**** BEGIN LOGGING AT Mon Jul 24 03:00:02 2017
Jul 24 05:39:58 <xes> freemangordon: tomorrow I'll be back to "usual" boring life after a few days without keyboards....of course, except N900 ;) I hope to have some free time to move a few resources and prepare the VM
Jul 24 05:45:33 <freemangordon> xes: ok. It would be good to have something ready when it is needed, to not lose momentum. And trust from the devuan guys.
Jul 24 05:48:05 <Juesto> nice
Jul 24 07:21:41 <DocScrutinizer05> I guess creating a (sub)domain will take longer than preparing a VM
Jul 24 07:22:10 <DocScrutinizer05> I also wonder if we run short of IPs
Jul 24 07:25:01 <DocScrutinizer05> devuan.maemo.org or maemo.devuan.org?
Jul 24 07:25:42 <DocScrutinizer05> I guess our sysops will prefer the former
Jul 24 07:28:53 <DocScrutinizer05> I guess you might pimp the scratchbox VM. It must be totally static and pretty low load
Jul 24 07:32:16 <DocScrutinizer05> and the DNS VM is maybe still totally idle and non-funct
Jul 24 07:32:39 <DocScrutinizer05> I'm not sure it ever went active
Jul 24 07:33:24 <DocScrutinizer05> http://monitor.maemo.org/ganglia/
Jul 24 07:35:42 <DocScrutinizer05> LOL  http://monitor.maemo.org/ganglia/?c=maemo&h=dns1&m=load_one&r=hour&s=by%20name&hc=4&mc=2
Jul 24 08:02:13 <DocScrutinizer05> Last login: Wed Feb 27 03:07:55 2013 from p5493b2e2.dip.t-dialin.net
Jul 24 08:02:44 <DocScrutinizer05> X-P
Jul 24 08:04:33 <DocScrutinizer05> warfare: ^^^  would we ever need dns1 VM?
Jul 24 08:08:27 <DocScrutinizer05> xes_: ^^^
Jul 24 08:55:51 <DocScrutinizer05> ~# lastb -a|grep -c 59.45.175 ;-->24978
Jul 24 08:56:31 <DocScrutinizer05> chinese gvmt hackers
Jul 24 08:57:45 <DocScrutinizer05> https://www.abuseipdb.com/check/59.45.175.66
Jul 24 09:02:34 <DocScrutinizer05> .24,56,64,66,86,88
Jul 24 12:51:51 <xes> DocScrutinizer05: hum interesting... Let me check
Jul 24 12:59:24 <DocScrutinizer05> xes: iirc we created dns1 back when we hoped for hidden primary to Nokia's DNS
Jul 24 12:59:43 <DocScrutinizer05> prolly nobody ever looked into it since
Jul 24 13:00:45 <DocScrutinizer05> but for a repo it would need way more storage, and I don't know where from to get that. Could it run from aux directly?
Jul 24 13:01:33 <DocScrutinizer05> or what are the requirements for that VM?
Jul 24 13:03:29 <xes> DocScrutinizer05: we have enough resources for a not too large VM (enough to fullfill freemangordon's req.) ...but then i hope we could buy some piece of hardware to expand a blade or activate another one
Jul 24 13:03:58 <DocScrutinizer05> I vote for activating blade3
Jul 24 13:04:09 <DocScrutinizer05> :-)
Jul 24 13:04:16 <xes> :) me too
Jul 24 13:04:59 <DocScrutinizer05> the CPU and drives must be dirt cheap meanwhile
Jul 24 13:05:08 <DocScrutinizer05> not so sure about RAM
Jul 24 13:07:06 <xes> ..more or less.. Refurbished ram and cpu are very cheap. Drives not 
Jul 24 15:06:40 <buZz> woop, the batteries i ordered already shipped :D
Jul 24 15:06:57 <buZz> they claim it might arrive tomorrow already O_o , that would be weird
Jul 24 15:08:23 <Juesto> maybe they are prepared
Jul 24 15:08:24 <Juesto> :D
Jul 24 15:08:35 <Juesto> i mean
Jul 24 15:08:55 <Juesto> maybe they placed you in a free slot in a shipping order or something
Jul 24 15:08:57 <Juesto> hmm
Jul 24 15:09:05 <Juesto> actually scratch that
Jul 24 15:09:13 <buZz> :)
Jul 24 15:09:24 <buZz> its just germany -> netherlands
Jul 24 15:09:37 <buZz> so might really be within a day, but doesnt seem likely :P
Jul 24 15:10:38 <Juesto> perhaps
Jul 24 15:10:44 <Juesto> :)
Jul 24 15:26:46 <buZz> :)
Jul 24 15:27:03 <buZz> i'll mostly be happy if batterylife is over 1 day in offlinemode
Jul 24 15:27:58 <sixwheeledbeast^> Should be easy. I get a weekend out of mine with light usage
Jul 24 15:28:17 <sixwheeledbeast^> that's a several (3?) year old polarcell
Jul 24 15:28:22 <buZz> cool cool, i dont even want to use GSM but sometimes its useful
Jul 24 15:28:32 <buZz> without simcard in n900, battery last about a week
Jul 24 15:28:41 <buZz> simcard inside, offline mode, barely a day
Jul 24 15:28:42 <buZz> :(
Jul 24 15:29:29 <Juesto> fancy
Jul 24 15:29:36 <buZz> although everyone tries to ensure me that sim doesnt become active in offline mode, it seems it totally does
Jul 24 15:29:53 <sixwheeledbeast^> battery is likely exhausted unless you have something eating in the background
Jul 24 15:29:59 <Juesto> theres probably a way
Jul 24 15:30:58 <sixwheeledbeast^> I don't believe it would but I couldn't say for certain
Jul 24 15:31:42 <buZz> ah whichever, it'll be over soonish hopefully :)
Jul 24 15:31:46 <bencoh> I don't think it does, but I have no way to find out for now
Jul 24 15:32:15 <buZz> i believe this is the culprit ; https://en.wikipedia.org/wiki/U-TDOA
Jul 24 15:32:39 <buZz> or https://osmocom.org/projects/security/wiki/RRLP
Jul 24 15:33:20 <sixwheeledbeast^> I just disable cellular in terminal so I can have WLAN. If I don't want to take phone calls but need internet.
