | eliel | may I ship the makefile/configure for asterisk-addons, to start working there with the xml docs? |
| 16:37.27 | russellb | someone should take a look at that review and sign off on it first .. |
| 16:37.51 | codefreeze-lap | Then Ask() I shall. I have converted all the park stuff to use the masq_park stuff. One call remains, in park_exec... one channel is involved, not two. It return(s/ed) a KEEPALIVE, whether targeted for the bridge, or for the pbx, I cannot say right now. Do I need to masq_park there also? |
| 16:38.13 | eliel | russellb: k |
| 16:43.47 | eliel | cli permissions has that sign off, but when that review was made i didn't have commit access, and no one commit it, maybe because someone don't want to commit that code for some reason not explained in the reviewboard? |
| 16:44.33 | russellb | i haven't had time to look at it or think about it ... |
| 16:45.28 | eliel | np |
| 16:47.23 | russellb | I'd rather you hold off on that one for now |
| 16:50.35 | eliel | no problem, it has been there since 2007-10-30 with minor conflicts, it can wait longer :-) heh, we are using it in a productive enviroment since then. |
| 16:53.55 | *** join/#asterisk-dev outtolunc (n=me@adsl-66-218-53-172.dslextreme.com) |
| 17:06.11 | oej | eliel: The CLI permissions touches the security framework stuff, so we might consider how it fits together too. |
| 17:06.24 | oej | The architecture formerly called ASA |
| 17:07.11 | eliel | oej: you are talking about pinemango? |
| 17:08.37 | oej | Well, the security stuff was separated from Pinemango into it's own little corner of the world |
| 17:08.56 | oej | For me, it's a requirement to implement Pinemango. But that's my personal opinion |
| 17:09.15 | oej | See edvina.net/asterisk/asa-intro.pdf |
| 17:12.04 | *** join/#asterisk-dev CunningPike (n=arodgers@vpn.dnv.org) |
| 17:13.31 | *** join/#asterisk-dev IgorG (n=IgorG@host-92-126-202-223.pppoe.omsknet.ru) |
| 17:19.26 | eliel | oej: I have read the slides, the actual cli permissions architecture is very simple (+1 heh), it is based on the *nix users being passed with the group currently opening the local socket, based on this credentials asterisk check permissions |
| 17:19.34 | eliel | against the cli commands |
| 17:26.57 | tzafrir_laptop | eliel, well, we also don't have any equivalent of Exec in the cli |
| 17:29.45 | eliel | like '!' ? |
| 17:30.00 | tzafrir_laptop | ! never gets to the Asterisk daemon |
| 17:30.10 | eliel | yeaph, i know |
| 17:31.01 | eliel | why you will want an Exec in the cli? |
| 17:31.13 | tzafrir_laptop | I f the client could set aliases and play with variables, we might be in the same mess |
| 17:31.28 | tzafrir_laptop | and no: I *don't* see a point in that |
| 17:31.45 | tzafrir_laptop | I do appreciate the simplicity |
| 17:33.54 | eliel | tzafrir_laptop: no, it wont be a mess, cause i am checking permissions in ast_cli_command and cli_aliases are being translated before calling ast_cli_command |
| 17:34.03 | eliel | so, I will check permissions on the final command |
| 17:34.28 | eliel | so, you could use the aliases you want, we just set the security on the "real" command |
| 17:35.08 | eliel | offcourse it is supposed that the asterisk administrator knows this, before applying security restrictions... |
| 17:41.20 | *** join/#asterisk-dev ManxPower (n=manxpowe@223.sub-70-221-254.myvzw.com) |
| 17:42.37 | ManxPower | It might be helpful to people if a reminder of the location of the doc directory and a reference to the READMEs and UPGRADE.txt be displayed to the user when they do a make install. |
| 17:46.41 | seanbright | putnopvut: we need a static analysis tool for strcasecmp i think |
| 17:47.10 | seanbright | if (!(strcasecmp(foo, bar) == 0))) { |
| 17:47.18 | seanbright | something that will print out a human readable form of that |
| 17:47.20 | seanbright | :) |
| 17:51.44 | tzanger | http://www.freshdot.net/img/637/wtfm.jpg |
| 17:51.47 | tzanger | seriously lol. |
| 17:52.34 | caio1982 | ManxPower: you can send in the (possibily) 2-lines-patch for this, although i highly doubt a "user" issuing make install needs to read it in order to get informed, otherwise she/he will complain anyway as distro packaging "should" do that too |
| 17:52.48 | russellb | tzanger: i have that on my wall, heh |
| 17:53.03 | caio1982 | ManxPower: PS: i'm not really following up your discussion in #asterisk |
| 17:53.11 | ManxPower | caio1982: spend some time on #asterisk and you'll change your mind. |
| 17:55.30 | caio1982 | i might end up not reading the upgrade or changes files anymore :P |
| 17:56.17 | seanbright | ManxPower: one of us will get to it today |
| 17:56.41 | *** part/#asterisk-dev jsmith (n=njsmith@72.21.36.138) |
| 17:56.41 | ManxPower | seanbright: anything to reduce problems |
| 17:56.46 | seanbright | sure, makes sense to me |
| 17:57.07 | ManxPower | also remember that there are the previous upgrade.txts are important to read as well. |
| 17:57.11 | seanbright | yup |
| 17:57.20 | seanbright | i'll post a patch somewhere for your thumbs up |
