[00:13] dsu (4d60575b@gateway/web/freenode/ip. joined #highaltitude.
[00:18] dsu (4d60575b@gateway/web/freenode/ip. left irc: Client Quit
[00:26] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) left irc: Remote host closed the connection
[00:36] Dently (456ec12e@gateway/web/freenode/ip. left irc: Ping timeout: 252 seconds
[00:37] fsphil_ (~fsphil@ joined #highaltitude.
[00:44] SpeedEvil1 (~user@tor/regular/SpeedEvil) joined #highaltitude.
[00:45] SpeedEvil (1000@tor/regular/SpeedEvil) left irc: Ping timeout: 240 seconds
[00:46] A01 (ae67c48e@gateway/web/freenode/ip. joined #highaltitude.
[01:00] <A01> why can't I get back to the CUSF balloon predictor for the US?
[01:00] <A01> http://farm5.static.flickr.com/4063/4624943312_abd50d90a9.jpg it looked like this
[01:02] <Randomskk> A01: new software version
[01:02] <Randomskk> it still works for the states
[01:02] <Randomskk> just key in your lat/lon in the box
[01:03] <A01> oh I see what I was doing wrong
[01:03] timbobel (~timboebl@5ED040EE.cable.ziggo.nl) joined #highaltitude.
[01:03] <A01> but it's doing the same thing, freezing at 10% or 0%
[01:03] <timbobel> i fixed the falcom: it was indeed the antenna
[01:03] <Randomskk> A01: just give it a bit of time
[01:03] <Randomskk> or possibly the data servers are broken again. we don't run them
[01:03] <timbobel> i broke off the sarantel helical antenna; then i semi soldered a proper GPS antenna sma connection on it: then it worked.
[01:04] <A01> I gave it like 8 hours last night to no avail
[01:04] <timbobel> then in fury i bruned the shit out of the fsa03
[01:04] <Randomskk> 8 hours is definitely too much
[01:04] <Randomskk> what UUID is it using? (in the debug window)
[01:04] <Randomskk> (near the top)
[01:04] timbobel (~timboebl@5ED040EE.cable.ziggo.nl) left irc: Client Quit
[01:05] <A01> bae775ae7639fd48180480f4f6647ca25a504375
[01:05] <Randomskk> launching near freeport?
[01:06] <A01> hopefully
[01:07] <Randomskk> looks like the NOAA servers are down again. sigh.
[01:07] <Randomskk> they are usually very reliable but the last two weeks they have been a bit unstable
[01:07] <Randomskk> they are who we get the data from
[01:08] <A01> come on NOAA
[01:09] <A01> how will I know when they're back up?
[01:09] <Randomskk> uh, it'l start working very quickly :P
[01:09] <A01> oh okay trial and error got it
[01:09] <Randomskk> usually it gets beyond 10% in about ten seconds or less, then shows a real ETA and actually goes up
[01:09] <Randomskk> typically takes ~40s to download data, but if the servers are slow up to a couple of minutes
[01:10] <A01> yeah I've gotten it to work before
[01:11] <Randomskk> the old predictor is still around but is equally broken
[01:11] <A01> I was just wondering if this new version didn't support me or something
[01:11] <Randomskk> both rely on the same upstream servers sadly
[01:11] <A01> but that's not the case it seems
[01:11] <Randomskk> nah, it works fine for the US
[01:11] <Randomskk> and globally in fact
[01:36] Nick change: SpeedEvil1 -> SpeedEvil
[01:44] SpeedEvil (~user@tor/regular/SpeedEvil) left irc: Quit: Leaving.
[01:53] fsphil_ (~fsphil@ left irc: Ping timeout: 264 seconds
[01:56] fsphil_ (~fsphil@ joined #highaltitude.
[02:02] fsphil_ (~fsphil@ left irc: Quit: Leaving
[02:03] SpeedEvil (1000@tor/regular/SpeedEvil) joined #highaltitude.
[02:28] juxta_ (~blah@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[02:51] juxta_ (~blah@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 276 seconds
[03:18] juxta (juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[04:38] A01 (ae67c48e@gateway/web/freenode/ip. left irc: Ping timeout: 252 seconds
[05:29] jedahan (~jedahan@pub253-166.resnet.stonybrook.edu) joined #highaltitude.
[05:52] natrium (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Ping timeout: 248 seconds
[06:10] jedahan (~jedahan@pub253-166.resnet.stonybrook.edu) left irc: Quit: Leaving
[07:01] spenguin[work] (~penguin@ left irc: Read error: Connection reset by peer
[07:01] spenguin[work] (~penguin@ joined #highaltitude.
[07:53] Upu (~Upu@ubn.upuaut.net) joined #highaltitude.
[07:57] Jasperw (~jasperw@212-98-44-230.static.adslpremium.ch) joined #highaltitude.
[08:47] <Upu> morning
[08:47] <Upu> http://www.youtube.com/watch?v=5V5FteVmdvc
[08:55] m1x10 (m1x10@ppp089210151148.dsl.hol.gr) joined #highaltitude.
[08:56] <m1x10> morning all
[08:57] <Upu> morning
[08:57] <Upu> http://www.youtube.com/watch?v=5V5FteVmdvc
[09:00] <m1x10> heh
[09:00] <m1x10> nice
[09:19] fsphil_ (~fsphil@ joined #highaltitude.
[09:34] rjg_ (4d60575b@gateway/web/freenode/ip. joined #highaltitude.
[09:36] fsphil_ (~fsphil@ left irc: Ping timeout: 246 seconds
[10:01] fsphil_ (~fsphil@ joined #highaltitude.
[10:16] fsphil_ (~fsphil@ left irc: Quit: Leaving
[10:31] <m1x10> cool, I used the coax instead of the wire and I can be heard by someone else :)
[11:16] GeekShadow (~Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[11:38] Jasperw (~jasperw@212-98-44-230.static.adslpremium.ch) left irc: Ping timeout: 240 seconds
[12:45] ejcweb (~chatzilla@ joined #highaltitude.