Jul 24 15:33:31 <buZz> reason i think this ; getting GPS lock without sim inside takes 15-20 minutes, with sim inside and offline mode, about 1 
Jul 24 15:33:55 <buZz> sixwheeledbeast^: yeah i still need to try if that has effect on gps lock speed
Jul 24 15:38:14 <sixwheeledbeast^> Are you comparing them like for like. If you have WLAN it can get A-GPS data via that.
Jul 24 15:38:41 <sixwheeledbeast^> You could have had different variables while testing
Jul 24 15:40:37 <jamesl> wait, A-GPS works over wifi?
Jul 24 15:40:42 <sixwheeledbeast^> I certainly don't get 1 minute GPS lock with data off. 
Jul 24 15:40:57 <sixwheeledbeast^> Well SUPL server
Jul 24 15:41:05 <sixwheeledbeast^> supl.nokia.com
Jul 24 15:41:37 <bencoh> actually not supl.nokia.com, but yeah
Jul 24 15:41:54 <bencoh> (I mean, you still need an entry in /etc/hosts)
Jul 24 15:42:10 <bencoh> I wonder if we could push a configuration file via cssu
Jul 24 15:42:27 <bencoh> something to update /etc/hosts, and/or dnsmasq configuration
Jul 24 15:43:53 <sixwheeledbeast^> It's a bit hacky, you still need supl.nokia.com in the settings but some locations need a /etc/hosts hack for DNS reasons I recall?
Jul 24 15:44:11 <bencoh> yeah
Jul 24 15:44:17 <bencoh> just some locations?
Jul 24 15:44:24 <sixwheeledbeast^> :nod: 
Jul 24 15:44:26 <bencoh> supl.nokia.com is an alias for nokia.supl.svc.ovi.com.
Jul 24 15:44:27 <bencoh> nokia.supl.svc.ovi.com is an alias for nokia.supl.svc.ovi.com.glb.as1248.net.
Jul 24 15:44:27 <bencoh> nokia.supl.svc.ovi.com.glb.as1248.net has address 127.0.0.1
Jul 24 15:44:32 <bencoh> this is what I get for now
Jul 24 15:45:13 <sixwheeledbeast^> doc had a link I'd have to find it now
Jul 24 15:46:05 <sixwheeledbeast^> It would be nice to fix the issue properly not provide a hack via CSSU.
Jul 24 15:46:34 <sixwheeledbeast^> ~logs
Jul 24 15:46:34 <infobot> All conversations are logged to http://infobot.rikers.org/%23maemo/ Lines starting with spaces are not logged. Logs are updated daily. http://mg.pov.lt/maemo-irclog/
Jul 24 15:46:54 <sicelo> doc's lin is/was https://www.whatismydns.com/#A/supl.nokia.com
Jul 24 15:47:53 <sicelo> fix what issue?
Jul 24 15:49:54 <sixwheeledbeast^> Get it working without having to hack hosts
Jul 24 15:50:30 <sicelo> but the problem isn't in maemo?
Jul 24 15:51:33 <sixwheeledbeast^> not in this instance but I also don't believe we can use another SUPL either if we wanted to host one
Jul 24 15:51:34 <sicelo> i doon't see how we can get supl.nokia.com to resolve to a real IP any other way
Jul 24 15:51:42 <sicelo> right.
