**** BEGIN LOGGING AT Wed Sep 27 03:00:01 2017 Sep 27 05:54:47 Morning! Sep 27 05:54:50 morphis: ping Sep 27 06:42:21 Morning! Sep 27 07:19:05 Tofe: morning! Sep 27 07:30:06 Morning! Sep 27 07:30:10 Tofe: Any clues for https://bpaste.net/show/6970d875ea3e ? Sep 27 07:30:14 MartinHov: Morning! Sep 27 07:30:30 I couldn't find much with Google :P Sep 27 07:32:17 Herrie|Work: pong Sep 27 07:33:21 morphis: Did this 3.4 kernel for Mako with the 4.2 BT backports ever make it? Sep 27 07:33:30 I saw it mentioned but I couldn't find the source with my search Sep 27 07:35:37 Herrie|Work: the patchset I did once ago should apply to that kernel but I think I never applied it for mako Sep 27 07:36:13 morphis: OK, so we could take the Hammerhead backports and try those on Mako Sep 27 07:36:35 https://wiki.ubuntu.com/Touch/BluetoothEnablement Sep 27 07:36:42 yeah Sep 27 07:37:08 https://code.launchpad.net/~bluetooth/+git/linux-bluetooth/+ref/bluetooth-3.4 has all the code which should apply on a 3.4 kernel Sep 27 07:38:26 morphis: OK we'll give that a try then Sep 27 07:44:59 morphis: that's the same thing as https://github.com/ubuntu-phonedations/backports right? Sep 27 07:48:44 Herrie|Work: it looks like your build error is due to an incomplete .c; where's your source code for this kernel ? Sep 27 07:49:35 Herrie|Work: the function *should* have been there: http://elixir.free-electrons.com/linux/v4.5/source/net/netfilter/xt_RATEEST.c#L62 but maybe this file was only partially backported Sep 27 07:51:07 Herrie|Work: just disable NETFILTER_XT_TARGET_RATEEST in defconfig Sep 27 07:55:41 Tofe: It's probably my Windows GH client :( Sep 27 07:55:53 When cloning large repo's like kernel it has issues :S Sep 27 07:56:22 I simply commited it as a single commit: https://github.com/Herrie82/android_kernel_lge_hammerhead/commit/dafcf6f192250ddb0ce9a3ad7b1fed4a378eeba7 Sep 27 07:56:28 I guess I would need to revert that one Sep 27 07:57:08 Tofe: I.e. I do a clean clone in GH Windows client and right away it tells me these 12 files have changes while I didn't do anything... Sep 27 07:57:40 Tried cloning multiple times... So I guess it's a bug in their client Sep 27 07:58:27 maybe, that's quite annoying yes Sep 27 07:58:49 OK testing it with that commit reverted Sep 27 07:58:59 BEcause disabling it in defconfig didn't work Sep 27 07:59:44 ok Sep 27 08:00:10 morphis: Thnx will give that a try Sep 27 08:00:29 Why are these backported patches needed exactly? Wasn't really clear to me, since BlueZ5 requires 3.4 kernel Sep 27 08:00:36 Or lots of bugs are fixed with these backports? Sep 27 08:05:24 Tofe: Kernel itself seems to build now :P Sep 27 08:05:31 Onto rest of image Sep 27 08:09:20 good! Sep 27 10:28:19 Tofe: The PID namespace commit was required for systemd right? Sep 27 10:28:42 bshah: ping Sep 27 11:12:29 Herrie|Pre3: right; could be it was for LXC too, I don't remember Sep 27 11:12:46 maybe even for chromium sandboxing :p Sep 27 11:18:42 Tofe: Ok well doesn't matter much anyway Sep 27 11:18:52 Shouldn't hurt upstream Halium right? Sep 27 11:26:25 Well the PID namespace has to be present in the kernel of course, but if it is then it's harmless Sep 27 12:18:29 Tofe: Seems my kernel clone issue is caused by API rate limit @ GH :P Sep 27 12:32:48 hehe Sep 27 13:46:48 Tofe: Hmmz seems builder @ home shut itself off again :s I suspect overheating... Will need to add some more paste or increase fan speed probably :P Sep 27 13:47:05 Now have ultra-silent cooler, might be too silent :P Sep 27 14:00:22 Herrie|Pre3: well, if it's off, it sure