[12:46] natrium42 (~natrium42@ joined #highaltitude.
[12:49] TraumaPony (~TraumaPon@203-214-123-83.dyn.iinet.net.au) left irc: Ping timeout: 255 seconds
[12:50] TraumaPony (~TraumaPon@ joined #highaltitude.
[13:13] SpeedEvil (1000@tor/regular/SpeedEvil) left irc: Ping timeout: 276 seconds
[13:15] Jasperw (~jasperw@212-98-44-230.static.adslpremium.ch) joined #highaltitude.
[13:29] SpeedEvil (~user@tor/regular/SpeedEvil) joined #highaltitude.
[13:34] SpeedEvil (~user@tor/regular/SpeedEvil) left irc: Ping timeout: 264 seconds
[13:47] SpeedEvil (1000@tor/regular/SpeedEvil) joined #highaltitude.
[14:00] rjg_ (4d60575b@gateway/web/freenode/ip. left irc: Quit: Page closed
[14:06] natrium42 (~natrium42@ left irc: Quit: natrium42
[14:15] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) joined #highaltitude.
[14:22] juxta|console (~juxta@ppp203-122-193-94.static.internode.on.net) joined #highaltitude.
[15:14] Jasperw (~jasperw@212-98-44-230.static.adslpremium.ch) left irc: Ping timeout: 265 seconds
[15:14] GeekShadow (~Antoine@reactos/tester/GeekShadow) left irc: Read error: Connection reset by peer
[15:24] <m1x10> silent assembly today !
[15:30] juxta|console (~juxta@ppp203-122-193-94.static.internode.on.net) left irc: Quit: Leaving.
[15:31] <Laurenceb> and a weeks detention for the whole school
[15:33] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) left irc: Remote host closed the connection
[15:44] natrium (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[15:46] Jasperw (~jasperw@212-98-44-230.static.adslpremium.ch) joined #highaltitude.
[15:51] TraumaPony (~TraumaPon@ left irc: Ping timeout: 252 seconds
[16:01] <m1x10> TraumaPony, what a lol nick
[16:13] gonzo_ (~gonzo_@cpc1-pool8-0-0-cust154.pool.cable.virginmedia.com) joined #highaltitude.
[16:15] rjharrison (~rharrison@ joined #highaltitude.
[16:15] <rjharrison> Afternoon all
[16:15] <m1x10> hello you
[16:15] <rjharrison> ping DanielRichman, Randomskk ...
[16:15] <rjharrison> Hi m1x10
[16:16] <DanielRichman> hi
[16:16] <Randomskk> hi
[16:16] <rjharrison> Hi DanielRichman, I'm going to make a start and probably a finish on getting the listener server onto habhub
[16:16] <DanielRichman> nice
[16:17] <rjharrison> I'm hoping to integrate History with the tracker and uload the past missions
[16:18] <rjharrison> One good project for some one is to create an xml generator for the XML definition files. I'm not sure how these will look in the furture but it may be sensible to keep them defined in XML
[16:18] <DanielRichman> the tracker is at
[16:18] <DanielRichman> /var/www/default/tracker
[16:18] <rjharrison> ok the listener will be at var/www/default/listener
[16:18] <DanielRichman> the files are owned by natrium:
[16:18] <DanielRichman> but you can change that if you need to.
[16:19] <rjharrison> I think I will move the ownership to a webgroup
[16:19] <DanielRichman> yes that's what we've done witht he other stuff
[16:19] <Randomskk> do we have a group for that? I think we planned on making one if nothing else
[16:19] <DanielRichman> what other stuff there is
[16:19] <rjharrison> hehe
[16:19] <DanielRichman> I suggest root:admin for now
[16:19] <DanielRichman> daniel@nessie:www$ ls -l
[16:19] <DanielRichman> total 8
[16:19] <DanielRichman> drwxrwxr-x 3 root admin 4096 2010-08-12 23:39 default
[16:19] <DanielRichman> drwxrwxr-x 3 root admin 4096 2010-08-06 10:03 status
[16:19] <DanielRichman> like so.
[16:20] <Randomskk> we should really have an architecture chat for how we plan to move forwards with the tracker, listener, predictor etc
[16:20] <DanielRichman> it's dangerous to chgrp www-data because then the web server and php will have write access to the files, which is dangerous given that the tracker & listener code isn't very secure
[16:20] <rjharrison> My aim is to get the exiting stuff integrated and working on hab hub and then let otheres take it to the next level
[16:20] <Randomskk> the xml generator is on my task list I think, I got quite sidetracker, but if anyone else wants to make it feel free
[16:20] <Randomskk> do we definitely want to stay with xml? if it's already being parsed okay in dl-fldigi there's probably no significant reason to change it
[16:21] <Randomskk> a generator would definitely be a good idea
[16:21] <DanielRichman> rjharrison: iirc php execution is only enabled in /tracker atm but I can also enable it in /listener
[16:21] <rjharrison> I think it's imperative that we keep a working system up 99.9% of the time.
[16:21] <Randomskk> definitely
[16:21] <rjharrison> DanielRichman cool I assume I can cahnge that in /etc/php.ini
[16:21] <Randomskk> nah, that's a cherokee setting
[16:21] <DanielRichman> no, that's cherokee config. I've got port forwarding already setup so I can do that now
[16:22] <rjharrison> Ok cool if you can do that it would be great
[16:22] <Randomskk> does the listener properly escape mysql queries?