Jul 24 15:52:10 <sixwheeledbeast^> bencoh: https://www.whatsmydns.net/#A/supl.nokia.com
Jul 24 15:52:21 <sicelo> we can use another. i know the google one definitely worked before
Jul 24 15:52:25 <sixwheeledbeast^> Resolves ok in USA for example
Jul 24 15:53:54 <sicelo> so yes, self-hosting could be a real possibility
Jul 24 15:56:12 <sixwheeledbeast^> I don't believe we can use the google one, I can't remember why. Otherwise the fix could be to switch to supl.google.com 
Jul 24 15:56:46 <sixwheeledbeast^> Not that I would want to let google handle my requests anyway :)
Jul 24 16:18:09 <DocScrutinizer05> buZz: please issue pnatd "at+cfun?" in "SIM, airplane" mode
Jul 24 16:18:58 <sicelo> yes google's doesnt work
Jul 24 16:19:20 <DocScrutinizer05> no way this behavior you observe makes any sense, except when an "STK application" on SIM runs circles when modem is in offline mode
Jul 24 16:21:49 <DocScrutinizer05> buZz: in airplane mode neither RRLP nor U-TDOA will work
Jul 24 16:22:28 <DocScrutinizer05> please issue pnatd "at+cfun?" in "SIM, airplane" mode, to save us from further speculating
Jul 24 16:24:46 <DocScrutinizer05> issues with TTFF differing between offline+SIM vs no-SIM may also be related to first rough estimate for location based on your last network provider you connected to. without SIM that is not available. However it's _not_ related to any RF transmission
Jul 24 16:26:32 <DocScrutinizer05> actually GPS even might enable GSM in +cfun=2 or whatever, to determine location from mere monitoring of visible networks, still with no RF TX. This also should work without SIM but might be broken since nobody in Nokia modem team ever tested it
Jul 24 16:26:59 <DocScrutinizer05> AT+CFUN? will tell us a lot about all this
Jul 24 16:30:05 <buZz> i will as soon as i'm back where the n900 is laying on charger :)
Jul 24 16:31:34 <DocScrutinizer05> ((an "STK application" on SIM runs circles)) particularly prone to that: SIM of providers with a "homezone" concept. They actually calculate your location on the SIM MCU
Jul 24 16:32:11 <DocScrutinizer05> bencoh: issuing a "AT+CFUN=0§ should fix any such issues for good
Jul 24 16:32:22 <DocScrutinizer05> bencoh: issuing a "AT+CFUN=0" should fix any such issues for good
Jul 24 16:32:56 <DocScrutinizer05> it's supposed to completely shut down modem (and GPS) and thus SIM
Jul 24 16:33:50 <DocScrutinizer05> obviously when you enable GPS, the middleware *will* re-enable moden via something like AT+CFUN=2 (or 4)
Jul 24 16:34:14 <DocScrutinizer05> otherwise the GPS in modem wouldn't be available at all
Jul 24 16:34:52 <DocScrutinizer05> but then, GPS itself eats LOTS of power
Jul 24 16:35:07 <DocScrutinizer05> you're not supposed to have it running all the time
Jul 24 16:36:49 <DocScrutinizer05> I could see how interleaving offline GPS-on online GPS-off commands could confuse the middleware
Jul 24 16:37:35 <DocScrutinizer05> possibly even a mere GPS-off doesn't exactly what it's supposed to do
Jul 24 16:38:19 <DocScrutinizer05> then you should try to fix that confusion with a offline->online->offline cycle
Jul 24 16:38:35 <DocScrutinizer05> or with pnatd AT+CFUN=0
Jul 24 16:43:37 <DocScrutinizer05> rotfl, sth funny with pnatd, at+cfun=0 makes it disconnect (surprise surprise - it connects to a modem it just has shut down completely)
Jul 24 16:43:59 <DocScrutinizer05> http://paste.ubuntu.com/25163802
Jul 24 16:44:47 <DocScrutinizer05> nfc what restarts modem when you restart pnatd
Jul 24 16:45:06 <DocScrutinizer05> blame csd and whatnot
Jul 24 16:46:07 <DocScrutinizer05> though... .oO(???)  AT interpreter should stay alive even with AT+CFUN=0
Jul 24 16:46:49 <DocScrutinizer05> prolly some watchdog kicks in when modem loses all radio activity ;-) unexpectedly
Jul 24 16:46:55 <sicelo> term frezes :)
Jul 24 16:47:24 <DocScrutinizer05> yeah, first time I had to disconnet ssj via ~.
Jul 24 16:47:31 <DocScrutinizer05> ssh*
Jul 24 16:47:56 <DocScrutinizer05> when you do locally, you need to close xterm or kill pnatd from another terminal
Jul 24 16:48:24 <DocScrutinizer05> for unclear reasons pnatd has no known way to quit it
Jul 24 16:49:22 <DocScrutinizer05> anyway I'm too lazy to switch my device to airplane mode the canonical way to check, right now
Jul 24 16:50:02 <DocScrutinizer05> just want to correct myself: AT+CFUN=0 might fail to work as expected, thanks to other stuff (watchdog) interfering
Jul 24 16:50:25 <DocScrutinizer05> might or might not, no idea
Jul 24 16:54:16 <DocScrutinizer05> playing with pnatd CFUN is considered safe anyway, except from maximal a reboot needed to fix stuff. ATS0=1 however is not as easy to fix again
Jul 24 16:54:34 <DocScrutinizer05> so don't do it unless you understand exactly what it does
Jul 24 16:55:15 <DocScrutinizer05> you could even ruin your SIM card by indiscriminately issuing AT commands you don't understand
Jul 24 16:56:07 <DocScrutinizer05> hint: POIN and PUK entry also is just a AT command. Don't touch those!
Jul 24 16:59:17 <DocScrutinizer05> http://seriss.com/people/erco/unixtools/hayes.html
Jul 24 17:15:29 <DocScrutinizer05> HAH!!! "please enter PIN-code" on device screen. That actually might explain ERROR reply to at+cfun?