| 17:58.04 | ManxPower | seanbright: dadhi "make install" should mention the README has the list of cards/drivers and the Asterisk make install should tell them all the major changes they have to know about are in the UPGRADE.txt files |
| 17:58.26 | seanbright | ok |
| 17:58.36 | ManxPower | over and over I see people on #asterisk having problems because they did not read those files |
| 18:06.02 | tzafrir_laptop | if people don't bother reading the README, what makes you think they actually read the text spewed by 'make install'? |
| 18:06.14 | seanbright | it won't hurt anything to add it |
| 18:06.22 | tzafrir_laptop | The longer it is, the less they read from it |
| 18:06.23 | seanbright | and if 1 user reads it, that is 1 less user to support |
| 18:08.09 | tzafrir_laptop | I suspect it already prints too much (and I'm not sure the part in the end of 'make config' is useful) |
| 18:08.09 | seanbright | well your suspicions and reality are not necessarily the same thing :) |
| 18:12.15 | ManxPower | If nothing else we can tell them they are an idiot for not even reading the stuff printed on the screen in a make install |
| 18:12.47 | seanbright | nothing says "welcome to the community!" like "you are an idiot" ;) |
| 18:13.08 | ManxPower | seanbright: you could also mention something like "If you are uprading from 1.4 many things will break if you don't read upgrade.txt |
| 18:15.33 | eliel | someone saw the output of the xml documentation with 'See Also' elements? I think it confuse a little, i think something more standard like: |
| 18:15.38 | eliel | [see also] |
| 18:15.46 | eliel | Gosub(), Macro(), Return() |
| 18:15.54 | eliel | is more apropiate than: |
| 18:16.07 | eliel | Gosub: for more info type 'core show application Gosub' |
| 18:16.19 | eliel | Macro: for more info type 'core show application Macro' |
| 18:16.21 | eliel | .. etc |
| 18:17.02 | tzafrir_laptop | any idea what are the plans regarding misdn? |
| 18:17.35 | eliel | tzafrir_laptop: at least cleanup the non-standard code! |
| 18:17.41 | ManxPower | tzafrir_laptop: I suggest always keeping it in a pentagram so it can't escape and destroy us all! >*-) |
| 18:17.57 | tzafrir_laptop | our chan_misdn uses misdn 1.1 . Aparantly there's a shiny new misdn 1.2 |
| 18:19.13 | tzafrir_laptop | but it also uses a completely new userspace interface |
| 18:20.20 | tzafrir_laptop | http://www.linux-call-router.de/ http://www.linux-call-router.de/download/lcr-1.2/ |
| 18:21.18 | tzafrir_laptop | I must admit they did a great job in picking a confusing name |
| 18:21.44 | kpfleming | what do you mean 'what are the plans'? chan_misdn is still in the tree and will still be supported as it has been (which hasn't been much, granted) |
| 18:22.11 | kpfleming | but now we'll be able to support BRI natively in chan_dahdi, which is a big improvement and simplifies configuration, plus it performs better |
| 18:22.32 | ManxPower | kpfleming: I heard that the Digium BRI card is finally getting DAHDI/Zaptel support. Is that true. |
| 18:22.53 | kpfleming | yes, in dahdi-2.1.0-rc1 which is on the download site |
| 18:23.05 | kpfleming | will be released as a real release once it gets through some more PQ testing |
| 18:23.13 | ManxPower | kpfleming: much of the world thanks you |
| 18:23.23 | Qwell | ManxPower: send lots of money. |
| 18:23.46 | ManxPower | Qwell: Oh, I would not touch a BRI for Asterisk in the USA for anything. |
| 18:24.08 | Qwell | Strom said he managed to get it working in the US.. No idea why... |
| 18:24.42 | ManxPower | Qwell: Strom is a TDM geek. There doesn't have to be a reason why other than "wow, that's cool!" |
| 18:24.48 | Qwell | true |
| 18:25.26 | ManxPower | He seems to be the sort of guy to use the 16Kbps D-channel on BRIs for lowspeed data in addition to signaling (as it was intended) |
| 18:26.25 | ManxPower | Qwell: If jsmith has room in the class next week for me, I'll be sure to track you down during a lunch break. |
| 18:26.47 | seanbright | i wonder if there is every going to be a class near me... |
| 18:26.49 | seanbright | looks it up |
| 18:30.07 | seanbright | so i have to go to AL or NV |
| 18:30.12 | seanbright | forget that. |
| 18:31.05 | putnopvut | seanbright: where in NV? |
| 18:31.08 | seanbright | vegas |
| 18:31.25 | seanbright | but if i went to vegas, i would not be able to focus on training |
| 18:31.35 | caio1982 | that reminds me that asterisk bootcamp and dcap tests will be given in Spanish here in Brazil this month |
| 18:31.38 | ManxPower | Not all of AL is unpleasant. The coast is very nice and Huntsville is a high tech city, has a NASA facility, as well as the HQ of companies like Adtran and large facilities from major high tech companies. |
| 18:32.17 | Qwell | looks out his window and sees A) A Saturn V rocket. B) Cotton fields. C) Farmland |
| 18:32.19 | ManxPower | caio1982: I guess they didn't have any in Portuguese. |
| 18:32.39 | anonymouz666 | caio1982: Spanish?!?! |
| 18:32.45 | putnopvut | Qwell: that about sums up Huntsville. |
| 18:32.47 | ManxPower | Qwell: You can see that rocket from 1/2 of huntsville. |