[16:22] <rjharrison> I think so most of my code does
[16:22] <Randomskk> excellent
[16:23] <DanielRichman> daniel@nessie:default$ ls -l
[16:23] <DanielRichman> total 12
[16:23] <DanielRichman> -rw-rw-r-- 1 root admin 1881 2010-08-06 13:07 index.html
[16:23] <DanielRichman> drwxrwxr-x 2 rharrison admin 4096 2010-08-15 15:21 listener
[16:23] <DanielRichman> drwxrwxr-x 10 natrium admin 4096 2010-08-08 22:14 tracker
[16:23] <Randomskk> (tracker... not so much)
[16:23] <rjharrison> I'll make sure it's in there
[16:23] <Randomskk> the tracker is very temporal atm
[16:23] <Randomskk> so it's more important from the listener
[16:23] <rjharrison> ok cool is there a webuser as such
[16:23] <Randomskk> we should/will make one
[16:23] <DanielRichman> php runs under www-data
[16:23] <DanielRichman> we are all members of the group admin
[16:23] <Randomskk> but atm you could probably just keep it owned by you and world readable
[16:23] <rjharrison> chown [webuser] /var/www/default/*
[16:24] <DanielRichman> rjharrison: having it owned by www-data is dangerous since then php could write to the files
[16:24] <Randomskk> having the web files owned by root makes me a little edgy but is probably harmless
[16:24] <Randomskk> having them owned by www-data is definitely a potential issue
[16:24] <Randomskk> at the moment, anyway
[16:24] <DanielRichman> they're owned natrium:admin or rharrison:admin
[16:24] <rjharrison> Ok coollets leave it like that
[16:24] <Randomskk> yea, the main thing is to have the group be admin I guess
[16:25] <DanielRichman> if you need to write to any flat files then you can chgrp www-data for that directory only and we can disable php execution in there
[16:25] <rjharrison> It will only take a sec to change them when we want to
[16:25] <Randomskk> yea
[16:25] <rjharrison> Ok at the moment I dump all data to the tracker to a file in ./data/data.dat
[16:25] <rjharrison> this is what view.php reads
[16:26] <DanielRichman> I wish `cherokee-admin` provided the login details in http://user:pass@ form so I could jsut click and go
[16:26] <Randomskk> DanielRichman: the -u switch
[16:26] <DanielRichman> hmmmm
[16:26] <Randomskk> given as it listens to localhost and is only on when you use it, nothing else should be able to access it
[16:27] <DanielRichman> true, but it runs as root and any user on the box can connect
[16:27] <Randomskk> true
[16:27] <DanielRichman> OK, I've got a rule for php cgi: "(Extensions php AND Directory /tracker)"
[16:27] <DanielRichman> I wonder if it's flexible enough to do php && (tracker || listener)
[16:27] <Randomskk> if not, regex
[16:28] <DanielRichman> ah, it is default ’ (Extensions php AND (Directory /tracker OR Directory /listener))
[16:28] <DanielRichman> very cool
[16:28] <Randomskk> excellent
[16:28] <rjharrison> I'm going to want to modify the client at some point to send data to habhub.org and I want the XML files to live in a subdirectory ./listener/xml
[16:29] <rjharrison> thought there may be an argument to put them in ./webroot/xml as other programs can make use of the mission definition xml files
[16:30] <rjharrison> The listener for example could be configured to lable custom data fields and format the data appropriately according to the xml definition of the mission
[16:30] <rjharrison> thought = though
[16:31] <DanielRichman> http://habhub.org/{listener/,tracker/,}test.php
[16:32] <DanielRichman> perhaps cherokee should be setup to 403 php files in directories where execution isn't allowed
[16:32] <DanielRichman> to prevent accidental exposure of database_passwords.php
[16:32] <rjharrison> hehe like the test.php code
[16:32] <Randomskk> probably xml isn't the best folder name
[16:33] <Randomskk> payloads or payload_definitions or payload_xml or such?
[16:33] <rjharrison> no sure but the idea that it's not linked to the listener only
[16:33] <Randomskk> yes
[16:33] <Randomskk> in the root makes more sense
[16:33] <Randomskk> the tracker should use it to decode and display sensor data, for instance
[16:33] <rjharrison> See above :)
[16:33] <rjharrison> Ahh I menat tracker
[16:34] <rjharrison> The tracker for example could be configured to lable custom data fields and format the data appropriately according to the xml definition of the mission
[16:34] <rjharrison> label :)
[16:35] <rjharrison> I have been at Farnell this w/e for their family funday and I was the guest of honnor to launch the Blimp for the re-launch of the trade counter
[16:35] <DanielRichman> :o
[16:36] <rjharrison> I'm off to Germany with farnell in November and they have agreed to fund HAB into schools
[16:36] <DanielRichman> very nice
[16:36] <rjharrison> Well up to about 10K per year which is more than is needed
[16:36] <Randomskk> :O
[16:36] <rjharrison> I have been working onthe UK Space Agency and they have finaly got behind the project too which is good
[16:37] <Randomskk> excellent
[16:37] <DanielRichman> btw rjharrison ages ago you were talking about 1-pin RTTY
[16:37] <rjharrison> Yep
[16:37] <DanielRichman> how does 1 pin 1 resistor rtty sound?
[16:38] <rjharrison> Hehe I guess it can be done. I thought that you would need the base R for 0 and another R for 1
[16:39] <DanielRichman> the ntx2 has an internal resistance so you can make a potential dividor with one resistor and the txd pin
[16:39] <rjharrison> Thinking a bit more I guess pin in and off via an R would do it
[16:39] <DanielRichman> you need something like a 3M resistor
[16:40] <DanielRichman> or a 1M6. We're not sure, I tried it with the NiM2 which might be different and I needed a 3M
[16:40] <DanielRichman> then it's simply pinmode OUTPUT and LOW and HIGH
[16:40] <rjharrison> -----R----NTX2
[16:41] Last message repeated 1 time(s).