Jul 24 17:21:29 <DocScrutinizer05> I think I know what happens: pnatd *is* the AT interpreter for ISI protocol to modem. pnatd is not prepared to handle CFUN states other than 1 (full). So after sending an ISI command to modem that's equivalent to +cfun=0 or +cfun=4, pnatd doesn't understand the answer from modem and so replies to +cfun? with "ERROR"
Jul 24 17:23:01 <APic> DocScrutinizer05 > *
Jul 24 17:23:01 <APic> 😉
Jul 24 17:23:02 <DocScrutinizer05> the modem however seems to act naturally, entering a CFUN state <>1 when you go airplane mode
Jul 24 17:23:10 <APic> 🙌
Jul 24 17:25:52 <DocScrutinizer05> and again: *no* transmission whatsoever in airplane mode is the whole purpose of that mode, so I can't see how RRLP or U-TDOA is involved. At best (worst) modem still *receives* aka scans for all visible networks and *might* work differently on that when a SIM is inserted to tell the modem which network(s) is the preferred one that needs particular attention
Jul 24 17:26:54 <DocScrutinizer05> so actually in airplane mode without SIM the modem might scan more bands and channels than with SIM inserted
Jul 24 17:27:35 <DocScrutinizer05> I don't see how it could be more busy with SIM than without, in airplane mode
Jul 24 17:28:00 <DocScrutinizer05> unless there's a STK app running on SIM
Jul 24 17:29:29 <DocScrutinizer05> http://paste.ubuntu.com/25163986
Jul 24 17:34:50 <DocScrutinizer05> at+cfun=4 (disable phone both transmit and receive RF circuits) doesn't seem to make the SIM PIN code requester pop up
Jul 24 17:35:05 <DocScrutinizer05> http://m2msupport.net/m2msupport/atcfun-set-phone-functionality/
Jul 24 17:35:53 <DocScrutinizer05> at+cfun=2/3 isn't supported according to at+cfun=?
Jul 24 17:36:58 <sicelo> N900 is a good/meek netizen, but I'd like for it to be more aggresive sometimes. gy 'good netizen' i mean - it is very happy to wait for network packets, etc. Droids don't do this. they demand attention ruthlessly. i'd like to (at least sometimes) have that on N900. any ideas?
Jul 24 17:37:06 <DocScrutinizer05> re TTFF also please note that a second fix, same day, is always way faster than the initial one
Jul 24 17:38:02 <DocScrutinizer05> sicelo: demand attention?
Jul 24 17:38:47 <DocScrutinizer05> what else than wait could a device do when it actually waits for a network packet
Jul 24 17:39:34 <DocScrutinizer05> you can adjust the duration it waits until the tcp/ip stack throws timeout error
Jul 24 17:40:06 <sicelo> i was wondering if iphbd couldn't be tweaked to controlthis somehow
Jul 24 17:40:10 <DocScrutinizer05> I bet there's a /proc node for that, at least in powerkernel
Jul 24 17:40:57 <sicelo> i'll try them too
Jul 24 17:41:38 <DocScrutinizer05> on my desktop PC I adjust that in "configuration" tool, so I don't know the right proc node off top of my head
Jul 24 17:42:11 <DocScrutinizer05> either /proc/net/whatever or even /dev/net/whatever, I guess
Jul 24 17:42:15 <sicelo> n900's network speeds are way slower than of any droid that i've seen .. even really 'weak' droids
Jul 24 17:42:31 <sicelo> i've seen those nodes before. will adjust and see
Jul 24 17:46:14 <DocScrutinizer05> http://www.sekuda.com/overriding_the_default_linux_kernel_20_second_tcp_socket_connect_timeout
Jul 24 17:48:16 <DocScrutinizer05> IroN900:~# cat  /proc/sys/net/ipv4/tcp_syn_retries 
Jul 24 17:48:17 <DocScrutinizer05> 5
Jul 24 17:49:00 <DocScrutinizer05> I'm still not sure I understand what exactly you want to achieve
Jul 24 17:50:58 <DocScrutinizer05> n900's perceived network speed depends on a friggin lot of factors, starting at WLAN chip and connection and probably ending at swap needed for rendering huge web pages in vrowser
Jul 24 17:52:27 <DocScrutinizer05> first define the problem please, very specifically. Then we can look into what's the culprit and what might get done about it
Jul 24 17:55:10 <sicelo> my problem is in cellular Internet. unacceptably slow, hangs. i know the ISP itself is bad, but droids somehow do way better. not high-end droids
Jul 24 17:55:16 <DocScrutinizer05> sftp for example is quite slow on my N900 but I think that's due to en/decrypting overhead which is heavy on CPU
Jul 24 17:56:03 <DocScrutinizer05> browsing is slow mainly due to a lack of available RAM
Jul 24 17:56:08 <sicelo> if i keep a running ping, it's a bit better
Jul 24 17:56:17 <DocScrutinizer05> oooh that
Jul 24 17:56:37 <DocScrutinizer05> that's about AP and PSM
Jul 24 17:56:44 <DocScrutinizer05> I think
Jul 24 17:57:13 <DocScrutinizer05> or about swappyness in N900 kernel
Jul 24 17:57:17 <sicelo> DocScrutinizer05: gprs
Jul 24 17:57:40 <DocScrutinizer05> gprs or umts?
Jul 24 17:58:05 <DocScrutinizer05> aka 2G or 3G?