| 18:32.58 | Qwell | ManxPower: I have a particularly nice view from my office. |
| 18:33.12 | Qwell | (of the rocket, that is..) |
| 18:33.35 | caio1982 | ManxPower: i was told it's an argentinian guy dcap-certified, though there are brazilian with a dcap badge capable of doing so... but i'm speculating about their knowledge on it |
| 18:33.44 | anonymouz666 | caio1982: English is easier than Spanish for us |
| 18:33.44 | caio1982 | Qwell: pictures? :) |
| 18:33.51 | Qwell | caio1982: nope.. |
| 18:34.07 | kpfleming | google 'huntsville space rocket center' |
| 18:34.10 | ManxPower | Qwell: My view right now is leaves in full fall colors, high enough up that I can't really even see the ground. |
| 18:34.26 | caio1982 | kpfleming: it's not the same like from his office :) |
| 18:34.31 | kpfleming | true |
| 18:36.57 | Qwell | caio1982: it more or less is - they'd just be closer |
| 18:37.04 | Qwell | (not by much) |
| 18:37.20 | Qwell | it's what, half mile from here? |
| 18:37.32 | Qwell | no, more than that.. |
| 18:38.38 | eliel | caio1982: I took the dcap in argentina, and i can say that ... dcap-- |
| 18:38.57 | eliel | but not because of the dcap itself |
| 18:39.05 | eliel | because the one trying to teach us |
| 18:39.18 | caio1982 | eliel: ouch, so it's probably the same person |
| 18:39.26 | caio1982 | i... suppose...? |
| 18:39.31 | seanbright | so there is no bootcamp in the states? |
| 18:39.42 | eliel | caio1982: do you know his name? |
| 18:39.49 | caio1982 | eliel: not really, no |
| 18:41.16 | eliel | caio1982: his surname was Bianchi, is a nice person |
| 18:41.25 | eliel | s/was/is/ |
| 18:41.53 | caio1982 | Mariano Bianchi? |
| 18:41.57 | eliel | yeaph |
| 18:58.19 | ManxPower | seanbright: last line of make install and make config output: These are the most important Asterisk documentation files you will ever read. Read them. UPGRADE-1.2.txt, Zaptel-to-DAHDI.txt, UPGRADE-1.4.txt, and UPGRADE.txt. |
| 18:58.32 | seanbright | k |
| 18:58.40 | ManxPower | Just a suggestion. 8-) |
| 19:00.16 | seanbright | heh |
| 19:00.19 | seanbright | no prob. |
| 19:03.27 | *** join/#asterisk-dev mvanbaak (n=michiel@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 19:03.27 | *** mode/#asterisk-dev [+o mvanbaak] by ChanServ |
| 19:04.08 | Corydon76-dig | mvanbaak: !!! |
| 19:04.28 | mvanbaak | heeeeeeeeey ! |
| 19:04.41 | Corydon76-dig | mvanbaak: I found a situation in which sometimes svnmerge locks up |
| 19:04.53 | mvanbaak | I haz internatz in my new house ! |
| 19:04.59 | Corydon76-dig | woot |
| 19:05.07 | mvanbaak | Corydon76-dig: yipes, that's bad |
| 19:05.16 | mvanbaak | Corydon76-dig: specific case ? |
| 19:05.39 | mvanbaak | hhmm, hang on. is it when the svnmerge cause conflicts ? |
| 19:05.48 | Corydon76-dig | Occasionally, during one of the steps after you merge a patch from 1.4 to trunk and trunk to 1.6, I get a solid lockup, because it's waiting for user input |
| 19:05.53 | tzafrir_laptop | ManxPower, this line of text adds noise. People learn to ignore noise. |
| 19:06.23 | Corydon76-dig | mvanbaak: right, because 1.6 doesn't have the 1.4-merged property |
| 19:07.04 | mvanbaak | Corydon76-dig: the other day I talked about the svnmerge stuff we have with russell |
| 19:07.16 | Corydon76-dig | When it attempts to merge the property, it expects you to type 'mf' in response, but since the command is not interactive, it seems to hard lock |
| 19:07.41 | mvanbaak | because I dont think we need the -P and -B flags anymore, because the new svnmerge can handle multiple merge sources right now |
| 19:07.47 | mvanbaak | they introduced the -S flag |
| 19:08.22 | mvanbaak | hhmm, I did remove the 'non-interactive' stuff from the svnmerge.py script |
| 19:08.29 | mvanbaak | but, good catch |
| 19:08.33 | mvanbaak | I'll have to look into it |
| 19:10.11 | *** join/#asterisk-dev stoffell (n=stoffell@d51A4D575.access.telenet.be) |
| 19:12.20 | Corydon76-dig | mvanbaak: to fix, I 'ps auxwww' to get the command that is hanging, cancel the svnmerge, run the hanging command, svn revert the properties, and svn pe trunk-merged . |
| 19:12.48 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 19:12.48 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 19:13.19 | mvanbaak | Corydon76-dig: wow, that's too much work |
| 19:16.19 | oej | mvanbaak! |
| 19:16.25 | mvanbaak | oej: ! |
| 19:17.33 | mvanbaak | oej: I haz internatz in my new house ! |
| 19:17.57 | mvanbaak | today at 2PM the signal arrived |
| 19:18.23 | oej | Congratulations and welcome back to the Internet. It was empty without you. |
| 19:18.42 | mvanbaak | :) |
| 19:25.38 | mvanbaak | brb, food |
| 19:32.46 | philipp64 | russelldb: saw that 0013833 was closed... couldn't figure out how to add a note to it (are notes closed to all but bug marshals after a bug is closed)? Anyway, wanted to stay in the loop for the discussion but I'm not on the dev mailing list. |
| 19:34.32 | philipp64 | Qwell: actually, you marked it resolved, right? |
| 19:34.36 | lmadsen | mvanbaak: !!! |
| 19:42.54 | mvanbaak | lmadsen: !!!!!!!!! |
| 19:43.43 | *** join/#asterisk-dev oej (n=olle@ns.webway.se) |