is silent ! Sep 27 14:00:35 True :P Sep 27 14:00:44 But it turns off too often under heavy load Sep 27 14:01:23 So I need to check bios settings if there's a too low temperature, increase fan speed or check if I have enough paste... Or upgrade to Threadripper :P Sep 27 14:14:42 Herrie|Pre3: what is you CPU's TDP ? Sep 27 14:15:07 95W Sep 27 14:15:23 ok, not that much Sep 27 14:15:52 mine is 65W, that was on purpose :) Sep 27 14:16:24 95 is max my miniITX board can handle :P Sep 27 14:17:40 Herrie|Pre3: For the paste, it's not necessarily that there isn't enough, it could be it's not properly distributed on the surface Sep 27 14:18:04 Yeah will first check bios settings just to be sure Sep 27 14:18:19 Could be I was too conservative in terms of temp before shut off :P Sep 27 14:18:24 Or fan speed too low Sep 27 14:18:43 Adjusted it when relatives were around, now nobody in that room :P Sep 27 14:19:00 could be that also, yes Sep 27 14:58:39 bshah: You around? Sep 27 14:58:56 Herrie|Pre3: yep.. sorry was travelling at morning Sep 27 14:59:03 JaMa: Any reason for the unstable build to take 6 hrs each :P ? Sep 27 15:00:30 Herrie|Pre3: rebuilding from scratch do to change in pseudo Sep 27 15:00:32 bshah: I noticed N4/Mako on your devices page for Halium but couldn't find where you're taking kernel from Sep 27 15:00:37 Any pointers? Sep 27 15:00:43 I've dropped rpi2 build, so the remaining builds should be faster Sep 27 15:00:45 Herrie|Pre3: ubports Sep 27 15:00:49 JaMa: OK Sep 27 15:00:53 bshah: OK Sep 27 15:01:04 Herrie|Pre3: and in addition I believe someone's local patch Sep 27 15:01:10 need to find who was porting to it Sep 27 15:01:21 we really need a way to track ports Sep 27 15:01:24 :/ Sep 27 15:02:05 bshah: I'm soon ready to propose a PR for GCC 5/6/7 for Hammerhead I guess to Halium kernel repo Sep 27 15:02:13 \o/ Sep 27 15:02:20 Just would need to be tested with your GCC 4.x toolchain Sep 27 15:02:33 Not sure someone @ Halium end could do this? Sep 27 15:03:54 Herrie|Pre3: I can test once you PR Sep 27 15:03:59 (easy to pull changes) Sep 27 15:04:21 bshah: Thnx Sep 27 15:04:36 Need to finish my build when home then cleanup commits a bit and will PR Sep 27 15:13:07 "Discussion about announced webOS-Ports projects only!" so... where do I discuss regular dev/porting talk? Sep 27 15:16:33 You can do that here too :P Sep 27 15:23:39 aah cool. I'm planning on porting the webOS UI to postmarketOS (https://postmarketos.org) as requested by someone in our community Sep 27 15:24:23 however, so far I find the building documentation all a bit vague. do I understand it right that parts of the LunaUI are on https://github.com/openwebos and other parts on https://github.com/webos-ports? Sep 27 15:27:23 PureTryOut[m]: do you mean the compositor? something else? Sep 27 15:29:27 our UI looks very much like openwebos, but we reimplemented a large part using more recent Qt Sep 27 15:31:07 Tofe: everything basically. I'm entirely new to webOS (besides having heard the name somewhere before) Sep 27 15:31:19 where can I find the compositor? what parts are required to get a working UI/shell? Sep 27 15:32:53 PureTryOut[m]: what build system is postmarketOS using? Sep 27 15:35:29 well, it uses it's own package manager "apk". it's build files are really close to Arch Linux build files, APKBUILD, which are basically just Shell scripts Sep 27 15:35:52 each package gets build separately using the regular build methods, `cmake`, `make`, etc Sep 27 15:36:18 do you already have packages for qt5? Sep 27 15:36:42 PureTryOut[m]: well, our