Jul 24 17:59:17 <DocScrutinizer05> on 3G there's indeed a several-seconds long grace span in which the channel stays established (and even TX constantly running, thus causing UMTS eating more power than GPRS on slow pings)
Jul 24 18:00:26 <DocScrutinizer05> at full rate however the situation trips: UMTS is way faster and thus finishes earlier and needs less battery
Jul 24 18:01:20 <DocScrutinizer05> so for shell and irc etc you want GPRS/2G while for browsing and bulk downloads you want UMTS/3G
Jul 24 18:02:24 <DocScrutinizer05> when you say a ping improves the situation, I conclude you're talking about 3G
Jul 24 18:03:44 <sicelo> even in 2G it's suboptimal, but on 3G it's completely bad. 
Jul 24 18:04:21 <DocScrutinizer05> try a ping -f
Jul 24 18:04:33 <DocScrutinizer05> -f like flood
Jul 24 18:05:03 <DocScrutinizer05> if that improves stuff in 2G, it's likely a problem on provider side
Jul 24 18:05:37 <DocScrutinizer05> particularly when you got some free RAM still
Jul 24 18:06:22 <DocScrutinizer05> use htop to make sure your system isn't running busy
Jul 24 18:07:23 <DocScrutinizer05> weird ipfilters/ipchains settings may cause issues
Jul 24 18:07:34 <APic> lol
Jul 24 18:07:53 * APic ♥ DocScrutinizer05 
Jul 24 18:08:04 <DocScrutinizer05> make sure you have no leftovers from an occasional experiment with "forwarding internet via USB" or whatever
Jul 24 18:08:06 <APic> (Not in a homosexual Way, just in a gayprideness-Way!  😉)
Jul 24 18:08:14 <APic> s/gay/happy/
Jul 24 18:08:14 <infobot> APic meant: (Not in a homosexual Way, just in a happyprideness-Way!  😉)
Jul 24 18:08:19 <APic> 😎
Jul 24 18:08:23 <KotCzarny> apic: isn't it the same?
Jul 24 18:08:27 <APic> KotCzarny: True.
Jul 24 18:08:45 <APic> KotCzarny: But i would never insert my Penis into DocScrutinizer05 Arse. That Stuff i refer to.   ☺
Jul 24 18:08:55 * APic just thinks You Gals/Guys here are nice!
Jul 24 18:08:56 <APic> 😎
Jul 24 18:09:00 <KotCzarny> so, passive then? :P
Jul 24 18:09:05 <APic> In a helpful Way.
Jul 24 18:09:12 <APic> No, not that Way around either. sorry.   😉
Jul 24 18:09:29 <APic> How would my Arse fit into DocScrutinizer05's Penis anyhow?     😉
Jul 24 18:09:30 <KotCzarny> platonic?
Jul 24 18:09:33 <APic> True!
Jul 24 18:09:36 <APic> You got it!
Jul 24 18:09:36 <APic> 😎
Jul 24 18:09:37 <DocScrutinizer05> no sex, I'm british err an old fart
Jul 24 18:09:41 <APic> Gut.
Jul 24 18:09:43 <APic> 🙌
Jul 24 18:21:00 <DocScrutinizer05> a pretty much OT post that nevertheless I like reading :-)   http://www.vanille.de/openmoko-10-years-after-mickeys-story/
Jul 24 18:21:32 <DocScrutinizer05> sentimental journey ;-D
Jul 24 20:55:30 <kalin> Hi all
Jul 24 20:57:12 <kalin> I was have many issues with my n900 and i flashed it with PR1.3
Jul 24 20:58:02 <sixwheeledbeast^> what sort of issues?
Jul 24 20:58:11 <kalin> Finaly i have installed pyOBD on my phone
Jul 24 20:58:27 <kalin> i was unable to install anything
Jul 24 20:59:14 <kalin> but now when try to start pyOBD it crashed
Jul 24 21:01:15 <kalin> can someone help me to setting up pyOBD?
Jul 24 21:02:33 <sixwheeledbeast^> I imagine you are having repo issues? I have never had an issue with pyOBD.
Jul 24 21:03:38 <kalin> no, it isn't a repo issue
Jul 24 21:04:14 <kalin> i think it is a side effect of instaling audacious
Jul 24 21:04:44 <sixwheeledbeast^> well what happens if you start pyOBD from a terminal? Is audacious python?
Jul 24 21:05:08 <kalin> no no no
Jul 24 21:06:03 <kalin> when i install audacios the phone had a broken/missing packages
Jul 24 21:06:37 <kalin> this is a second time that i have to reflash the phone after audacious
Jul 24 21:07:12 <kalin> so now i have fresh flashed n900 with only pyOBD installed
Jul 24 21:09:05 <kalin> and when i try to start pyOBD the phone loading something few seconds and close the app
Jul 24 21:09:41 <sicelo> 23:04 < sixwheeledbeast^> well what happens if you start pyOBD from a terminal?
Jul 24 21:09:45 <sicelo> ^
Jul 24 21:10:10 <kalin> whait a second to try
Jul 24 21:10:25 <sixwheeledbeast^> ~audacious
Jul 24 21:11:22 <kalin> did i have to be root?
Jul 24 21:11:49 <sixwheeledbeast^> er I would hope not
Jul 24 21:12:28 <kalin> the terminal says not found
Jul 24 21:14:01 <sicelo> root's needed to establishh the rfcomm connection .. butt iiirc it calls sudo for that .. also haven't used it in a hwile
Jul 24 21:14:44 <sixwheeledbeast^> python /opt/pyobd/pyobd.py
Jul 24 21:16:30 <kalin> "No module named dbus"
Jul 24 21:17:38 <sicelo> how did you install pyobd ...