| 19:43.43 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 19:46.25 | mvanbaak | meh, my new oven can cook 4 pizza's at the same time |
| 19:46.37 | mvanbaak | I love huge kitchen equipment |
| 19:49.04 | kpfleming | mvanbaak: welcome back |
| 19:49.06 | kpfleming | where are you living now? |
| 19:49.20 | seanbright | kpfleming: look out your window... |
| 19:49.21 | mvanbaak | kpfleming: thank you |
| 19:49.23 | seanbright | heh |
| 19:49.30 | kpfleming | lol |
| 19:49.34 | mvanbaak | kpfleming: The Hague |
| 19:49.37 | kpfleming | ahh |
| 19:50.09 | mvanbaak | the big city ! |
| 19:50.22 | mvanbaak | Lunteren is a very small town hidden in the forests |
| 19:50.43 | mvanbaak | always had to take a car, no matter what I wanted to do |
| 19:50.43 | kpfleming | 'small' means quite a lot in the Netherlands as well, since the whole country is small |
| 19:50.51 | mvanbaak | lol |
| 19:50.57 | mvanbaak | yeah |
| 19:51.11 | mvanbaak | it was like 2000 citizens |
| 19:51.25 | mvanbaak | during holiday season around 4000 |
| 19:52.03 | mvanbaak | The Hague is: 478.948 |
| 19:53.20 | mvanbaak | Corydon76-dig: you have an easy way to reproduce the svnmerge hang ? |
| 19:53.26 | mvanbaak | so I can test a patch I have here |
| 19:53.59 | mvanbaak | kpfleming: is there a viewvc setup on origsvn ? |
| 19:54.35 | seanbright | there is not, no. |
| 19:55.05 | kpfleming | nope |
| 19:55.08 | seanbright | although i am not kpfleming, i feel confident in my answer. |
| 19:55.13 | kpfleming | it would require certifcates for access |
| 19:55.19 | kpfleming | which we could do... but we haven't |
| 19:56.21 | mvanbaak | k |
| 19:56.26 | mvanbaak | no problem |
| 19:58.48 | snuff-home | kpfleming: manage to get 2.6.26+ booting? |
| 19:59.09 | kpfleming | yes, the firmware upgrade fixed that |
| 19:59.35 | mvanbaak | Corydon76-dig: it must be because svnmerge runs under expect |
| 19:59.35 | kpfleming | and asterisk compiles fine on Linux, but that's because the makefile only includes -Wcast-align for Solaris, not based on CPU type (which it should) |
| 20:00.00 | kpfleming | mvanbaak: it is |
| 20:00.20 | kpfleming | so, Qwell is setting up the T2000 to dual-boot Solaris 10 and Linux |
| 20:00.42 | mvanbaak | it's....sad |
| 20:00.49 | kpfleming | and then we'll have to get something setup to test to see what actually works and what doesn't, because the compiler warnings are just indications of possible problems, many of them might not be actual problems |
| 20:01.04 | mvanbaak | good news is, we dont need the branch specific merge props anymore with the new svnmerge |
| 20:01.28 | mvanbaak | but that means we have to pick a date and time where we drop compatibility with the older svnmerge versions |
| 20:02.21 | mvanbaak | because digium svnmerge learned this neat trick with multiple merge sources on Sept 12 |
| 20:02.34 | mvanbaak | commit repotools 487 |
| 20:02.40 | MuffinMan | [487] Committed by mvanbaak on 2008-09-12T16:58:17.627298Z: Update svnmerge to rev 32857 from upstream svn repo |
| 20:02.40 | MuffinMan | .. |
| 20:02.40 | MuffinMan | ..NOTE: I had to remove the 'svnmerge avail -B' option from upstream. |
| 20:02.40 | MuffinMan | ..Previous versions of our svnmerge had this -B replaced by -b but that |
| 20:02.44 | MuffinMan | ..option is now taken by the bi-directional functionality. |
| 20:02.46 | MuffinMan | ..You can still access this functionality with it's long option version --blocked: |
| 20:02.47 | mvanbaak | that one |
| 20:02.49 | MuffinMan | ..svnmerge avail --blocked |
| 20:02.50 | seanbright | commit commit commit commit |
| 20:02.50 | snuff-home | np kpfleming |
| 20:02.51 | MuffinMan | .. |
| 20:02.54 | MuffinMan | .. http://svn.digium.com/view/repotools?view=rev&revision=487 |
| 20:03.17 | seanbright | commit seanbright is_awesome_yay |
| 20:03.19 | seanbright | heh. |
| 20:08.14 | *** join/#asterisk-dev bird_of_Luck (n=melifaro@secured.by.ipfw.ru) |
| 20:12.07 | mvanbaak | kpfleming: what was it again you wanted to have tested on non-linux machines ? |
| 20:12.13 | mvanbaak | something with menuconfig right ? |
| 20:12.32 | mvanbaak | oh wait, it was with shells other then bash correct ? |
| 20:13.51 | tzafrir_laptop | if it works with posh and dash, chances are it'll work with ksh |
| 20:14.07 | mvanbaak | it works with zsh |
| 20:14.26 | tzafrir_laptop | "ksh" on my system is zsh |
| 20:14.38 | mvanbaak | that's evil |
| 20:14.53 | mvanbaak | I can try with ksh as well |
| 20:15.43 | mvanbaak | real ksh |
| 20:16.45 | lmadsen | I can try with sea shell |
| 20:19.21 | mvanbaak | In file included from res_config_ldap.c:44: |
| 20:19.28 | mvanbaak | <PROTECTED> |
| 20:19.30 | mvanbaak | meh |
| 20:19.32 | mvanbaak | nice |
| 20:26.03 | mvanbaak | basename(3) needs #include <libgen.h> on OpenBSD |
| 20:53.20 | *** join/#asterisk-dev jameswf (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 20:55.08 | *** join/#asterisk-dev JimVanM (n=jimvanm@bas1-toronto63-1096578109.dsl.bell.ca) |
| 20:55.59 | Corydon76-dig | JimVanM: Don't be going berzIRC now |
| 20:58.33 | JimVanM | Corydon76-dig: I can't. I have responsibilities I must not shIRC |