[16:41] <rjharrison> -----|
[16:41] <rjharrison> Humm this is not going to work
[16:41] <Matt_soton> apex used one pin rtty, but used three resistors
[16:41] <rjharrison> but R to gnd as above
[16:41] <DanielRichman> Output Pin>-------[ 3M ] -------- TXD pin
[16:41] <DanielRichman> when output is 0v then txd is 0v
[16:41] <rjharrison> Cool
[16:41] <DanielRichman> when output is 5v, then the internal resistance of 100k makes a dividor 3M and 1k
[16:42] <DanielRichman> and that is a tiny voltage somewhere >0.3 that gives you the 425 shift
[16:42] <rjharrison> Cool
[16:42] <rjharrison> I like it
[16:42] juxta (juxta@ppp203-122-193-94.static.internode.on.net) left irc: Ping timeout: 265 seconds
[16:42] <DanielRichman> it requires science to perfect it but we're now 99% sure it works
[16:42] <rjharrison> Trial and Error :)
[16:43] <griffonbot> @dbsnyder: SurfUp 6 flew successfully, reaching 95,000 ft. Easy recovery in bean field near road. Beer afterward #ARHAB [http://twitter.com/dbsnyder/status/21239797264]
[16:43] <rjharrison> Right let me get on for a bit I'll shout if I get stuck with permission errors
[16:48] <m1x10> DanielRichman, I replaced the wire with coax and now I can be heard by other 3 stations near my city !!
[16:48] <Randomskk> :D
[16:48] <m1x10> I think coax helped :)
[16:48] <DanielRichman> with the antenna at the right height now?
[16:48] <m1x10> 50cm
[16:48] <DanielRichman> good
[16:49] <m1x10> no RF signal through breadboard anymore :):)
[16:50] <Randomskk> hehe much better
[16:51] <m1x10> all packets show qAR which mean they heard me directly on radio
[17:22] <rjharrison> DanielRichman is there a password for the mysql db?
[17:22] <Randomskk> it's in /root
[17:22] <rjharrison> Thanks
[17:25] <DanielRichman> yeah the mysql root pw is in there, natrium created an account rjharrison
[17:25] <DanielRichman> I forget its details but I believe the p/w is still in the logs; /root/.mysql_history
[17:25] <rjharrison> Ok shall I create a db called listener and show my tables in there or preferably create the required tables in the tracker database
[17:25] Darkside (~Darkside@hpavc/darkside) left irc: Ping timeout: 240 seconds
[17:26] <rjharrison> show = shove
[17:26] <DanielRichman> it might be easier to have it all as one user one database
[17:26] <rjharrison> Yep that is my thought
[17:26] natrium42 (~natrium42@ joined #highaltitude.
[17:27] <DanielRichman> rjharrison: ^^
[17:27] <rjharrison> Now what currently happens is that all succesfull data is written to table that holds all the fields for the sentence sent for each user
[17:27] Darkside (~Darkside@hpavc/darkside) joined #highaltitude.
[17:27] <rjharrison> hi natrium
[17:32] <rjharrison> The tracker table is also updated with a successfull string. Now given that many people can send a single datapoint unique indexes in the tracker tables reject multiple entries
[17:33] <rjharrison> natrium I'm in the process of integrating the tracker with the listener server
[17:33] <rjharrison> natrium42 this will take a few hours
[17:34] <rjharrison> I'm going to make a couple of changes to the code to allow for historical missions to be displayed
[17:35] <natrium42> ah, cool
[17:35] <rjharrison> I havn't been inside the new tracker code yeat but I'm assuming it's not a million miles away from the old tracker
[17:35] <Randomskk> rjharrison: you can probably remove the executable permission from all the files, by the way - stops ls bolding and greening them if nothing else, but also prevents mishaps if someone were to put something actually executable in there
[17:36] <rjharrison> Yep will do
[17:36] <natrium42> yep, it didn't change much
[17:50] <rjharrison> T time
[17:53] RocketBoy (~steverand@ joined #highaltitude.
[18:02] <natrium42> Mr. T Tea
[18:14] natrium42 (~natrium42@ left irc: Quit: natrium42
[18:49] <rjharrison> Hey RocketBoy
[18:57] RocketBoy (~steverand@ left irc: Quit: RocketBoy
[19:04] <Upu> evening
[19:09] Simon-MPFH (~simon@phantom.mpfh.co.uk) joined #highaltitude.
[19:14] RocketBoy (~steverand@ joined #highaltitude.
[19:32] Dently (456ec12e@gateway/web/freenode/ip. joined #highaltitude.
[19:36] Hiena (~Hiena@ joined #highaltitude.
[19:36] <rjharrison> Hi Upu
[19:36] <Upu> Evening Rob
[19:37] <Upu> quick question whilst you're here Icarus 1 what was the shift on your NTX2 ? 425 ?
[19:37] <rjharrison> DanielRichman is there a way I can access mysql using mysql workbench from my machine
[19:37] <rjharrison> or Randomskk
[19:38] <rjharrison> Putting it another way can I have access from my static ip to the mysql port
[19:38] <rjharrison> Assuming I dont have this atm
[19:38] <Randomskk> yes, ssh forwarding
[19:38] <Randomskk> but let me check
[19:38] <Randomskk> I don't think mysql is listening
[19:38] <Randomskk> yes, only from localhost
[19:38] <Randomskk> oh, well, that'l still work.