Jul 24 21:18:14 <sicelo> because python-dbus *is* a dependency  .. so install that
Jul 24 21:18:21 <kalin> App manager
Jul 24 21:18:40 <jamesl> ~flasher
Jul 24 21:18:40 <infobot> refer ~flashing; http://www.jedge.com/n810/flasher/maemo_flasher-3.5_2.5.2.2.tar.gz (also .exe!), or http://www.chakra-project.org/ccr/packages.php?ID=5027 or generally http://www.google.com/search?q=maemo_flasher-3.5_2.5.2.2.tar.gz.   HARMattan(N9): https://aur.archlinux.org/packages/fl/flasher-harmattan/flasher-harmattan.tar.gz; list of filenames/md5sums: http://pastebin.com/sYKdNJSH, or http://galif.eu/nokia/
Jul 24 21:19:09 <jamesl> ~emmc
Jul 24 21:19:09 <infobot> emmc is probably is http://nds2.fds-fire.nokia.com/fdp/interface/FiRe/2010/5/--FID--A0A22YHFSICNA/--LID--FiRe1275051276916/AE98ED9D_RX-51_2009SE_10.2010.13-2.CENTRAL-EUROPE_PR_EMMC_MR0_ARM.bin or see ~emmc2
Jul 24 21:22:13 <kalin> thanks sicelo , i will try today because first i have to install rootshell to use apt-get and add repo mirror
Jul 24 21:22:24 <kalin> now i will go to get some sleep
Jul 24 21:22:30 <kalin> GN to all
Jul 24 21:28:36 <ff_> sicelo: try load CPU to 100% and then check internet connection
Jul 24 21:28:57 <ff_> accidently I found that this matter
Jul 24 21:29:16 <ff_> accidently I found that this matter
Jul 24 21:29:29 <ff_> when I download through wifi with 100% load CPU by for example encoding audio, average speed is about 96% of max available
Jul 24 21:30:02 <ff_> but when nothing use CPU, download average speed decrease to about 65% of max available
Jul 24 21:30:19 <ff_> maybe the same effect is with 2G/3G GSM internet connection
Jul 24 21:34:57 <sixwheeledbeast^> Er so there is an issue with repos or broken packages of some sort
Jul 24 21:48:37 <ff_> it's look like CPU's sleep timer introduce quite a lot delay into connection response
Jul 24 22:31:20 <DocScrutinizer05> hmm, that might make sense. Though probably because of packet / timer aggregating that's done in maemo to keep CPU in suspend mode as much as possible, and only do pending jobs (timer, IO) on bursts every now and then
Jul 24 22:34:15 <DocScrutinizer05> to avoid this, probably you don't need 100% load on CPU, what you need is frequent forced wakeups so the IO can get done,  while sleep 0.02; do :; done
Jul 24 22:37:14 <DocScrutinizer05> or simply slide with your fingernail over the screen in a circular or other constant movement and see if IO speeds up during that. Touch controller will produce IRQs plenty
Jul 24 22:37:55 <DocScrutinizer05> sicelo: ^^^
Jul 24 22:39:00 <DocScrutinizer05> sixwheeledbeast^: I thik that's the right direction
Jul 24 22:39:59 <DocScrutinizer05> ~lf
Jul 24 22:40:00 <infobot> i guess #maemo lazyflashing is http://wiki.maemo.org/Updating_the_tablet_firmware#The_Lazy_Approach
Jul 24 22:41:36 <sixwheeledbeast^> busybox sleep is only whole numbers?
Jul 24 22:41:54 <DocScrutinizer05> yeah
Jul 24 22:41:57 <DocScrutinizer05> sucks
Jul 24 22:42:02 <DocScrutinizer05> ~messybox
Jul 24 22:42:02 <infobot> messy... err busybox is meant for lean scripting. Regarding all the missing options and immanent limitations (see su, passwd, nice, ps, diff as used by mc...) it's not really the interactive shell of choice. A lot of people hate busybox because a lot of system integrators don't understand the difference between busybox and a decent user interactive shell plus unix utils
Jul 24 22:42:26 <DocScrutinizer05> it might have msleep or usleep
Jul 24 22:42:39 <sixwheeledbeast^> in fact busybox sleep has no arguments I recall. just full numbers
Jul 24 22:42:40 <DocScrutinizer05> don't know
Jul 24 22:43:17 <Juesto> busybox is just a swiss army knife unix utils
Jul 24 22:43:18 <DocScrutinizer05> sucks, it has neither msleep not usleep
Jul 24 22:43:23 <Juesto> it is intended for embedded devices
Jul 24 22:43:47 <DocScrutinizer05> jiemy take on busybox you see above
Jul 24 22:43:48 <Juesto> go compare the size of a bunch of unix utils to busybox binary with the same utils
Jul 24 22:44:03 <DocScrutinizer05> who cares?
Jul 24 22:44:18 <Juesto> :/
Jul 24 22:45:06 <sixwheeledbeast^> er yes, the issue is our little devices are useful to have some better functionality than other embedded devices.