compositor uses QtWayland, is named LunaNext and can be found on our github, next to all wr have Sep 27 15:36:59 I know how to package and build stuff, I just like some list of required components of the webOS UI to get a working shell Sep 27 15:37:20 JaMa: most of them y es Sep 27 15:37:22 *yes Sep 27 15:37:41 most Qt5 packages are already packaged, and I've packaged QtWayland myself for Plasma Mobile before Sep 27 15:37:55 so is the compositor this one? https://github.com/webOS-ports/luna-next Sep 27 15:38:09 as it fails building immediately due to CMake not being able to find a "webos" directory in the root of the source Sep 27 15:38:49 PureTryOut[m]: but be aware you might need some of infrastructure stuff too, like LS2... we never tried to run luna-next on another os Sep 27 15:39:18 PureTryOut[m]: yes Sep 27 15:39:18 you'll need the cardshell too Sep 27 15:39:24 well it's about time then! 😄 Sep 27 15:39:41 and there are some modifications e.g. for qtwebengine which you'll need to take as well Sep 27 15:39:58 which is the big qml description of our UI Sep 27 15:40:10 not sure how that would play with other UIs which might want to use the same qtwebengine package Sep 27 15:40:13 JaMa: Does he? Sep 27 15:40:27 so, why does CMakeLists.txt at line 8 (in https://github.com/webOS-ports/luna-next) have an entry which makes it look for a directory which not exists then? Sep 27 15:40:40 If he doesn't use browser or Enyo apps he shouldn't require our webengine mods? Sep 27 15:40:50 but if you're lucky enough, you'll get a working wayland compositor... Who knows :) Sep 27 15:41:38 PureTryOut[m]: https://github.com/openwebos/cmake-modules-webos Sep 27 15:41:39 PureTryOut Sep 27 15:43:44 I think he'll be fine with a regular Qt 5.9 Sep 27 15:44:12 But the apps won't be 'luna-next' aware Sep 27 15:45:13 sure it depends on the end-goal, I doubt a bit that people were asking just for compositor itself :) Sep 27 15:45:34 JaMa: ooh thanks! Sep 27 15:46:05 ah, well, in that case, they should install the whole os :) Sep 27 15:46:36 should I use submissions/19 or 1.0.0-rc6 of cmake-modules-webos? Sep 27 15:46:44 Tofe: I see some advantages of pmOS for the legacy webOS devices too :) Sep 27 15:48:50 PureTryOut[m]: I don't remember, but master should be fine Sep 27 15:49:45 Herrie|Pre3: in the end pmOS should be an advantage for *every* Linux distro, as the main goal is to get as many devices mainlined as possible. this should make it possible to install any distro in the end Sep 27 15:54:29 PureTryOut[m]: I think I understand the project, but one difficulty is that mobile "glibc/linux" distro are more deeply integrated than on desktop Sep 27 15:56:10 e.g. we hacked qtwebengine to extend it, etc Sep 27 15:57:52 but still, it'd be interesting to know how far this can go Sep 27 15:58:09 not sure what you mean with more deeply integrated Sep 27 16:01:53 well for instance we expect the apps to negociate their startup in a certain way Sep 27 16:02:42 I'm on an unstable connection, be back in a little hour Sep 27 16:07:29 "Qt5Compositor", is that Qt5Wayland or is it it's own package? Sep 27 16:28:22 PureTryOut[m]: that's part of QtWayland Sep 27 16:28:57 yeah I just found it. I was checking the build of qt5-wayland, and according to qmake Qt Wayland Compositor is compiled in Sep 27 16:29:00 * PureTryOut[m] sent a long message: PureTryOut[m]_2017-09-27_16:28:59.txt Sep 27 16:29:12 however, CMake can't find it Sep 27 16:29:24 (this is while trying to compile `luna-next`) Sep 27 16:32:40 well, we don't any tricks here, we use standard