| 21:07.39 | lmadsen | oooooooooooooo |
| 21:07.41 | lmadsen | nerd jokes |
| 21:08.34 | *** join/#asterisk-dev easycrypt (n=savek@lak-67-121.wohnheime.ruhr-uni-bochum.de) |
| 21:09.17 | easycrypt | is there any reason, why chan-sip needs to recieve a frame first before it starts sending frames itself? |
| 21:11.17 | easycrypt | just trying CLI=> originate sip/100 application echo |
| 21:11.23 | easycrypt | (100 is local sip phone) |
| 21:11.32 | easycrypt | won't work this way |
| 21:11.42 | kpfleming | symmetric rtp |
| 21:12.29 | easycrypt | so it's not a bug, it's a feature? ;) |
| 21:12.39 | kpfleming | right |
| 21:14.05 | mvanbaak | kpfleming: the menuselect changes you made do work on zsh and ksh |
| 21:14.57 | kpfleming | yay |
| 21:15.48 | mvanbaak | hurray |
| 21:17.32 | easycrypt | so if i were to write an application that is called via a chan_sip - i should send a garbage frame first in order to be able to "talk" to the channel? |
| 21:18.15 | easycrypt | (assuming, i'm waiting for data from the sip side and don't really have anything to send at that time) |
| 21:25.36 | mvanbaak | hhmm |
| 21:28.25 | mvanbaak | svn merge --ignore-ancestry https://origsvn.digium.com/svn/asterisk/trunk https://origsvn.digium.com/svn/asterisk/team/mvanbaak/res_clialiases_unconditional . |
| 21:28.39 | mvanbaak | that's how I should merge a developer branch back to trunk right ? |
| 21:28.52 | russellb | in theory, yes |
| 21:28.56 | Qwell | and usually revert ., to remove props |
| 21:29.00 | russellb | assuming the branch is in sync |
| 21:29.02 | russellb | and yes, what Qwell said |
| 21:29.17 | mvanbaak | the branch is in sync |
| 21:29.26 | mvanbaak | svnmerge avail shows no avail revs |
| 21:29.31 | mvanbaak | it compiles without trouble |
| 21:29.46 | mvanbaak | but still, the merge is touching files it shouldn't |
| 21:30.03 | Qwell | mvanbaak: it might not be modifying them |
| 21:30.30 | *** join/#asterisk-dev eliel (n=eliel@200.61.172.61) |
| 21:30.30 | *** mode/#asterisk-dev [+o eliel] by ChanServ |
| 21:30.38 | mvanbaak | k, I'll wait till it's done merging |
| 21:33.39 | mvanbaak | lmadsen: vino ? |
| 21:33.47 | lmadsen | mvanbaak: I just got a bottle! |
| 21:33.57 | lmadsen | just doing a bit of work, then will pour a little glass while I make dinner for the g/f |
| 21:34.24 | mvanbaak | aaaaaaah, things got worse since astricon eh ? |
| 21:34.26 | mvanbaak | ;) |
| 21:35.03 | lmadsen | heh |
| 21:39.19 | mvanbaak | man, I'm stupid |
| 21:39.51 | putnopvut | I say that about you a lot. |
| 21:39.54 | putnopvut | j/k |
| 21:41.40 | mvanbaak | lol |
| 21:42.59 | lmadsen | putnopvut: funny, I was just gonna say the same thing (about you) |
| 21:43.04 | lmadsen | j/k <3 |
| 21:43.07 | putnopvut | oh snap! |
| 21:43.21 | mvanbaak | of course it should touch all those files |
| 21:43.35 | mvanbaak | deprecated CLI commands be gone ! |
| 21:52.36 | seanbright | kpfleming: ping |
| 21:52.41 | kpfleming | gnip |
| 21:53.10 | seanbright | this fixes a res_snmp.c build error over here... look safe/right? |
| 21:53.12 | seanbright | http://seanbright.pastebin.com/d3158d4db |
| 21:53.56 | seanbright | and it's only a build error because of --enable-dev-mode i believe. |
| 21:55.55 | kpfleming | seanbright: yes, that looks right |
| 21:56.05 | seanbright | splendid. |
| 21:56.43 | seanbright | another question... is this new API abuse? http://seanbright.pastebin.com/d40676af8 |
| 21:58.39 | russellb | No, I think that is a perfect candidate for the new API call. |
| 21:59.02 | seanbright | the callback just never returns 1 :) |
| 21:59.04 | russellb | The places where it wouldn't be appropriate necessarily, are when the operation you are performing on each channel is non-trivial |
| 21:59.05 | russellb | right |
| 21:59.11 | seanbright | right |
| 21:59.21 | russellb | if what you're going to do might block or take a while, you do _not_ want to do that with the channel list locked |
| 21:59.27 | russellb | if you're incrementing an integer, then fine |
| 22:00.54 | mvanbaak | russellb: http://reviewboard.digium.com/r/26/diff/2/ |
| 22:01.03 | mvanbaak | The file '/Makefile' (r684) could not be found in the repository: PROPFIND request failed on '/svn/asterisk-addons/!svn/bc/684/Makefile' '/svn/asterisk-addons/!svn/bc/684/Makefile' path not found |
| 22:01.36 | russellb | UMMMM |
| 22:01.49 | russellb | mirror out of sync i suppose? |
| 22:03.02 | russellb | no, it's in sync ... |
| 22:05.04 | kpfleming | seanbright: hahaha... no, that's a perfectly valid performance improvement |
| 22:05.11 | seanbright | cool |
| 22:05.43 | russellb | mvanbaak: eliel ... I don't know what the problem is. Maybe something happened when uploading that diff? something not set right? |
| 22:05.56 | eliel | russellb: i fixed it |
| 22:05.58 | russellb | diff rev3 works |
| 22:06.00 | russellb | thanks |
| 22:06.26 | eliel | it was the path, i though it will work with / but /trunk was needed |
| 22:06.31 | russellb | right |
| 22:09.28 | *** join/#asterisk-dev tzafrir_laptop (n=tzafrir@bzq-179-75-202.static.bezeqint.net) |