Jul 24 22:45:18 <DocScrutinizer05> I installed quite a bunch of gnu utils plus mandb and still no problems with free space on /
Jul 24 22:45:49 <sixwheeledbeast^> bash4 has an opt version I believe
Jul 24 22:47:15 <DocScrutinizer05> the major advantage of busybox is: it's a single monolithic binary, so the system doesn't need to load it again for 99% of scripts written for it
Jul 24 22:47:42 <Juesto> yep
Jul 24 22:47:46 <DocScrutinizer05> it's basically an interpreter of its own, for scripts
Jul 24 22:48:11 <DocScrutinizer05> this is mainly a performance advantage
Jul 24 22:48:43 <DocScrutinizer05> thus buybox is nice for lean scripting, like e.g. by using it instead of init
Jul 24 22:49:15 <DocScrutinizer05> or to run init scripts
Jul 24 22:49:53 <DocScrutinizer05> but honestly, messybox is pathetic, do e.g a `ps --help`
Jul 24 22:51:06 <DocScrutinizer05> I know there's a pimped version that has more compile options for functions
Jul 24 22:51:29 <DocScrutinizer05> but it still stays sub-par compared to proper unix toolbox
Jul 24 22:51:47 <Juesto> o.o
Jul 24 22:52:12 <DocScrutinizer05> particularly when it comes to cmds_of(8)
Jul 24 22:52:37 <DocScrutinizer05> nice is borked, su is borked and can't ever get fixed
Jul 24 22:53:02 <DocScrutinizer05> meh, I ranted so many times about that thing I get bored doing it
Jul 24 22:56:46 <sixwheeledbeast^> I wonder if you could replace it for the latest version. Seen as Maemo is 1.10.2 and the latest is 1.27.1
Jul 24 22:57:26 <Juesto> probably if you compile from good sources sixwheeledbeast^ 
Jul 24 22:59:21 <Wizzup> $ busybox sleep --help
Jul 24 22:59:21 <Wizzup> BusyBox v1.25.1 (2017-01-25 13:22:31 CET) multi-call binary.
Jul 24 22:59:21 <Wizzup> Usage: sleep [N]...
Jul 24 22:59:21 <Wizzup> Pause for a time equal to the total of the args given, where each arg can
Jul 24 22:59:24 <Wizzup> have an optional suffix of (s)econds, (m)inutes, (h)ours, or (d)ays
Jul 24 22:59:36 <Wizzup> $ time busybox sleep 0.1s
Jul 24 22:59:36 <Wizzup> real	0m0.105s
Jul 24 22:59:36 <Wizzup> user	0m0.001s
Jul 24 22:59:36 <Wizzup> sys	0m0.005s
Jul 24 22:59:54 <Wizzup> further questions?
Jul 24 23:00:53 <Wizzup> Being stuck forever on an old version of software doesn't make it crap nor messy. It just illustrates how important a healthy ecosystem is, where packages are kept up to date
Jul 24 23:00:56 <Juesto> :p
Jul 24 23:01:20 <Wizzup> busybox is one of the best things out there
Jul 24 23:01:49 <Wizzup> And you can easily build a whole distro with/around it - see alpine linux
Jul 24 23:02:51 <DocScrutinizer05> IroN900:~# time busybox sleep 0.1s
Jul 24 23:02:52 <DocScrutinizer05> sleep: invalid number '0.1s'
Jul 24 23:03:52 <Wizzup> >software that is 7 years old
Jul 24 23:04:41 <DocScrutinizer05> budybox is a monolithic binary that does multiple things and none of them correctly. Just like systemd ;-)
Jul 24 23:04:54 <Wizzup> This is the most retarded thing I've read today
Jul 24 23:05:03 <DocScrutinizer05> great :-P
Jul 24 23:05:37 <Wizzup> busybox is the exact opposite of systemd, it's lean, minimal, clean. And as I said, you can build an entire distro with decent init system with it.
Jul 24 23:05:52 <Wizzup> alpine linux is really just openrc+busybox+apk (their package tool)
Jul 24 23:05:57 <Wizzup> it's amazingly fast and lean
Jul 24 23:06:08 <Wizzup> their networking scripts are at least as decent as debian ones
Jul 24 23:06:10 <DocScrutinizer05> a bag of bugs
Jul 24 23:06:32 <Wizzup> smh
Jul 24 23:09:33 <sixwheeledbeast^> So if busybox can be updated it should be done via CSSU? Seen as it's a system package in the SSU repo
Jul 24 23:10:13 <Wizzup> maybe, but I'm not sure what you'd gain.
Jul 24 23:10:29 <Wizzup> You'd have to check if it still supports such old kernels (likely)
Jul 24 23:10:47 <Wizzup> can probably compile it statically, too
Jul 24 23:11:03 <sixwheeledbeast^> well sleep to decimal places for starters. I assumed you ran the above on an N900?
Jul 24 23:11:07 <Wizzup> in fact, I have a armv7 static busybox here on my laptop
Jul 24 23:11:16 <Wizzup> I will scp it to my n900 and see
Jul 24 23:12:46 <Wizzup> oh, it's a PIE
Jul 24 23:12:52 <Wizzup> doubt the n900 can handle PIE :p 
Jul 24 23:12:57 <Wizzup> the n900 kernel*
Jul 24 23:13:15 <Wizzup> herp.
Jul 24 23:13:20 <Wizzup> I copied the powerpc binary.