cmake dependencies Sep 27 16:33:21 hmm Sep 27 16:33:51 let me just check if we do patch qtwayland to fix something Sep 27 16:35:43 btw in your CMakeLists.txt, the if statements just after the "find_package()" bits never get called, as CMake already breaks on not being able to find the package Sep 27 16:36:22 ah, good remark, that'll simplify the file Sep 27 16:37:11 so you have usr/lib/cmake/Qt5WaylandCompositor/* files ? Sep 27 16:38:13 You are using Qt 5.9, right ? Sep 27 16:38:41 because when we switched from Qt 5.6 to Qt 5.8, we needed to update the QtWayland dependencies: https://github.com/webOS-ports/luna-next/commit/81b8b853c1450948dd15ed96794f509d260c3f47 Sep 27 16:39:01 (among many other details) Sep 27 16:40:00 yes Qt 5.9 Sep 27 16:40:05 I'll check for those files hold on Sep 27 16:43:02 * PureTryOut[m] sent a long message: PureTryOut[m]_2017-09-27_16:43:01.txt Sep 27 16:43:05 yup, they're there ^ Sep 27 16:43:47 however, CMake complains about "Qt5Compositor", not "Qt5WaylandCompositor" Sep 27 16:44:14 but we never ask about it for luna-next Sep 27 16:45:26 yeah I noticed Sep 27 16:45:40 does CMake point to the origin of the error? Sep 27 16:46:13 ooh nvm I get it Sep 27 16:46:20 I downloaded the latest release of Luna-next, which was in 2014 Sep 27 16:46:25 I need the latest from Git Sep 27 16:46:38 hmm, isn't it time for a new release then? 😉 Sep 27 16:46:45 yes, always take from Github Sep 27 16:47:41 ah, we switched to a sort-of rolling release, we have a stable release about each month Sep 27 16:47:58 well I downloaded from Github, but that release is way too old 😛 Sep 27 16:48:05 don't you guys tag your releases anymore? Sep 27 16:48:37 We tag our build system, not the sources Sep 27 16:49:14 the build system then has a precise set of recipes, which in turn point to precise commits Sep 27 16:50:36 ooh shit... does Luna have a hard dep on systemd? Sep 27 16:50:59 about the releases: hmm too bad. tagging the source is easier for other distros (obviously) Sep 27 16:51:28 PureTryOut[m]: there's just a .service file; the dependency in the CMakeList is a bit too much I guess Sep 27 16:51:53 quite. can I just patch it as optional for now? Sep 27 16:52:31 patch as much as you want :) We'll see a bit later what cleaning up we'll do Sep 27 16:54:16 But the hard stuff will be Luna*... here, you'll get an example of "deeper integrated than usual" Sep 27 17:05:05 just now got the first Luna dep. wish me good luck 😆 Sep 27 17:07:31 :) Sep 27 17:23:00 the most interesting part will be if it all works with musl, rather than glibc Sep 27 17:25:08 what are those "submissions/" releases of openwebos projects? Sep 27 17:25:21 are they like betas? Sep 27 17:28:37 eeuhm... this is interesting. "libpnbjson" depends on "pmloglib" which depends on "libpnbjson" again... Sep 27 17:30:18 either that or "pbnjson_c" != "libpnbjson" Sep 27 17:34:59 for openwebos, I quite don't know, it might be something used by HP in the past Sep 27 17:35:27 and for libpnbjson, I don't know either :) not much help there, eh Sep 27 17:36:20 lol Sep 27 17:36:56 I can only find a pbnjson.h file in libpbnjson, which led me to believe libpbnjson provides pbnjson_c Sep 27 17:37:08 but since that causes a reverse dependency, I guess not Sep 27 17:37:12 where can I find webos-ports build scripts? Sep 27 17:38:33 PureTryOut[m]: https://github.com/webos-ports/webos-ports-setup Sep 27 17:38:41 ^ (we use OpenEmbedded) Sep 27 17:39:01 this checkouts used metadata layers, and the rest is OE parsing the metadata and building based on that Sep 27 17:39:29 