[19:39] <Randomskk> just have your SSH client forward port 3306
[19:39] <rjharrison> Ok I'm on windows box wanting to run the workbench and connect to the mysql instance on the habhub server
[19:39] <Randomskk> from linux, `ssh -L3306:localhost:3306`
[19:39] <Randomskk> well whatever ssh client you use on windows then
[19:39] <Randomskk> putty can do it
[19:39] <Upu> it can
[19:39] <rjharrison> Cool
[19:40] <rjharrison> Learn something new every day :)
[19:40] <Randomskk> once you have port 3306 forwarded, tell mysql workbench to connect to port localhost:3306
[19:41] <rjharrison> Yep
[19:41] <rjharrison> Supa cool
[19:41] <Upu> tunnel D3306 in Putty
[19:41] <Randomskk> not D
[19:42] <Randomskk> D would make a socks proxy
[19:42] <Randomskk> L
[19:42] <Upu> oops yes
[19:43] <Upu> afk a few
[19:43] m1x10 (m1x10@ppp089210151148.dsl.hol.gr) left irc: Quit: I eat electrons for breakfast and I vomit thunders
[19:51] <DanielRichman> if you already have a ssh connection open you can type ~C on a new line and then enter -L3306:localhost:3306. I don't know if putty will respond to that
[19:51] <DanielRichman> * -L 3306:localhost:3306
[19:53] <Randomskk> oh, cool, I did not know that
[19:58] <DanielRichman> There's also ~. and ~(Ctrl-Z)
[19:59] <DanielRichman> for kill and suspend
[19:59] <DanielRichman> might be another few; it's in man ssh under "-e escape char"
[19:59] <Randomskk> oh excellent, ~. will be handy
[19:59] Simon-MPFH (~simon@phantom.mpfh.co.uk) left irc: Quit: Leaving
[20:00] <DanielRichman> hmm there's loads; try ~? for a list
[20:21] Hiena (~Hiena@ left irc: Ping timeout: 252 seconds
[20:23] GeekShadow (~Antoine@reactos/tester/GeekShadow) joined #highaltitude.
[20:23] <rjharrison> DanielRichman, Randomskk ok have to stop for a bit
[20:24] <rjharrison> Wife etc ...
[20:24] <rjharrison> bbiab
[20:24] rjharrison (~rharrison@ left irc:
[20:24] <DanielRichman> Interesting. Increasing the NOMADS DAP request from a 3x3 (HD 0.5) latxlon square to a 20x20 one doesn't dramatically increase the request time (~10%) - however, the difference in asking for just time[0] and time[0..1] doubles the time it takes for the server to respond
[20:27] <DanielRichman> to that extent, asking for worldwide data, all altitude levels, for just one particular time rather than for the whole week, results in a quick response
[20:28] <DanielRichman> it waits for a few seconds, then gives you ~50MB of floats at 500K/s
[20:28] <Randomskk> curious
[20:29] <DanielRichman> maybe last time I was testing this the servers were having a slow day
[20:32] <DanielRichman> there seems to be some throttling of the download, but I can do a few in paralell and it doesn't seem to care
[20:32] <Randomskk> time deltas are eighths of days?
[20:32] <Randomskk> uh, or three hours, then?
[20:33] <DanielRichman> yeah
[20:33] <Randomskk> which is probably enough for one flight?
[20:34] <DanielRichman> well the software wants time[0] and time[1] so that it can interpolate between
[20:34] <Randomskk> ugh
[20:34] <DanielRichman> and the chances are that a flight will cross over and we'll require time[0] 1 and 2
[20:35] <Randomskk> interpolating is definitely a good idea?
[20:35] <DanielRichman> idk, ask whoever wrote it
[20:38] <Upu> BINGO ! 425Hz shift
[20:38] <Upu> 0nly taken me a week
[20:38] <Randomskk> :D
[20:39] <Upu> ok technically if it didn't fall apart when I move it this could actually fly
[20:39] <DanielRichman> ok. For every time[] that you request, regardless (almost) of the alt lat and lon size, it will add 4s to the time that it spends thinking and not downloading
[20:39] <DanielRichman> if that reaches 100s then it will timeout
[20:39] <Randomskk> I see
[20:40] <Randomskk> is it faster to request time[0] and then time[1]
[20:40] <DanielRichman> so the largest amount of data we can ask for in one go is, as far as I can tell, [0:1:25][0:1:46][0:1:360][0:1:719]
[20:40] <Randomskk> rather than time[0:1]
[20:40] <DanielRichman> if we request in paralell, then yes
[20:40] <Randomskk> curious
[20:40] <Randomskk> what about in series?
[20:40] <DanielRichman> then no
[20:40] <DanielRichman> requesting time[0] will (roughly) be 4s + 40mb download at 500K/s
[20:41] <DanielRichman> time[1] will be the same
[20:41] <DanielRichman> if you do it in paralell you get two streams at 400-500K/s
[20:42] <DanielRichman> ie 4s + 80mb at 1000K/s
[20:42] <DanielRichman> if we're only downloading a 3x3 latxlon square then the download is practically instant
[20:42] <DanielRichman> note that this is only for one variable out of the three that we need
[20:43] <DanielRichman> time[0:1] would be 8s + 80mb at 500K/s
[20:46] <DanielRichman> if the nomads servers will let us, we could download the entire dataset in a little over a minute, doing 64 paralell requests.
[20:46] <DanielRichman> infact, I'm going to try that now >.>
[20:46] <Randomskk> hehe
[20:47] <DanielRichman> ps: I'm playing with the HD data here. A full dataset with all 3 variables that we need is 18GB; aka not really worth it
[20:49] <DanielRichman> as I see it the best solution would be a predictor that can make gfs requests itself, with some default values chosen for it so that it doesn't waste too much of its time on network latency (ie. too many requests) or downloading data it doesn't need, and a central cache with decent flocking to avoid programs crashing into each other. Then a cronjob to download data for popular areas in advance
[20:50] <Randomskk> yup
[20:52] <DanielRichman> hmm. I appear to have created... a few processes
[20:53] <Randomskk> 65 or so
[20:53] <DanielRichman> yeah. all the wget processes are competing for curses on my terminal
[20:53] <DanielRichman> it's goign crazy
[20:53] <Randomskk> haha
[20:54] <DanielRichman> and it's producing quite a lot of ssh traffic in the process
[20:54] <DanielRichman> $ du -sh .
[20:54] <DanielRichman> 1.2G .
