| 11:51.41 | *** join/#asterisk-doc trelane_ (i=trelane@pdpc/supporter/sustaining/trelane) [NETSPLIT VICTIM] |
| 12:36.51 | *** join/#asterisk-doc jsmith (n=jsmith@000-242-432.area3.spcsdns.net) |
| 12:36.51 | *** mode/#asterisk-doc [+o jsmith] by ChanServ |
| 13:22.06 | *** join/#asterisk-doc blitzrage (n=blitzrag@asterisk/documenteur-extraordinaire/blitzrage) |
| 13:22.06 | *** mode/#asterisk-doc [+o blitzrage] by ChanServ |
| 13:28.13 | *** join/#asterisk-doc trelane (n=trelane@unaffiliated/trelane) |
| 13:59.58 | *** join/#asterisk-doc trelane_ (i=trelane@pdpc/supporter/sustaining/trelane) [NETSPLIT VICTIM] |
| 14:02.55 | *** join/#asterisk-doc blitzrage (n=blitzrag@asterisk/documenteur-extraordinaire/blitzrage) [NETSPLIT VICTIM] |
| 14:02.56 | *** join/#asterisk-doc trelane (n=trelane@unaffiliated/trelane) |
| 14:02.56 | *** join/#asterisk-doc jsmith (n=jsmith@000-242-432.area3.spcsdns.net) [NETSPLIT VICTIM] |
| 14:02.56 | *** mode/#asterisk-doc [+oo blitzrage jsmith] by irc.freenode.net |
| 14:06.14 | blitzrage | jsmith: got a question for you... maybe you can point me in the right direction |
| 14:06.34 | jsmith | I can try... |
| 14:25.31 | blitzrage | Lets say I have a single Asterisk box and multiple softswitch servers. It does not answer the call as soon as the INVITE is sent since its a proxy server. The 200 OK is done (afaict) only once the channel is answered (phone is answered). So my problem is trying to determine how to "failover" from asterisk through the various proxy servers. If I send all calls in parallel to the SS, then it seems like I just end up w |
| 14:25.38 | blitzrage | Sorry it took a while... Louie called |
| 14:26.59 | blitzrage | to simplify the question, I guess I'm asking how I could approach the problem of failing over between multiple SS's |
| 14:27.44 | blitzrage | I will admit I have not done *heavy* testing (looking at multiple SIP captures etc...), but those appear to be the issues in preliminary testing |
| 14:33.04 | jsmith | Sorry, I'm missing something. |
| 14:33.28 | jsmith | Why can't you just say something like Dial(SIP/ss1&SIP/ss2&SIP/ss3)? |
| 14:33.38 | jsmith | The first one to respond w/ a 200 OK will get the call. |
| 14:33.58 | jsmith | You could even get really fancy and do this: |
| 14:34.18 | jsmith | exten => 1,1,Dial(SIP/ss1&Local/2@context) |
| 14:34.28 | jsmith | exten => 2,1,Wait(1) |
| 14:34.46 | jsmith | exten => 2,2,Dial(SIP/ss2&Local/3@context) |
| 14:34.53 | jsmith | exten => 3,1,Wait(1) |
| 14:35.11 | jsmith | exten => 3,2,Dial(SIP/ss3) |
| 14:36.03 | jsmith | In other words, give the first switch a 1 second headstart, but if they don't answer, start dialing the number out the second softswitch |
| 14:36.04 | jsmith | and on down the line. |
| 14:41.11 | blitzrage | hrmmm interesting. The issue is that the calls could be sent back into Asterisk, then I have multiple calls hitting Asterisk, and I don't always do an Answer() right away |
| 14:41.27 | blitzrage | Although the delayed dial thing you thought up could be quite useful |
| 14:41.44 | blitzrage | I could just call one SS, then wait a second, 2nd one, and so on down the line... |
| 14:41.58 | blitzrage | Since it takes too long to wait for a ${DIALSTATUS} |
| 14:42.35 | blitzrage | takes over a minute for SIP to fail since it does 6 retransmits, doubling the time (exponential) between each retry |
| 14:42.52 | blitzrage | although it fails a lot quicker if I put the retry to only 1 or 2 in the SIP source |
| 14:59.23 | jsmith | It re-transmits the INVITE 6 times? |
| 14:59.33 | jsmith | The softswitch doesn't send a 100 Trying? |
| 15:05.21 | blitzrage | jsmith: if the end point doesn't respond (down), then it retransmits the INVITE 6 times, doubling the time between retries at each try, before it finally says, "Can't reach the end point" and continues on in the dialplan |
| 15:05.28 | blitzrage | so for over a minute, Dial() will just "hang" while it determines the endpoint is not responding |
| 15:05.38 | jsmith | Shouldn't take a minute. |
| 15:05.39 | blitzrage | or not responding with what it wants |
| 15:05.41 | blitzrage | it does |
| 15:05.48 | blitzrage | I've seen it just sit there |
| 15:05.56 | blitzrage | which is why I couldn't just look through |
| 15:06.06 | blitzrage | because it was taking like 5 mins for it to fail on a call |
