[00:16] * dejello pokes the tablet sitting next to him, quit bootlooping! :P [02:00] I realize it's not a high priority, but is it planned that Ubuntu Touch will be able to run on galaxy S4? [02:42] Debolaz, This was an earlier answer to the same question referencing an sgs4. It's rather less trivial than on PCs, yes. https://wiki.ubuntu.com/Touch/Porting [02:43] same phone? === chihchun__afk_af is now known as chihchun_ [04:14] Hey [04:14] Does anyone know if any further specs have been discussed? [04:14] Such as ARMv7 vs ARMv8? [04:15] As I'm curious to how they plan on running Andriod if they go the ARMv8 route === Namidairo`bnc is now known as Namidairo === chihchun_ is now known as chihchun__afk === chihchun__afk is now known as chihchun__afk_af [06:17] Hey guys, has anyone encountered an issue where you can't click some of the settings? === chihchun__afk_af is now known as chihchun === chihchun is now known as chihchun_afk [07:08] good morning [07:13] good morning === nik90 is now known as nik90|Office [07:28] will work for a Ubuntu Edge. :P === chihchun_afk is now known as chihchun [07:49] ForbiddenX, there is nothing such a click could open yet for most of them :) === dholbach_ is now known as dholbach === chriadam is now known as chriadam|away [08:16] i have a samsung galaxy s2, and i want to install ubuntu touch, but apparently calls /sms doesn't work. https://wiki.ubuntu.com/Touch/Devices/i9100 is there a fix [08:23] ogra_: hello [08:23] ogra_: when are 25 images coming out? [08:23] seems we were like super close yesterday [08:27] Good morning all, happy Louise Brown Day! :-D === schwuk_away is now known as schwuk [08:29] balloons: ping [08:41] hi ,can I just follow "Manual Installation" section to install ubuntu-touch (no previous step needed??) [08:41] https://wiki.ubuntu.com/Touch/Install [08:45] asac, build starts at 10:00 UTC === Ursinha-afk is now known as Ursinha [08:45] (and takes 50min) [08:46] ogra_: cool. can it start before i wake up? [08:46] at best even finish before that :) [08:46] :-P [08:46] asac, no [08:47] i didnt expect a no :) [08:47] (i wish it could, but then daily-release doesnt end up on the image) [08:47] what can make this not possible? [08:47] asac, daily-release would have to land earlier [08:47] asac, hire people who are in a !european tz so reviews are done at 3am utc [08:48] yeah [08:48] there you got your solution :) [08:48] asac, the issue is that atm didrocks&co review/approve change when they start their day [08:48] and they are all european based [08:48] so time for them to start, review, etc it's at least 7utc before things get in [08:48] asac, before 8 UTC is hard because we share the livefs builder among all images and it would mean to re-schedule all of them [08:48] then publisher time, etc [08:48] seb128: so punching the packages from daily-release ppa to the archive? thats done when didrocks starts his day? [08:49] ogra_: why can't we have two or three livefs builders? [08:49] is that a non-reproducible machine? or an unreproducible concept? [08:49] asac, well, in practice it's done automatically ... excepted when there are changes that need manual review (e.g packaging changes, api changes, etc) [08:49] asac, we do have a fallback one thats highly unstable (we have two) [08:49] err unscalable [08:49] asac, once infinity has the calxeda machine up that should change, we're just waiting for that [08:49] ogra_: livefs build happens medival style on native panda boards? [08:49] asac, the issue is that things move in a way that we have changes to reviews manually almost every day, always new api, compat changes, new depends, etc [08:50] asac, failing tests, flacky utah,e tc [08:50] seb128: no more rants about utah. the flakiness is everywhere [08:50] asac, it has to, there are still packages in the archive that break under qemu [08:50] asac, sorry, was an easy one, just ignore that part [08:51] seb128: but it is... its actually everywhere where we have jenkins :) [08:51] asac, but yeah, the issue is that for a number of reasons, we need manual reviews to get things in, and those are done by european, and their work hours lead to the current build start at 10utc [08:51] seb128: right. i see [08:51] i actyually believe that manual review isn't such a bad thing [08:52] imho we should do two rounds a day [08:52] especially since