[20:54] <DanielRichman> quarter of the way there
[21:00] <DanielRichman> job done
[21:01] <DanielRichman> took about 5 minutes and I now have worldwide v component wind data for the next week
[21:02] <Randomskk> nice
[21:02] <Randomskk> would just have to run that every six hours... :P
[21:02] <DanielRichman> :D
[21:02] <DanielRichman> not the most... efficient solution
[21:03] <Randomskk> to be honest though, that could work. do that, output it to the predictor's csv format, everything else becomes super duper simple
[21:03] <Randomskk> horribly inefficient
[21:03] <Randomskk> breaks if the noaa servers are late or down
[21:03] <DanielRichman> they might get a bit annoyed
[21:03] <Randomskk> uses stupid amounts of bandwidth and disk i/o needlessly for almost all the data
[21:03] <Randomskk> I doubt they'd notice to be honest, but maybe. they should be annoyed that it's so wasteful
[21:04] <DanielRichman> heh we shouldn't have setup the RDNS
[21:05] <DanielRichman> or set it up as haha.your.bandwidth.not.mine.org
[21:05] <Randomskk> haha
[21:05] <Randomskk> except that like, it is our bandwidth too
[21:06] <DanielRichman> yeah but it's our download; and as a webserver, we don't really need that much
[21:06] <DanielRichman> they should care about their upload
[21:07] <Randomskk> usually there isn't really any difference
[21:07] <Randomskk> as far as the server operators are concerned
[21:07] <Randomskk> bandwidth is either unmetered, metered by total pipe speed, or metered by total traffic transferred
[21:08] <DanielRichman> hmm, and I 'spose we're not exactly going to fill their pipe
[21:08] <DanielRichman> to download that I was using... ~15mb/s
[21:08] <Randomskk> 15MB/s or 15Mb/s?
[21:09] <DanielRichman> 'dunno, what does wget report it in?
[21:09] <Randomskk> what did it say exactly?
[21:10] <DanielRichman> 1 sec
[21:10] <DanielRichman> it said "220K/s" and there were 64 of them
[21:11] <Randomskk> blimey
[21:11] <Randomskk> are you sure all 64 were getting 220K/s
[21:11] <DanielRichman> I can't be 100% sure.
[21:11] <Randomskk> that would imply 13MB/s exactly
[21:12] <DanielRichman> which is a bit more than a 100 megabit ethernet
[21:12] <Randomskk> yes
[21:12] <Randomskk> there are programs to monitor total data flow, like iftop maybe
[21:12] <DanielRichman> well I'm trying it again on another dataset, and I've used `time` properly (I hope) this time
[21:12] <DanielRichman> so I can compare du -sh with that
[21:13] <Randomskk> also try iftop while it runs
[21:13] <DanielRichman> hmm some of them appear to have failed this time
[21:13] <DanielRichman> maybe I triggered the anti-DoS
[21:13] <DanielRichman> real 10m40.384s
[21:13] <DanielRichman> 2.5G mess/
[21:14] <DanielRichman> that's more like 4MB/s
[21:16] <DanielRichman> If we've got a 100mbit ethernet connection then the maximum it makes sense to run in paralell is 20 before we fill the pipe
[21:16] <DanielRichman> ish.
[21:16] <DanielRichman> but that might make a few other services laggy
[21:17] <Randomskk> heh
[21:17] <DanielRichman> damn, I want an internet connection like this
[21:17] <Randomskk> go to uni
[21:17] Jasperw (~jasperw@212-98-44-230.static.adslpremium.ch) left irc: Ping timeout: 265 seconds
[21:17] <DanielRichman> yeah but if I pulled the stunt I just did at uni I'd get hit by the supercap
[21:17] <Randomskk> it's pretty good times, the 100mbps internet connection especially
[21:18] <Randomskk> depends where you go to uni
[21:18] <DanielRichman> true. couple of years to go anyway
[21:18] <Randomskk> yea :P
[21:18] <Randomskk> sadly schools don't tend to have that awesome internet and even if they did it doesn't tend to be quite as available
[21:18] <Randomskk> also you don't live at school
[21:18] <DanielRichman> so basically the answer to "how do I choose a college" is "pick the one with no caps and the gbit infrastructure
[21:18] <Randomskk> well I mean there are potentially other considerations
[21:19] <Randomskk> but, yes
[21:19] <DanielRichman> :P
[21:19] <Randomskk> door locks matter for assassins
[21:19] Action: DanielRichman checks if ed is logged in
[21:19] <DanielRichman> do they actually pick locks?
[21:19] <Randomskk> yea I just checked that too
[21:19] <Randomskk> don't highlight him
[21:19] <DanielRichman> d/w
[21:19] <Randomskk> :P
[21:22] <Randomskk> and ofc you want to avoid oxford, "no p2p at all. this includes spotify"
[21:22] <DanielRichman> wat
[21:23] <Randomskk> http://www.telegraph.co.uk/technology/news/7019157/Oxford-University-bans-students-from-Spotify.html
[21:23] <DanielRichman> nice
[21:23] <DanielRichman> despotify iirc has the p2p component disabled
[21:24] <Randomskk> colleges to avoid wrt shitty network settings include jesus, homerton, new hall, st catherine's, hughes hall
[21:24] natrium (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc: Ping timeout: 252 seconds
[21:25] <Randomskk> http://wiki.camdc.net/intel
[21:25] <DanielRichman> Sorry, this document isn't available from your location. If you feel this is a mistake, please email camdc-support at googlegroups.
[21:25] <Randomskk> ah, you might need to be internal
[21:25] <DanielRichman> what do oxford think about ssh+socks and then using spotify?
[21:26] <Randomskk> if you are capable of doing that kind of thing there's probably nothing they can do about it
[21:26] <DanielRichman> heh
[21:26] <Randomskk> can you tunnel through SSH twice? if I put an SSH tunnel to one server from one server and you connected to that server and tunneled through that, for instance?