Jul 24 23:13:22 <Wizzup> go me.
Jul 24 23:14:04 <Wizzup> Nokia-N900:~# ./bb --help
Jul 24 23:14:05 <Wizzup> FATAL: kernel too old
Jul 24 23:14:05 <Wizzup> Segmentation fault
Jul 24 23:14:06 <Wizzup> boom
Jul 24 23:14:10 <Wizzup> toldya
Jul 24 23:14:33 <Wizzup> software needs to stay in sync ;) 
Jul 24 23:16:09 <DocScrutinizer05> sixwheeledbeast^: please sonsider that all scripts are tested against exactly this version of busybox with all its bugs
Jul 24 23:16:44 <DocScrutinizer05> there's a busybox-power which everybody interested can install instead or on top of the system busybox
Jul 24 23:17:52 <DocScrutinizer05> users rendered their system fcked up simply by defining bash instead busybox as user user's default shell
Jul 24 23:18:44 <sixwheeledbeast^> The latest version has allsorts of extra utils as standard. There's also a size question too.
Jul 24 23:29:05 <DocScrutinizer05> you're aware that when you use busybox with anything not-busybox as shell, the binaries your system needs to load for each command are way larger than what they would be in genuine decent unix toolbox? Since each time the multi-purpose huge busybox binary needs to get loaded and started when your shell calls $whatever in stat, ls, dir, date, `busybox --help`
Jul 24 23:32:53 <DocScrutinizer05> I just might mention that e.g. jolla/mer decided against busybox and for a 'normal' sheel afaik, after long pondering
Jul 24 23:33:20 <DocScrutinizer05> shell* plus gnu utils
Jul 24 23:49:53 <DocScrutinizer05> http://paste.ubuntu.com/25166165
Jul 24 23:57:16 <DocScrutinizer05> and http://paste.ubuntu.com/25166208
Jul 25 00:03:35 <DocScrutinizer05> btw it seems like for N900, busybox-power as system shell actually reduces availble free space on /, vs busybox('normal') and gnu utils on opt
Jul 25 00:04:47 <DocScrutinizer05> reason: system shell can't live on /opt. And busybox-power of course is larger than busybox-'normal'
Jul 25 00:06:31 <Wizzup> 'huge busybox binary needs to get loaded' sounds kind of weird
Jul 25 00:06:39 <Wizzup> It usually lives in the cache
Jul 25 00:13:38 <Wizzup> and linux also doesn't need to load the whole binary either
Jul 25 00:14:10 <Wizzup> that's usually loaded on demand
Jul 25 00:14:41 <Wizzup> so I'm not sure how much that really matters.
Jul 25 00:15:59 <Wizzup> #justtechnicalthings
Jul 25 00:29:43 <DocScrutinizer05> I'm not going to embarrass myself by answering with a quote
Jul 25 00:33:55 <Juesto> its difficult to find a n900 nowdays
Jul 25 00:33:56 <Juesto> derp
Jul 25 00:34:38 <Juesto> how does busybox behaves on something like Mach
Jul 25 00:34:40 <Juesto> or BSD?
Jul 25 00:36:19 <Wizzup> busybox is for linux
Jul 25 00:36:32 <Wizzup> actually, I am not sure about that
Jul 25 00:37:04 <Wizzup> I know BSD was working on their own - http://landley.net/code/toybox/
Jul 25 00:37:42 <Wizzup> https://www.freshports.org/sysutils/busybox/
Jul 25 00:37:49 <Wizzup> there's your answer
Jul 25 00:37:55 <Wizzup> the same as on most posix compliant OSes I guess
Jul 25 00:40:09 <Juesto> hm
Jul 25 00:43:01 <Wizzup> almost 3am.
Jul 25 00:43:03 <Wizzup> ttyl
Jul 25 00:52:45 <Oksana> Like Nokia N900, but without keyboard and FM transmitter, apparently. http://www.gsmarena.com/samsung_s8500_wave-3146.php Also, capacitive touchscreen, not resistive. Similar processor, same 800x480 display resolution.
Jul 25 00:53:44 <Oksana> Possible to change OS on it: https://www.sammobile.com/2012/06/26/android-port-for-the-samsung-wave-gt-s8500/
Jul 25 00:55:26 * Oksana wonders whether it would be a nice porting target for Maemo 5... But let's check first if the Android port managed to get cellular to work.
Jul 25 00:55:44 <Wizzup> why not droid4?
Jul 25 00:59:06 <Oksana> Either I am very poor at search, or something, because I cannot find droid 4 on EBay
Jul 25 00:59:19 <Oksana> Or Gumtree
Jul 25 00:59:24 <Oksana> At the same time, https://www.gumtree.com.au/s-ad/bonbeach/other-phones/samsung-wave-gt-s-8500-mobile-phone/1153376589
Jul 25 01:01:44 <Wizzup> motorola droid 4
Jul 25 01:02:58 <Oksana> Yes, I searched for motorola droid 4. Nice article in Wikipedia...
Jul 25 01:04:43 <Wizzup> i got 4 from ebay, arriving tomorrow
Jul 25 01:12:16 <Oksana> Nice :-)
Jul 25 01:41:28 <stryngs> Anyone at DEFCON this yr?
**** ENDING LOGGING AT Tue Jul 25 03:00:01 2017