we found that our daily-release testing doesnt really match what we see on real images (so calibration has to happen first) [08:52] ++ [08:52] the U.S guys should do one in their end of day [08:52] we should also have two image builds [08:52] but that requires thee tests to finish in a reaonable time [08:52] if the u.s guys were doing reviews at end of day we would have a mostly current image by european morning [08:52] the 4-5h it currently takes arent reasonable yet [08:52] seb128: tell me who works as reliable and as strongly as the euro guys from US that could do it? [08:52] reality is that there is not a lot happening during the night [08:53] yeah you are right [08:53] we should have 24/5 [08:53] 5 ? [08:53] push on the CI machine [08:53] only ... ? [08:54] ogra_: sure. why not get such a great coverage and still hav weekend [08:54] pfft, weekend [08:54] hire someone for the weekend shifts :) [08:54] better than have like a machine that always needs to be kickstarted every morning :) [08:54] is to just need to do that once on monday morning tokyo time [08:54] ogra_: we want folks to not work on weekends [08:54] hire persia then :) [08:55] if there is a firedrill then everybody will have to work weekend [08:55] asac, well, we have a few people in the U.S in our team, e.g kenvandine is in didrocks' team and do review, he just does them when he wakes up (which is middle of the day for us) ... if we were to do an evening build we could probably get Ken to do end of day reviews rather than start of day ones [08:55] * ogra_ bets there are people that would hapiily apply for a weekend job :) [09:12] Hi will ubuntu touch work on my samsung galaxy s 4g? [09:12] Hi will ubuntu touch work on my samsung galaxy s(1) 4g? [09:15] anyone? [09:15] asac,ogra_: we're working on the livefs build issue in the releng sprint this week [09:16] you mean migrating to the calxeda box ? [09:16] no, I mean moving them into Launchpad [09:16] And also in parallel moving Launchpad's armhf builders to Calxeda, but that's more IS' problem [09:16] will ubuntu touch work on my samsung galaxy s(1) 4g? [09:16] ? [09:17] asac, ^^^ [09:17] A prerequisite is build cancellation (so that we don't end up being unable to progress livefs builds due to long-running package builds that we can't do anything about), which I'm personally working on right now [09:17] yay [09:22] cjwatson: hey. is livefs nowadays just live-build? [09:22] e.g. can developers reproduce what launchpad will do easily on their own desk? [09:24] asac: None of the work we're doing changes how developers would reproduce it, at all - we're just moving the trigger [09:24] asac: It's livecd-rootfs on top of live-build. Instructions for reproducing: https://lists.ubuntu.com/archives/ubuntu-devel/2011-June/033458.html [09:25] cjwatson: yeah. hoped we would make it easier as well :)... but seems we dont tackle that for now. so fine :) [09:25] or wait... let me quickly read the post [09:25] We should make it easier as well, certainly [09:25] i guess we should field test it with a few engineers [09:25] :) [09:26] e.g. can you try to produce an image [09:26] like how design folks do it [09:26] and then be depressed seeing how lost they are :) [09:27] Well, I think I know there are problems without the need for "user" testing here - it'd be a worthwhile project to make that much simpler for developers to run locally [09:27] It's just not today's project and not related to making the official builds less bottlenecked :) [09:28] if you want an actually identical live builder setup it is pretty complex to do (with BuildLiveCD etc) .... i think we should have a setup script for this [09:28] BuildLiveCD itself is functionally trivial [09:28] surely not a super high prio task for 13.10 though [09:29] cjwatson, the stacked chroot setup isnt if you dont know which chroots, how they are named and where the /build dir has to live etc [09:29] It's not necessary to make it identical to the datacentre setup, with multi-machine dispatch etc. [09:29] For a local-developer setup a simple stub would suffice [09:29] no, but the stacked chroots are needed if you wany it as close as possible [09:30] which requires a certain structure we dont have documented [09:30] (well, not publically at least) [09:30] I don't think that's necessary. As long as you have the dependencies