| 15:06.19 | blitzrage | you have to remember we're dealing with a not so good SS |
| 15:06.33 | jsmith | But this should solve that, because after a second it'll start going down the second SS |
| 15:06.47 | blitzrage | and now I can't seem to figure out how to get MSN to verify my passport account since I never got an email from them |
| 15:06.49 | jsmith | Or two seconds, or whatever you feel is an acceptable time. |
| 15:07.04 | blitzrage | jsmith: yes, I know, hence why I said that would be a useful thing to try earlier :) |
| 15:07.16 | jsmith | Aha! |
| 15:30.18 | *** join/#asterisk-doc ihassin (n=ihassin@c-69-249-83-168.hsd1.nj.comcast.net) |
| 15:31.09 | *** part/#asterisk-doc ihassin (n=ihassin@c-69-249-83-168.hsd1.nj.comcast.net) |
| 15:33.22 | *** join/#asterisk-doc ToTo (n=ToTo@host138-138-dynamic.2-87-r.retail.telecomitalia.it) |
| 15:34.14 | *** join/#asterisk-doc jsmith (n=jsmith@000-242-432.area3.spcsdns.net) |
| 15:34.14 | *** mode/#asterisk-doc [+o jsmith] by ChanServ |
| 15:42.09 | *** join/#asterisk-doc Alric_ (n=nbowyer@avantacom.com) |
| 15:49.44 | Alric | Some jerk stole my nick! |
| 15:56.10 | blitzrage | I just found a routing issue in Roger's AA |
| 15:56.11 | blitzrage | hahaha |
| 15:56.28 | blitzrage | seems like their queue system is down or something |
| 15:58.56 | Alric | Hrm. |
| 15:59.02 | Alric | Debug option for console... |
| 16:00.03 | blitzrage | lol, I just reported a bug :) |
| 16:02.27 | Alric | Asterisk has some weird default options, imo. |
| 16:06.17 | blitzrage | you have some weird default options, imo |
| 16:06.46 | Alric | That seems like an intelligent, adult response... |
| 16:07.45 | blitzrage | oh come on, I'm just playin' |
| 16:07.49 | Alric | Maybe its just cause odbc was detected. |
| 16:08.21 | Alric | Just seems odd that odbc storage of vm would be a default 'on' option. |
| 16:10.34 | Alric | 1.4 seems to be a steep learning curve. |
| 17:00.34 | Alric | Atacomm wants to charge me $270 to repair a IP300 and have it here next day. |
| 17:05.18 | blitzrage | lol |
| 17:05.28 | blitzrage | why wouldn't you just buy a new IP501 then? :D |
| 17:13.58 | Alric | Exactly. |
| 17:14.05 | Alric | Or two 301s and have them here next day. |
| 17:16.27 | blitzrage | yah |
| 17:18.31 | Alric | Luckily I have one 301 in stock to replace the one that just fried in the field. |
| 17:21.18 | blitzrage | hrmmmm.... I think I should work on the outbound dialing logic in the DP now.... |
| 17:22.07 | blitzrage | gotta alter my dialplan script I think to put all callers into the same context, then do the logic inside there by controlling access based on the info in the DB |
| 17:22.11 | blitzrage | thats probably my best approach |
| 17:22.25 | Alric | Why not different contexts? |
| 17:22.52 | blitzrage | because then I have about a million contexts generated and no centralized code |
| 17:23.17 | blitzrage | I want to write static code that pulls info from the DB instead of generating dialplan logic from a script and having to perform a reload when something changes |
| 17:23.25 | blitzrage | this way when something changes in the DB, its available immediately |
| 17:23.37 | Alric | Realtime? |
| 17:25.49 | jsmith | No, this is FAKETIME! |
| 17:26.18 | jsmith | Alric: Did you see him write "I want to write *static* code that pulls...." |
| 17:26.43 | blitzrage | Alric: RT will be in the future, but the call handling still has to be done somewhere |
| 17:26.55 | Alric | I saw that, but that seemed more like the means to the end, rather than something he wanted to do himself. |
| 17:27.08 | Alric | I was just suggesting another means to that end... |
| 17:27.15 | blitzrage | Alric: you obviously haven't been paying attention to the wicked DP logic I've been writing lately :) |
| 17:27.58 | blitzrage | RT will be implemented when I want to move users registration to the DB as well, but for now I need to build something before Astricon, and this is the best interim method until I redo everything again for 1.4 w/ SER integration |
| 17:28.15 | blitzrage | those registrations are going to be moved to SER at some point, so RT is probably a moot point |
| 17:29.19 | Alric | subnet for /29 is... |
| 17:29.44 | Alric | .248? |
| 17:31.31 | blitzrage | 0/24, 128/25, 192/26, 224/27, 240/28, 248/29 |
| 17:31.33 | blitzrage | yep, you're right |
| 17:31.57 | blitzrage | 256, 128, 64, 32, 16, 8 |