| 22:09.28 | *** mode/#asterisk-dev [+o tzafrir_laptop] by ChanServ |
| 22:11.41 | *** join/#asterisk-dev mvanbaak_ (n=michiel@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 22:11.42 | *** mode/#asterisk-dev [+o mvanbaak_] by ChanServ |
| 22:11.45 | mvanbaak | hhmm |
| 22:12.49 | mvanbaak | need some help with svn |
| 22:12.58 | mvanbaak | I merged a branch into trunk |
| 22:13.04 | mvanbaak | now, if I run 'svn diff' |
| 22:13.09 | mvanbaak | the two new added files dont show up |
| 22:13.11 | mvanbaak | :( |
| 22:14.02 | *** join/#asterisk-dev jameswf (n=james@dsl093-157-131.phx1.dsl.speakeasy.net) |
| 22:14.46 | mvanbaak | how can I include them in the diff ? |
| 22:14.50 | seanbright | well, i can't get res_snmp to compile... :( so i won't be able to compile and test that patch |
| 22:14.56 | seanbright | if someone else wants to, by all means. |
| 22:15.37 | eliel | mvanbaak: svn copy |
| 22:15.54 | eliel | svn copy http://branch/file localworkingcopy |
| 22:19.31 | mvanbaak | nope |
| 22:19.33 | mvanbaak | same thing |
| 22:20.06 | putnopvut | mvanbaak: are the two new files in your working copy of trunk? |
| 22:20.07 | eliel | hmmm i have tested it with: svn copy http://svn.digium.com/svn/asterisk/team/group/res_clialiases/res/res_clialiases.c res/ |
| 22:21.10 | mvanbaak | putnopvut: they were in team/mvanbaak/res_clialiases_unconditional |
| 22:21.35 | putnopvut | Did you ever svn add those? |
| 22:21.36 | mvanbaak | putnopvut: I then used: svn merge $trunk_url $clialiases_url . |
| 22:21.38 | mvanbaak | in trunk |
| 22:21.44 | mvanbaak | that added the two files |
| 22:21.48 | mvanbaak | anywayz |
| 22:21.53 | mvanbaak | cd trunk |
| 22:22.42 | mvanbaak | svn remove $those_two_files && cp ../team/mvanbaak/res_clialiases_unconditional/res/res_clialiases.c res/res_clialiases.c && svn add res/res_clialiases.c && svn diff |
| 22:22.43 | putnopvut | Hmm, that seems weird. If you added the files in the branch, and then merged it to trunk, then the new files should have automatically been added. |
| 22:22.46 | mvanbaak | that did the trick |
| 22:22.54 | putnopvut | Cool |
| 22:26.06 | mvanbaak | hhmm, bugs.digium.com down ? |
| 22:26.11 | *** join/#asterisk-dev dioedu (n=dioedu@201.7.117.114) |
| 22:26.12 | seanbright | yes |
| 22:26.16 | seanbright | maintenance |
| 22:26.27 | seanbright | there was a post to the -dev list about it |
| 22:26.52 | mvanbaak | ah |
| 22:27.35 | kpfleming | seanbright: you know what happens when you get a new hammer? everything looks like a nail |
| 22:30.59 | mvanbaak | hhmm, we are going to use nested functions all over the place now eh ? |
| 22:32.01 | seanbright | kpfleming: those couple changes were a perfect fit |
| 22:32.23 | seanbright | kpfleming: the other instances where walk_locked are used were not |
| 22:32.39 | russellb | hits seanbright with a hammer |
| 22:32.43 | russellb | i mean, hi! |
| 22:33.01 | seanbright | i can revert |
| 22:33.04 | seanbright | but i don't wanna |
| 22:33.05 | seanbright | :) |
| 22:33.12 | mvanbaak | no, it's ok |
| 22:33.53 | seanbright | mvanbaak: is what problematic on OpenBSD? |
| 22:34.01 | mvanbaak | it's just that asterisk code never had it, till kevin and olle were on sippit and introduced a nested function, and now I see it happening in more places |
| 22:34.19 | mvanbaak | seanbright: it was, but I think I fixed it earlier |
| 22:34.27 | *** join/#asterisk-dev CGMChris (n=chris@mail.cgmyes.com) |
| 22:34.39 | seanbright | how so? |
| 22:34.47 | seanbright | how did you fix it, i mean. |
| 22:34.51 | mvanbaak | seanbright: the openbsd gcc does not generate trampolines by default. you have to add a flag to the compile line to enable generation of trampolines |
| 22:34.58 | seanbright | ahhh, i see. |
| 22:35.41 | mvanbaak | from the gcc-local(1) manpage: |
| 22:35.42 | mvanbaak | <PROTECTED> |
| 22:35.45 | mvanbaak | <PROTECTED> |
| 22:35.49 | mvanbaak | <PROTECTED> |
| 22:35.52 | mvanbaak | <PROTECTED> |
| 22:36.08 | seanbright | is there a performance penalty or something? |
| 22:36.21 | Corydon76-dig | seanbright: slight |
| 22:36.26 | seanbright | ahhh |
| 22:36.33 | mvanbaak | http://www.openbsd.org/cgi-bin/man.cgi?query=gcc-local&apropos=0&sektion=0&manpath=OpenBSD+Current&arch=i386&format=html |
| 22:36.45 | mvanbaak | the openbsd ppl documented all changes to stock gcc |
| 22:36.50 | mvanbaak | that's the manpage about it |
| 22:36.56 | seanbright | well we could change the callback passed to ast_channel_search_locked to take a void * |
| 22:37.04 | seanbright | but i kinda like it the way it is :) |
| 22:37.53 | *** join/#asterisk-dev Pazzo (n=ugelt@sadsl-246059.rol.raiffeisen.net) |
| 22:38.38 | seanbright | i'll leave it up to the powers-that-be |
| 22:38.44 | seanbright | points to people that aren't him |
| 22:38.53 | mvanbaak | I think it's fine |
| 22:39.28 | mvanbaak | still dont see the real benefit over a seperate function, but that's me :) |
| 22:39.48 | seanbright | you get access to all of the local variables in the parent function |
| 22:39.50 | putnopvut | mvanbaak: russellb sent out an e-mail on the dev list about the use of nested functions. If you want to comment on it, you can reply there. |