installed (livecd-rootfs basically) then live-build will deal with creating the inner chroot. [09:30] it didnt for me :) [09:30] I'm not concerned about having an outer chroot. [09:30] Even if you happened to do something wrong in one test :) [09:31] I run livecd-rootfs without the need for an outer chroot all the time. [09:31] It's fine. [09:33] What I'd probably like to do is to add a mode to cdimage where you can have it do the livefs build locally rather than sshing to a remote machine to do it (and generally make that setup much more table-driven while we're there), and generally fix up all the bits that make it hard to run cdimage locally. [09:34] Doing that right would make it easy for developers to run a full local image build. It's all possible right now (there's no secret code or anything), but it requires too much setup for most people. [09:34] is there any instruction or guide about how to port a custom android device to ubuntu-touch? [09:34] cjwatson, yeah [09:34] It wasn't until fairly recently that we got permission to open the full image build code at all; the previous state where part of it was closed held us back from doing this kind of productisation. [09:34] though thats definitely not a 13.10 task :) [09:35] dlan, see the channel topic, there is a link to the porting guide [09:39] ogra_: :) [09:39] ogra_: can you give me confirm that pitti is right? [09:40] ogra_: forwarded mail [09:40] pitti us usually right ... but what are you referring to ? [09:40] he says media player is really busted [09:40] ah, heh [09:40] so the test failure we see there might be a real one [09:40] ogra_: can you confirm that its busted for you? [09:40] on your phone? [09:40] he hasnt tried the yet unreleased new gstreamer stak i suspect [09:40] well he tries the image [09:40] should land this week [09:40] i am only bothered about the image yet [09:40] so you say mediaplayer is busted? [09:41] it works fine on maguro, is completely busted on grouper is what i know [09:41] ogra_: on maguro we have success, so that matches what you say [09:41] (havent tested today yet and my maguro is busy ... but i can check later) === schwuk is now known as schwuk_away [09:41] ogra_: on mako its busted according to dashboard. can you try that? [09:41] yeah, grouper being broken with the current implementation is known [09:42] asac, if you send me a mako [09:42] ogra_: buy one [09:42] expense [09:42] * ogra_ has no mako and never touched a manta [09:42] (well, i have a mako as phone, but not a company one and i wont use it for company stuff) [09:47] Hi all! === schwuk_away is now known as schwuk [09:55] Wellark: ping === Ursinha-afk is now known as Ursinha [10:31] pete-woods: ping [10:41] hello [10:41] oSoMoN: hi! [10:42] given you pinged both Wellark and I, I'm guessing that HUD is broken? [10:42] pete-woods: heh :) nothing’s broken (that I’m aware of), but I do have a question about the HUD though [10:43] oSoMoN: okay, what's the question? [10:44] stgraber, ping: How will hardware-specific tarballs/partitions merge with the main ubuntu image? [10:44] pete-woods: I’m porting the HUD code in the browser app to the new unity actions API, and I was wondering about the applicationIdentifier property of the HUD object, has that disappeared, and can the related code be safely removed? [10:46] Does the Ubuntu Touch have the ability to attach to a monitor and run Ubuntu like the Edge? === MacSlow is now known as MacSlow|lunch [10:47] Person, not yet [10:47] oSoMoN: I'm not actually familiar with the old hud-client API, but I'm pretty confident that the unity action API uses the APP_ID environment variable [10:47] okay [10:47] thanks [10:48] pete-woods: ok, so it should be safe, right? [10:48] john-mcaleely: the upgrader will simply apply multiple tarballs one on top of the other, first one is the common rootfs, then hardware specific bits, then customizations. So everything is on the same system partition [10:48] oSoMoN: if the old code was just there to feed the application ID into the API, then yes, I think so [10:49] pete-woods: cool, I love removing code [10:49] oSoMoN: :D [10:50] stgraber, ok. makes sense for the hardware specific bundle. I thought customisations were a separate partition though? [10:50] john-mcaleely: nope, they're just a separate reserved path on the system partition [10:50] john-mcaleely: adding partitions is pretty tricky on the current devices, so we try not to do that [10:51] stgraber, oh, I see [10:58] asac, new pending image is up [10:59] * ogra_ glares at his chroot ... [10:59] Get:1 http://192.168.2.125:9999/ubuntu-ports/ saucy/multiverse android all 0.20130723-0ubuntu9 [464 MB] [11:00] xnox, i think we should produce one deb per subarch :) [11:00] seb128, Laney: for Online Accounts I need to implement a way of launching it as a stand-alone application; are you aware of similar requirement for other System Settings plugins? [11:01] seb128, Laney: my idea is to create a simple container to launch individual System Setting plugins [11:02] mardy: No I don't know that that's been asked for [11:02] mardy: We already have system-settings plugin-name to go staight to a plugin in the full s-s app === rickspencer3_ is now known as rickspencer3 [11:03] seb128, Laney: like, for example: "system-settings-launcher " [11:03] Laney: ah!! [11:03] I wouldn't bother unless someone comes up with a reason to need it to be more minimal [11:07] mardy, why do you need as a standalone app? is opening the settings app on the panel, as Laney said, good enough? [11:08] seb128: third party apps need to open the Online Accounts panel [11:08] seb128, Laney: but we also need to be able to pass some option [11:09] seb128, Laney: for instance to say "open Online Accounts for creating a Google account" [11:09] mardy, well, we have the infra to call a panel, we can probably add arguments after that ... it's going to be useful for other panels as well I'm sure [11:09] and when the panel is closed we should go back to the caller application, not to the System Settings [11:09] how do you close it? [11:10] (since the devices don't have a back button) [11:10] do you use back in the toolbar? [11:10] seb128: let me check the design [11:10] other topic: does anyone know if polkit is working on the touch image? [11:11] cjwatson, ^ (I think I read you mention needing it for click) [11:13] seb128: https://docs.google.com/a/canonical.com/document/d/1UwAQTXgEyZSD3di6fAUS0W18rKxh8TXb1TwsmkgbGG0/edit#heading=h.41rrn69y8201 [11:14] seb128: looks like there's a cancel button, and I presume that once the account is successfully created, we automatically go back to the app [11:14] mardy, I guess we just need a way to close the settings app with an action or something then [11:14] can you .exit() ? ;-) [11:15] seb128: I was told that it would work better once we switched to lightdm so that the session's registered with logind [11:15] Laney, ^ [11:15] loginctl shows me the session already is registered [11:15] cjwatson, so it doesn't work atm? (we are trying to figure out why we can't change the timezone through systemd) [11:15] but maybe it is missing something [11:16] do we have anything using polkit successfully on touch? [11:16] it doesn't have an active seat for example [11:16] I guess let's just wait for that [11:16] seb128: It works in terms of being able to handle AuthAny, but it doesn't know that the session is active [11:16] we know it works in principle [11:16] yeah, that's what I'm seeing [11:19] cjwatson, Laney: ok, thanks [11:19] Laney, yeah, let's assume it's working from our side and test again when lightdm start being used [11:20] cjwatson, do you have any idea when the switch to lightdm is supposed to happen? [11:20] didrocks, sil2100 can we push https://code.launchpad.net/~phablet-team/platform-api/trunk through? === chihchun is now known as chihchun_afk === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha [11:32] Saviq: platform-api from this daily release is in distro, right? [11:33] didrocks, yes [11:33] you need rev 98? [11:35] seb128: Afraid not [11:35] Only know as much as I told you :) I think I heard that from lool [11:36] cjwatson, ok, thanks, I'll check with mterry when he gets online [11:39] slangasek: hi steve, the image is already a daily thing [11:39] kgunn: slangasek and it updates once a day from the Mir staging ppa [11:40] seb128, ^^^ this image has lightdm too [11:41] ogra_, "this image"? the one with Mir? [11:41] seb128, right [11:41] Saviq: I guess we can re-roll the release to have that change [11:41] ogra_, do you know if it works? [11:41] not to my knowledge [11:41] ogra_, and where I can get it? [11:41] Saviq: is it 100% needed? Rev 98? [11:41] it