[21:26] <DanielRichman> hmm... yes, I think so
[21:27] <Randomskk> e.g. cleric now has port 7070 setup as a socksv5 proxy to a ucam internal server, and you have access to cleric
[21:27] <DanielRichman> allright so if I -L 7070:localhost:9000
[21:27] Action: DanielRichman tries
[21:28] <DanielRichman> great success, it works
[21:28] <Randomskk> haha
[21:28] <Randomskk> nice
[21:28] <Randomskk> there's not really much on that page, just some info about what colleges can/can't access camdc
[21:30] <Randomskk> but anyway when your school's uni advisor person is like "so you should pick a college based on how it feels to you" they're talking crap
[21:30] <DanielRichman> nice, ty
[21:31] <Randomskk> the internet access is obviously much more important :D
[21:31] <DanielRichman> also, "Cambridge University Networking and Technical Support."
[21:31] <DanielRichman> hurr
[21:32] <Randomskk> generally speaking the cambridge network is quite good and you can do whatever
[21:32] <Randomskk> I run my own internal network for my stuff via an openwrt router
[21:32] <Randomskk> including my own wifi etc and my college are cool with that
[21:33] <Randomskk> they're also cool with us setting up this weather station though, they're generally quit egood
[21:33] <Randomskk> quite good*
[21:33] <DanielRichman> they do give you really nasty usernames though
[21:33] <Randomskk> ag611?
[21:34] <DanielRichman> mmm
[21:34] <Randomskk> compared to?
[21:34] <DanielRichman> almost as bad as my school's
[21:34] <DanielRichman> b02547
[21:34] <Randomskk> I'd say that's much worse
[21:34] <Randomskk> mine has my initials and a not-too-long number
[21:34] <DanielRichman> true
[21:35] <Randomskk> it's definitely not hard to remember and is pretty short
[21:35] <Randomskk> I don't know what else they could do
[21:35] <Randomskk> you can't let people pick one, that'd go so wrong
[21:35] <Randomskk> and you'd very quickly run into dupes with something like agreig or adam.greig
[21:35] <DanielRichman> yeah, even those that start out with the best intentions would eventually be forced to choose rediculous unames as tonnes are taken
[21:35] <DanielRichman> I guess 611 is yours for life?
[21:35] <Randomskk> yea
[21:36] <Randomskk> well I think it can get retired if I graduate and don't come back, but never reassigned
[21:36] <Randomskk> ag611@cam.ac.uk is only active for as long as I'm at the uni
[21:38] <DanielRichman> but there was no ag611 before you to sign up teh address for tonnes of spam
[21:38] natrium (~natrium@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) joined #highaltitude.
[21:38] <Randomskk> correct
[21:39] <Randomskk> maybe one day they'll have to delete old ones
[21:39] <Randomskk> if they're at 611 'ag's since they started
[21:39] <Randomskk> though of course just moving to four digits would give you a long time (some initials are already four digits)
[21:39] <DanielRichman> js?
[21:40] <Randomskk> he's 843
[21:40] <DanielRichman> nearly there
[21:40] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) joined #highaltitude.
[21:40] <Laurenceb> hi
[21:40] <Randomskk> yo
[21:41] <Laurenceb> wifi disconnects each time i use a lot of bandwidth :-/
[21:41] <DanielRichman> jesusC does indeed look nice, >31gb a few times and you get a £250 fine
[21:42] <Randomskk> !
[21:42] <Randomskk> blimey
[21:42] <Randomskk> selwyn is a bit middle I guess, it's not "have as much bandwidth as you like" but the fees are fair enough ish
[21:43] <Randomskk> the throttling is mean, but you can queue big downloads for the 2-7AM window
[21:43] <Laurenceb> useful for making my wifi behave :P
[21:44] <Laurenceb> think its the stupid non free drivers
[21:45] <Laurenceb> - Atheros, anyone else come across this?
[21:45] <Randomskk> I'd only heard good things about atheros on linux before
[21:46] <Randomskk> hmm, the latest version of the PCB I have for this thing I made is r4, but in eagle I have a PCB labelled r5 and I cannot remember what changed
[21:47] jnd (~jenda@unaffiliated/jnd) joined #highaltitude.
[21:48] <Randomskk> oh, god.
[21:48] <Randomskk> I just got out the box for these PCBs and found a few loose and about five wrapped in duct tape with some squiggles on it
[21:49] <Randomskk> which I instantly recognised from my weather station PCBs as their way of saying "these have been e-tested"
[21:49] <Randomskk> all the ones I made up were on non-tested boards
[21:49] <Randomskk> which rather explains the fault I found on one of them
[21:49] <Randomskk> grrr
[21:49] <Randomskk> that's silly
[21:49] <Randomskk> so, future reference, if getting PCBs from seeed studio, the ones with two lines on one board edge have been tested, the ones without haven't :P
[21:51] <Laurenceb> can they test all boards?
[21:51] <Randomskk> yes, but it costs a little more, usually they do 50%
[21:51] <Laurenceb> for an extra fee?
[21:51] <Randomskk> I don't think I realised that when I got these ones made
[21:51] <Randomskk> usually I only need one to five boards anyway so don't bothe
[21:51] <Randomskk> +r
[21:51] <Laurenceb> yeah
[21:52] blomlet (~quassel@unaffiliated/blomlet) left irc: Ping timeout: 240 seconds
[21:53] <Laurenceb> eek - 240B/s wifi
[21:53] <Laurenceb> maybe its interference
[21:53] <Laurenceb> really odd - its going from a few meg to vitually nothing
[21:55] <Randomskk> hah I have done something glorious on this PCB and I cannot remember what
[21:56] <Randomskk> there are components going across pads for totally different things in orders that were not intended
[21:56] <Randomskk> time to try and work out what the heck I was doing here
[22:03] SpeedEvil (1000@tor/regular/SpeedEvil) left irc: Read error: Operation timed out
[22:05] ejcweb (~chatzilla@ left irc: Ping timeout: 240 seconds
[22:11] SpeedEvil (1000@tor/regular/SpeedEvil) joined #highaltitude.