| 22:40.25 | putnopvut | russellb tried to set out some basic guidelines for when they are useful and when they are not. |
| 22:40.28 | seanbright | oh... maybe i should have read that. |
| 22:41.08 | mvanbaak | seanbright: yeah, I know. but that's also the thing that scares me ;) |
| 22:41.58 | seanbright | putnopvut: when was that posted? |
| 22:42.04 | mvanbaak | as far as I can see it also makes the variable space on the stack executable |
| 22:42.08 | putnopvut | last week. I'll find the date... |
| 22:42.30 | mvanbaak | part of the reason why openbsd disabled it by default |
| 22:43.34 | seanbright | putnopvut: found it |
| 22:43.35 | putnopvut | seanbright: it was posted Oct 31 |
| 22:43.37 | russellb | If it _does_ make stack space executable, we need to understand that |
| 22:43.38 | putnopvut | cool |
| 22:43.40 | seanbright | staring right at me :-/ |
| 22:43.45 | russellb | because that's a pretty good reason _not_ to use it |
| 22:43.51 | russellb | my understanding was that was not the case |
| 22:44.15 | putnopvut | what is meant by "making stack space executable" |
| 22:44.42 | russellb | normally, executable code is in a different section in memory |
| 22:44.54 | russellb | and your stack would be non-executable |
| 22:45.00 | russellb | (helps with security issues.) |
| 22:45.15 | russellb | because then with a stack buffer overflow, you can't insert arbitrary code to be executed ... |
| 22:45.18 | mvanbaak | russellb: |
| 22:45.18 | mvanbaak | <PROTECTED> |
| 22:45.22 | mvanbaak | <PROTECTED> |
| 22:45.24 | mvanbaak | <PROTECTED> |
| 22:45.33 | mvanbaak | russellb: that's a mod in the OpenBSD gcc |
| 22:45.40 | russellb | interesting. |
| 22:45.44 | mvanbaak | russellb: normal gcc does not use the mprotect thing |
| 22:45.57 | russellb | so perhaps normal gcc makes the whole section executable when you use this? |
| 22:45.59 | mvanbaak | russellb: it simply makes the whole stack executable |
| 22:46.11 | mvanbaak | exactly |
| 22:46.23 | russellb | ok, well i'd like to verify that ... should be able to looking at the memory map for a running process |
| 22:46.39 | russellb | and if that is the case, we'll have to revisit this and likely revert all of the uses of it in asterisk |
| 22:46.46 | seanbright | yeah... :( |
| 22:47.02 | russellb | no big deal, adding an extra argument to the functions isn't the end of the world ... |
| 22:47.15 | mvanbaak | that's my opinion as well |
| 22:47.28 | seanbright | well, no, but they're sexy |
| 22:47.31 | russellb | ok, so, who wants to do the investigation on this to see what happens :) |
| 22:47.37 | russellb | you should be able to run asterisk 1.4, and then asterisk trunk |
| 22:47.40 | russellb | and look at the difference |
| 22:47.43 | russellb | 1.4 shouldn't have any usage of it |
| 22:49.02 | seanbright | http://seanbright.pastebin.com/d164b957 |
| 22:49.23 | seanbright | hmmm, that's dumb... hod |
| 22:49.26 | seanbright | hold* |
| 22:50.07 | *** part/#asterisk-dev ManxPower (n=manxpowe@223.sub-70-221-254.myvzw.com) |
| 22:50.45 | seanbright | http://seanbright.pastebin.com/d1159b14d |
| 22:50.51 | seanbright | i'm not sure how to read that, however :) |
| 22:51.24 | mvanbaak | me neither |
| 22:52.03 | mvanbaak | can you do the same thing but with moving the inline function to a real function ? |
| 22:52.23 | seanbright | sure. |
| 22:52.36 | *** join/#asterisk-dev jtodd (n=jtodd@blob.fox-den.com) |
| 22:52.43 | mvanbaak | jtodd: ! |
| 22:53.27 | seanbright | http://seanbright.pastebin.com/d28108083 |
| 22:53.55 | seanbright | looks identical |
| 22:54.24 | seanbright | let me try one more thing |
| 22:54.44 | mvanbaak | eliel: thanks for the review. let me fix those issues |
| 22:55.06 | eliel | mvanbaak: minors |
| 22:55.20 | eliel | just learned what russellb already review about my code :) |
| 22:55.37 | seanbright | http://seanbright.pastebin.com/d28e5a393 |
| 22:56.11 | seanbright | looks the same to me... but i'm not sure if that is meaningful at all :) |
| 22:57.34 | seanbright | russellb: try to make sense out of those pastebins ^^^^ |
| 22:57.36 | seanbright | you're smarter than me. |
| 22:57.40 | seanbright | and i have to get a haircut |
| 22:57.42 | seanbright | heh |
| 22:57.44 | seanbright | wanders off |
| 22:59.07 | eliel | seanbright: you could take a look at the generated assembler to see how gcc is implementing it (???) |
| 23:03.59 | *** join/#asterisk-dev `Sean (i=Shaan@CPE0016d3fe20a0-CM001ade84fd0a.cpe.net.cable.rogers.com) |
| 23:06.06 | snuff-work | in sh how can you make a variable have 0 length.. |
| 23:06.17 | mvanbaak | AST_MAX_ARGS is not going to work |
| 23:06.21 | mvanbaak | it's 64 |
| 23:07.14 | mvanbaak | in main/cli.c I see 'cmd[512]' |
| 23:07.17 | mvanbaak | for example |
| 23:20.00 | mvanbaak | brb |
| 23:21.30 | kpfleming | just catching up on the discussion here.. if nested functions require executable stack, then i agree they'll have to go |
| 23:24.59 | *** join/#asterisk-dev mvanbaak_ (n=michiel@asterisk/contributor-and-bug-marshal/mvanbaak) |