might be installed, but its not the used greeter [11:41] oh ? [11:41] sil2100, yes [11:41] didrocks: can I re-roll platform? [11:41] ogra_: nop, lightdm work is deprioritized right now [11:41] ricmm, i thought that was a prereq for Mir inclusion [11:42] we had to work around it and go direct [11:42] didrocks, yes, sorry, missed your ping [11:42] seb128, sorry then ... seems i was wrong [11:42] as robert was held up with xmir fixes and such [11:42] sil2100: if it's what saviq needs, fine with me :) [11:42] ricmm, hum [11:42] ricmm, do you plan to fix logind session registration in another way? [11:42] ricmm, lightdm needs to be a high prio ... we need the pam/logind registration [11:43] ricmm, not having a correct session is breaking auth through polkit, which breaks system setting and other things [11:43] thats actually urgent to get rid of the NM hacks [11:43] and for getting rid if the PPA [11:43] ricmm, who depriorized it ? [11:43] the circumstances [11:43] I'll touch base with robert today [11:43] great [11:44] thats really high prio ... lets not drop it :) [11:44] its supposed to be relatively simple, 3-4 days work [11:44] Saviq, ricmm: it doesn't break any API/ABI's? [11:44] sil2100: what doesnt? [11:45] and no, theres no API breakage from latest rev of platform-api [11:45] sil2100, no [11:45] ricmm: platform-api ;) Thanks === MacSlow|lunch is now known as MacSlow === om26er is now known as om26er|afk [12:00] ogra_, 0725 seems to reboot over and over :/ [12:01] Saviq, just pulling it, i'll check [12:01] Saviq, what devi.ce ? [12:01] ogra_, manta [12:01] * ogra_ only has grouper and maguro [12:02] Saviq, meanwhile boot into recovery: adn shell cat /proc/last_kmsg | pastebinit [12:02] *adb [12:03] ogra_, not much... http://paste.ubuntu.com/5910914/ [12:03] (though manta has a history of currpting its ramconsole) [12:04] Saviq, try again until you get a proper log ... thats a mako specific issue [12:04] one out of ten boots it shoudl work [12:05] err [12:05] manta specific issue indeed [12:05] Saviq: ping [12:05] ricmm, pong [12:06] Saviq: o/ did you guys fix the unity-mir API problem? [12:06] I updated with the latest from the PPA and unity8 still segfaults [12:07] ricmm, Gerry was on it, but seems to have only tried to bump unity-mir version [12:07] ricmm, if it is an API issue like Gerry suspects, we should probably rebuild unity8, too [12:08] barry: you broke phone updates again ;) [12:08] Saviq: ok, ill re-run my recipe [12:08] ogra_, got it http://paste.ubuntu.com/5910932/ [12:09] barry: though that should have been the last time, I just need to get rid off the code in lxc-android-config and the initramfs-tools package now and we should be good starting tomorrow [12:09] Saviq: ah, it never build [12:09] its a colo branch [12:09] hmmm [12:09] ogra_: do you remember how we fix colocated branches? for recipe building [12:09] Saviq, i bame the kernel [12:09] *blame too [12:09] ogra_, yeah, looks like it [12:09] * Saviq too! [12:10] Saviq: the platform stack failed one test, but I'll anyway try to publish once the build step finishes [12:10] ricmm, nope, sorry ... thats sergiuiens speciality [12:12] ogra_: fixed, just bzr upgrade [12:12] Saviq: should build fine now, will retry [12:12] :) [12:12] none of the builds had actually made it through before [12:13] ricmm, :/ === _salem is now known as salem_ === chihchun_afk is now known as chihchun [12:19] tsdgeos, do you know where are the copies of the gvariant<>qvariant? [12:20] tsdgeos, besides dee-qt and the unity plugin === chihchun is now known as chihchun_afk === jhodapp is now known as jhodapp|bbiab [12:28] greyback: hi! [12:29] ricmm: hey [12:29] ricmm: so, what's broken? :) [12:29] well the unity8 branch was in a format that doesnt work with recipes [12:29] so it never built [12:29] I just pushed the upgraded one [12:31] ricmm: what format does work, for my future knowledge? [12:32] oh god [12:32] wait, doing the migration changed the target of the recipe and it built against the old one again [12:32] I think we hit the daily quota for building one sec :p [12:35] Hello All [12:37] When I try to use Breakfast command "breakfast yuga" I get [12:37] build/core/product_config.mk:239: *** _nic.PRODUCTS.