[22:13] jiffe (~jiffe@nsab.us) left irc: Ping timeout: 240 seconds
[22:13] jiffe (~jiffe@nsab.us) joined #highaltitude.
[22:13] stilldavid (~dave@68-64-214-18.static.forethought.net) left irc: Ping timeout: 240 seconds
[22:14] stilldavid (~dave@68-64-214-18.static.forethought.net) joined #highaltitude.
[22:14] <DanielRichman> hmm that's an inefficeint way to flush the kernel's disk cache. Download 3gb of dods data and then rm -Rf it
[22:19] ejcweb (~chatzilla@42.128.gr5.adsl.brightview.com) joined #highaltitude.
[22:32] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) left irc: Ping timeout: 245 seconds
[22:34] SpeedEvil (1000@tor/regular/SpeedEvil) left irc: Ping timeout: 240 seconds
[22:39] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) joined #highaltitude.
[22:40] RocketBoy (steverand@ left #highaltitude.
[22:46] Lunar_Lander (~lunar_lan@p54883848.dip.t-dialin.net) joined #highaltitude.
[22:46] <Lunar_Lander> hello
[22:46] <Upu> evening
[22:46] <Lunar_Lander> how's life?
[22:46] <Upu> very good I have a working flight computers
[22:46] <Upu> http://www.youtube.com/watch?v=C-bSepQ4CNs&feature=player_embedded
[22:46] <Upu> computer even
[22:46] <Lunar_Lander> cool
[22:47] <Lunar_Lander> I got my balloon jumpsuit done :D
[22:47] <Upu> so all I have to do now is make it a bit more flight worthy
[22:47] <Upu> balloon jump suit ?
[22:47] <Lunar_Lander> http://img230.imageshack.us/img230/3196/imag0029yg.jpg
[22:47] <DanielRichman> Upu: what resistor setup did you end up using?
[22:47] <Upu> http://ava.upuaut.net/wp-content/uploads/2010/08/ntx2schematic.png
[22:48] <Upu> close to the Alien one
[22:48] <Upu> Jumping out of a balloon Lunar ?
[22:49] <Lunar_Lander> nope
[22:49] SpeedEvil (~user@tor/regular/SpeedEvil) joined #highaltitude.
[22:49] <Lunar_Lander> I just thought having a jumpsuit is cool
[22:49] <Upu> lol
[22:49] <Lunar_Lander> even with unmanned HAB
[22:49] <Lunar_Lander> :D
[22:49] <Upu> haha
[22:49] <Randomskk> Upu: four whole resistors and /two/ IO lines? :P
[22:49] <Lunar_Lander> the blue flag is the one of Minnesota btw
[22:49] <Lunar_Lander> as this is the balloon state in the US so to speak xD
[22:49] <Upu> Randomskk it works
[22:50] <Randomskk> that's like 12.5% as good as one resistor and one IO :P
[22:50] <Upu> yadda :)
[22:50] <Randomskk> haha yea I know :P so long as it's working
[22:50] <Randomskk> what's more important is the status of the pink
[22:50] <Upu> taken me all week
[22:51] <Upu> I'm sure there are better ways of doing it but it seems to work nicely
[22:51] <DanielRichman> :D
[22:52] <Upu> my main issue is I only have 6 different values of resistors :)
[23:06] SpeedEvil (~user@tor/regular/SpeedEvil) left irc: Ping timeout: 265 seconds
[23:21] SpeedEvil (1000@tor/regular/SpeedEvil) joined #highaltitude.
[23:21] Upu (~Upu@ubn.upuaut.net) left irc:
[23:25] <fsphil> speaking of resistor setups, here's what I'll be using this time: 3.3v -> 330 ohm -> NTX2 tx pin <- 2.2k ohm <- rtty output pin
[23:25] <fsphil> two resistors and a single pin
[23:27] <Randomskk> not bad, but not as good as one resistor and one pin :P
[23:27] <fsphil> nope ;)
[23:33] earthshine (~earthshin@cpc3-orpi1-0-0-cust867.2-3.cable.virginmedia.com) left irc: Read error: Connection timed out
[23:34] <DanielRichman> http://hexoc.com/hab/predict/predict/#!/uuid=419062571f9d81af97314712f57cfa221abe6e10
[23:34] <Randomskk> that seems unlikely
[23:34] <DanielRichman> mmm
[23:34] <DanielRichman> doesn't do that in non-hd
[23:35] <DanielRichman> Randomskk: also have a look at ~daniel/dods/download_whole
[23:35] <DanielRichman> that script is a bit more gentle on the nomads servers. It takes 7 minutes to run and downloads the non-hd dataset
[23:35] <DanielRichman> resulting in 1.3gb of worldwide wind data
[23:35] <Randomskk> nice
[23:36] <Randomskk> that's another potential way forward then I guess
[23:36] <DanielRichman> I still don't think it's efficient. I'm going to edinburgh for a week tomorrow so won't be doing any more on my pred/ branch for a while but when I get back I'll consider it
[23:36] <Randomskk> jonsowman's away for a week as well
[23:36] <Randomskk> and I'm at work, boo
[23:37] <DanielRichman> oh and when that script was running it took ~20MBits; so about a fifth of our pipe
[23:37] <DanielRichman> so it shouldn't slow sstuff down too badly
[23:37] <DanielRichman> Randomskk: work :(
[23:37] ejcweb (~chatzilla@42.128.gr5.adsl.brightview.com) left irc: Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100723084720]
[23:41] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) left irc: Ping timeout: 245 seconds
[23:52] Laurenceb (~laurence@host86-145-126-39.range86-145.btcentralplus.com) joined #highaltitude.
[23:59] GeekShadow (~Antoine@reactos/tester/GeekShadow) left irc: Quit: The cake is a lie !
[23:59] Lunar_Lander (~lunar_lan@p54883848.dip.t-dialin.net) left irc: Quit: Nettalk6 - www.ntalk.de
[00:00] --- Mon Aug 16 2010