| 23:24.59 | *** mode/#asterisk-dev [+o mvanbaak_] by ChanServ |
| 23:25.24 | snuff-work | mm.. insects seems nice n fast.. |
| 23:26.52 | mvanbaak | yup |
| 23:27.53 | mvanbaak | but it's not on bugs.digium.com yet |
| 23:28.10 | mvanbaak | so traffic to it will be no where as much as when it goes back into production |
| 23:28.23 | Qwell | mvanbaak: try again |
| 23:28.39 | *** join/#asterisk-dev bbryant (n=brett@c-68-59-20-153.hsd1.sc.comcast.net) |
| 23:28.39 | *** mode/#asterisk-dev [+o bbryant] by ChanServ |
| 23:28.58 | Qwell | oh, you weren't trying. nevermind |
| 23:29.12 | mvanbaak | trying what ? |
| 23:29.18 | kpfleming | bugs is back up |
| 23:29.19 | Qwell | nothing.. |
| 23:30.43 | mvanbaak | I noticed :) |
| 23:31.45 | snuff-work | thinks he may have submitted first bug on new host for bugs.. |
| 23:33.40 | mvanbaak | hhmm, muffinman not running on that machine eh ? |
| 23:33.54 | kpfleming | probably not yet |
| 23:33.56 | mvanbaak | erm, the bugs bot I mean |
| 23:34.00 | mvanbaak | I know |
| 23:34.03 | mvanbaak | I saw them leave |
| 23:34.52 | mvanbaak | that reviewboard is a great tool btw |
| 23:36.21 | kpfleming | sure is |
| 23:36.54 | *** join/#asterisk-dev otherwiseguy (n=otherwis@72.128.38.41) |
| 23:38.02 | putnopvut | otherwiseguy linked me to this page, regarding trampolines: |
| 23:38.03 | putnopvut | http://gcc.gnu.org/onlinedocs/gccint/Trampolines.html |
| 23:38.19 | putnopvut | What jumped out at me was the ENABLE_EXECUTE_STACK macro |
| 23:38.38 | putnopvut | It seems like if you don't explicitly define it, then the stack will not be made executable. |
| 23:38.53 | putnopvut | I could be misinterpreting though. |
| 23:40.04 | snuff-work | The operating system may also require the stack to be made executable before calling the trampoline. To implement this requirement, define the following macro. |
| 23:40.53 | snuff-work | suggests that certian OSs might require executable stack for trampoline |
| 23:41.17 | kpfleming | no, that means that if you need to do anything to enable executable stack, that is where you define the code to do it |
| 23:41.29 | kpfleming | this is for people porting gcc to platforms, not for code being compiled by gcc |
| 23:42.05 | russellb | oic. |
| 23:42.05 | mvanbaak | main question remains: do trampolines make the stack executable or not |
| 23:42.10 | mvanbaak | as far as I see it does |
| 23:42.20 | kpfleming | my reading of that is that taking the address of a nested function always generates a trampoline, which by definition must live on the stack, which means the stack must be executable |
| 23:42.36 | russellb | hrm ... |
| 23:42.42 | mvanbaak | yeah, that's what I was thinking |
| 23:42.46 | kpfleming | so nested functions are OK if we don't need to take their addresses, but in every case we've allowed their usage, it was specifically for that purpose |
| 23:42.58 | russellb | if that's the case, we seriously need to consider whether we should keep using this or not |
| 23:43.02 | russellb | ok, so, we should document that |
| 23:43.03 | kpfleming | agreed |
| 23:43.51 | russellb | i should actually add that info to the coding guidelines like I said I was going to |
| 23:43.51 | russellb | heh |
| 23:44.45 | mvanbaak | no wonder I had to add all those switches on OpenBSD ;) |
| 23:44.54 | russellb | kpfleming: wait .... aren't we taking the address of the function every time we pass it as a callback? |
| 23:44.59 | kpfleming | yes |
| 23:45.02 | russellb | so ..... |
| 23:45.31 | mvanbaak | that's not trampolines |
| 23:45.32 | russellb | oh, ok, you said that already |
| 23:45.37 | russellb | i misread your statement earlier ... |
| 23:45.37 | kpfleming | right |
| 23:45.51 | kpfleming | trampolines are used because the nested function needs a pointer to its owner's stack frame |
| 23:45.56 | russellb | So ... does that mean we should nix this usage completely? |
| 23:46.14 | kpfleming | i think so |
| 23:46.17 | russellb | nods |
| 23:46.25 | russellb | I still need to read more about how this gets implemented |
| 23:46.30 | mvanbaak | we should not allow it in my opinion |
| 23:46.42 | russellb | nods |
| 23:46.43 | kpfleming | i will email gcc-help and verify that nested function address taking always requires executable stack, but i suspect it does |
| 23:46.54 | russellb | kpfleming: ok, that would be great |
| 23:47.03 | mvanbaak | I think it does |
| 23:47.11 | russellb | and I'll handle making sure we change what needs to be changed tomorrow |
| 23:47.16 | russellb | (unless someone beats me to it) |
| 23:47.43 | mvanbaak | reading the remark about the gcc-changes in OpenBSD where they implemented some extra checks to limit the area that's marked executable ... |
| 23:48.44 | mvanbaak | russellb: you have 2 minutes to answer some non-asterisk-dev questions ? |
| 23:49.04 | russellb | not right now, sorry |
| 23:49.10 | mvanbaak | no problem |
| 23:49.14 | mvanbaak | that's why I asked :) |
| 23:49.26 | mvanbaak | I'll mail you tomorrow (I hope) |
| 23:54.04 | *** join/#asterisk-dev CunningPike (n=arodgers@S010600095b33697f.vc.shawcable.net) |