[[device/sony/yuga/cm.mk]]: "vendor/sony/qcom-common/qcom-common-vendor.mk" does not exist. Stop. ** Don't have a product spec for: 'cm_yuga' ** Do you have the right repo manifest? [12:38] rsalveti, hey [12:52] Saviq, if you need this image to work, try if making /data/ubuntu/usr/lib/lxc-android-config/update-fstab executable and rebooting fixes it [12:53] ogra_, trying [12:54] (teh kernel should still not fall over just because fstab is missing) [12:54] Hi Guys I get this build/core/product_config.mk:239: *** _nic.PRODUCTS.[[device/sony/yuga/cm.mk]]: "vendor/sony/qcom-common/qcom-common-vendor.mk" does not exist. Stop. ** Don't have a product spec for: 'cm_yuga' ** Do you have the right repo manifest? [12:54] when i try to breakfast [12:54] why do i get it?>? [12:54] ogra_, +1 [12:55] Geniuswork, did you run repo sync after adding it to the xml file ? [12:55] Saviq, +1 for my comment or because the image works ? [12:55] :) [12:55] ogra_, yes [12:55] lol [12:55] ogra_, no media on manta still, though [12:55] yeah [12:56] ogra_: how do i see that binfmt works correctly? [12:56] chroot tells me that /bin/bash is not tehre [12:56] ooof [12:56] , but it surely is [12:56] * ogra_ didnt have to use the binfmt stuff in years ... [12:56] QEMU_LD_PREFIX = /etc/qemu-binfmt/arm [12:56] first look if the qemu handler is inside your chroot [12:56] that doesnt exist [12:57] that /etc file/dir [12:57] before blaming the kernel :) [12:57] is qeu-user-static installed ? [12:57] ogra_: how do i know? [12:57] ogra_: yeah [12:57] i have qemu-arm-static [12:57] the binary ? [12:57] oh i can copy that in [12:57] right [12:57] let me try [12:57] that needs to live somewhere inside the chroot [12:57] to /sbin? [12:58] /usr/sbin or /usr/bin [12:58] i *think* it was the latter [12:58] * ogra_ is so spoiled with his chromebook and USB 3.0 disk :) [12:58] yep worked :) [12:58] nice [12:58] i dont have to use all these weirdo workaround [12:58] s [12:58] get an edge ! [13:00] * tvoss mutters pledge for an edge [13:00] eh it all doesnt work :) [13:00] ping [13:00] mhr3: the thinkg larsu did is the unity plugin, no? [13:00] even though i mounted /proc [13:01] jibel, YOU ROCK ! [13:01] stgraber: apparently my tablet refuses to hold a charge. i was going to test it this morning [13:01] tsdgeos, no [13:01] asac, did you mount /sys too ? and did you do it inside the chroot ? [13:03] mhr3: qmenumodel has another one [13:03] QVariant Converter::toQVariant(GVariant *value) [13:03] tsdgeos, yea, right, lars told me already thx [13:04] asac, note that name resolution will indeed not work unless you copy the respective files from the host [13:04] pinging IPs should work if /proc and /sys are mounted though [13:04] mhr3: so yeah dee-qt, unity and qmenumodel [13:05] i think that's the 3 [13:06] ogra_: no sys [13:06] but i could ping wihtout that in the past [13:06] ogra_: can i mount sysfs or do i need to bindmount? [13:06] mount -t sysfs sys /sys [13:06] when chrooted [13:06] mhr3: wait and gsettings-qt too [13:06] ogra_, :) ty [13:07] mhr3: so dee-qt, unity, qmenumodel and gsettings-qt [13:07] hello! any chance to get ubuntu touch to install onto an YP-G70 device? [13:07] cjwatson, https://code.launchpad.net/~saviq/qtubuntu/respect-path/+merge/176904 [13:07] cjwatson, https://code.launchpad.net/~saviq/qtubuntu/more-robust-desktop-parsing/+merge/176931 [13:08] daveid, if it isnt on the device page you might need to port it yourself [13:08] !devices |daveid [13:08] daveid: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices [13:08] the link to the porting guide is in the channel topic [13:13] ogra_: thanks! [13:15] Guys Could Please someone help me?? [13:15] Geniuswork, did you see my answer above ? [13:15] Wait [13:17] which one? [13:17] the link?? [13:18] Geniuswork, did you run repo sync after adding it to the xml file ? [13:18] i used breakfast [13:18] didn't manually make ot [13:18] well, it seemingly didnt pull or didnt find the tree [13:19] So Can I use The CM10.1 precompiled Zip for that?? [13:19] no [13:19] Is there a simulator of some sort that can be used to test ubuntu touch with intended screen resolution on a PC? [13:19] Then How?? [13:20] Debolaz, nope, but it is desired that we have one at some point .... patches gracefully accepted ;) [13:20] Geniuswork, make sure the tree gets pulled ... if automated (breakfast) doesnt work i'd go step by step through the manual variant and see if they worked out [13:21]