| 17:37.26 | trelane | NOW DO IT IN BINARY AND MATH! |
| 17:37.35 | trelane | MUAHAHAHAHAHA |
| 17:37.39 | blitzrage | I don't do subnetting with binary :) |
| 17:37.43 | trelane | nor do I |
| 17:37.48 | trelane | cisco does, but then they're on crack |
| 17:37.57 | blitzrage | I use a tree structure |
| 17:38.28 | jsmith | <PROTECTED> |
| 17:38.32 | blitzrage | I developed it at school and now they teach it in the program I graduated from |
| 17:38.34 | jsmith | 3 bits gives us 8 addresses |
| 17:38.39 | jsmith | 256 - 8 = 248 |
| 17:38.42 | jsmith | It's not hard |
| 17:38.58 | blitzrage | some people use some crazy bit twiddling strategies |
| 17:39.10 | blitzrage | working with moving the bits around on paper |
| 17:39.19 | jsmith | There's nothing to move! |
| 17:39.26 | jsmith | It's just counting in binary, that's all. |
| 17:40.21 | blitzrage | jsmith: yah, but the method I use it for is when you have a subnet and have to break it up into multiple peices for different LANs |
| 17:40.30 | blitzrage | TMTOWTDIT |
| 17:52.16 | Alric | TEOTWAWKI! :) |
| 18:16.10 | *** join/#asterisk-doc jsmith (n=jsmith@000-242-432.area3.spcsdns.net) |
| 18:16.10 | *** mode/#asterisk-doc [+o jsmith] by ChanServ |
| 18:38.53 | *** join/#asterisk-doc jsmith-cell (n=jsmith@000-242-432.area3.spcsdns.net) |
| 18:38.56 | Alric | ... |
| 18:39.34 | jsmith-cell | Cool... |
| 18:39.41 | Alric | Cell phone? |
| 18:40.03 | jsmith-cell | Yeah |
| 18:40.31 | jsmith-cell | kind of trippy |
| 18:41.05 | Alric | Cool cool |
| 18:42.29 | jsmith-cell | Yeah, now I miss the keyboard on my old phone |
| 18:45.16 | trelane | AAAAAAAAAAAAAAAAAAAAARGH |
| 18:45.43 | jsmith | trelane: Why the scream? |
| 18:45.43 | trelane | the test dialplan is set up on DID XXX-XXXX, someone is going to need to record the following prompts: |
| 18:45.54 | trelane | My first impression is the call never ringed. It simply went directly into a queue and I was presented with music on hold messaging until someone picked up the phone. I think this is wrong and we need a better approach. |
| 18:45.54 | trelane | <PROTECTED> |
| 18:45.58 | trelane | NO SHIT |
| 18:45.59 | trelane | RECORD |
| 18:45.59 | trelane | THE |
| 18:46.01 | trelane | FUCKING |
| 18:46.03 | trelane | PROMPTS |
| 18:46.06 | trelane | DAMMIT! |
| 18:46.35 | trelane | one you record the prompts there will be people saying stuff so that the music on hold makes sense |
| 18:46.43 | blitzrage | ? |
| 18:46.44 | trelane | like thanks for calling please hold a moment while we connect you to an operator |
| 18:46.45 | blitzrage | I'm confused :) |
| 18:46.54 | jsmith | blitzrage: Me too! |
| 18:47.04 | trelane | ok |
| 18:47.14 | Alric | I think he's saying that someone who is tasked with recording prompts is annoyed at the function without the prompts |
| 18:47.21 | trelane | BINGO |
| 18:47.23 | Alric | Nevermind that it isn't permanent, and they're being waited on. |
| 18:47.32 | trelane | nevermind that it isn't production either |
| 18:47.35 | trelane | just record the prompts |
| 18:48.03 | trelane | because until you record the prompts there won't be prompts! |
| 18:48.38 | jsmith | Unless you provide defaults ;-) |
| 18:48.56 | Alric | "THIS IS WHERE YOUR STUPID PROMPT WOULD GO" Then queue. |
| 18:48.57 | Alric | :) |
| 18:48.59 | jsmith | Maybe I can get Allison to record a "This prompt has not yet been recorded" prompt |
| 18:49.12 | jsmith | Which, of course, would be a lie. |
| 18:49.27 | trelane | it's a lie merely because of it's existence |
| 18:49.42 | Alric | Festival :) |
| 18:49.53 | Alric | Or cepstal or whatever is in these days. |
| 18:49.56 | Alric | say()? |
| 18:51.06 | trelane | how's about they just record the fucking prompts like good little sheeple so they can hear the new dialplan! |
| 18:51.28 | Alric | I love the word sheeple. |
| 18:52.08 | jsmith | "Beat the sheeple with a clue-by-four..." |
| 18:52.44 | trelane | no we're up to 4x6 hardwood beams |
| 20:55.11 | *** join/#asterisk-doc ToTo (n=ToTo@host138-138-dynamic.2-87-r.retail.telecomitalia.it) |
| 22:52.07 | Alric | You around jsmith? |
| 22:55.44 | Alric | Eh, catch ya later. |
| 22:56.05 | Alric | Got an opportunity to leave and I'm jumping that! |
| 22:56.30 | *** part/#asterisk-doc Alric (n=nbowyer@avantacom.com) |
| 23:59.34 | *** join/#asterisk-doc blitzrage (n=blitzrag@165.154.103.164) |