our next release is based on the "pyro" branch Sep 27 17:40:31 are there build scripts for each individual package? Sep 27 17:40:38 I just want to find which package provides pbnjson_c Sep 27 17:41:37 https://github.com/webOS-ports/meta-webos-ports/blob/master/meta-luneui/recipes-webos/libpbnjson/libpbnjson.bb Sep 27 17:41:53 aah found it, webos-ports/meta-webos-ports Sep 27 17:43:48 hmm Sep 27 17:44:08 CMake can't find "PmLogLib-headers" when compiling libpbnjson, however you guys do not have a dependency even close to it in the build script Sep 27 17:47:56 mmh right, good question Sep 27 17:49:15 PureTryOut[m]: should be provided by https://github.com/openwebos/pmloglib Sep 27 17:49:30 ha, like I said, that causes a reverse dependency Sep 27 17:49:38 as pmloglib depends on libpbnjson Sep 27 17:50:48 yup, there's a dependency issue, but not if you have NO_LOGGING to FALSE, which is the default Sep 27 17:51:06 which is done where? Sep 27 17:51:24 https://github.com/webOS-ports/meta-webos-ports/blob/master/meta-luneui/recipes-webos/libpbnjson/libpbnjson.bb Sep 27 17:51:25 ah no sorry, if it's TRUE, like here https://github.com/webOS-ports/meta-webos-ports/blob/pyro/meta-luneui/recipes-webos/libpbnjson/libpbnjson.bb#L18 Sep 27 17:51:50 ooh just a CMake variable Sep 27 17:52:31 JaMa: he has a point tough :p this logging dependency shouldn't exist, not like this Sep 27 17:53:01 similarly like this dependency? :) https://github.com/webOS-ports/meta-webos-ports/blob/master/meta-luneui/recipes-webos/pmloglib/pmloglib-private.bb Sep 27 17:53:11 well that fixed it. now I can't find `lemon-native` Sep 27 17:53:58 PureTryOut[m]: ./meta-oe/recipes-devtools/lemon/lemon_3.7.3.bb Sep 27 17:54:05 in meta-openembedded repository Sep 27 17:54:19 PureTryOut[m]: the "-native" suffix means it will be compiled for the host arch, not the target one Sep 27 17:54:42 probably because it provides a build tool or such Sep 27 17:55:38 JaMa: I see no such directory https://github.com/openwebos/meta-oe/tree/master/recipes-devtools Sep 27 17:55:53 wrong repository Sep 27 17:56:24 I figured Sep 27 17:56:27 https://github.com/webOS-ports/webos-ports-setup/blob/pyro/conf/layers.txt#L3 Sep 27 17:56:47 found it Sep 27 17:57:00 https://github.com/openembedded/meta-openembedded/tree/master/meta-oe/recipes-devtools/lemon Sep 27 17:57:36 interesting, it's part of sqlite? Sep 27 17:57:45 yes Sep 27 18:02:31 PureTryOut[m]: on what arch are you building this ? Sep 27 18:03:32 x86_64 for now Sep 27 18:03:39 eventually it will be compiled for armhf and aarch64 as well Sep 27 18:05:01 ok Sep 27 18:05:17 x64 and armv7 should work Sep 27 18:05:33 Herrie: we also have a qemux86_64 working, isn't it ? Sep 27 18:05:37 x86 too Sep 27 18:05:46 ah sorry misread Sep 27 18:05:48 Tofe: UI yes Sep 27 18:06:22 Just hw kbd not Sep 27 18:06:31 oh, ok Sep 27 18:06:33 I forgot Sep 27 18:07:37 Could be something smalll really Sep 27 18:07:52 For the rest looked ok-ish Sep 27 18:07:57 ok good Sep 27 18:09:18 I recall something similar on x86 Sep 27 18:23:17 is http://jenkins.nas-admin.org down for you as well? Sep 27 18:24:38 novaldex|away: ^^ Sep 27 18:25:56 damn, a script in libpbnjson depends on Bash instead of regular sh. replacing the shebang just makes it fail on syntax Sep 27 18:26:06 not only jenkins.nas-admin.org it seems, I cannot connect to kwaj over vpn as well (bonaire works) Sep 27 18:26:10 "https://github.com/openwebos/libpbnjson/blob/master/src/pbnjson_c/validation/apply_lemon.sh#L37" Sep 27 18:26:15 unexpected "(" Sep 27 18:27:44 at least it has corrent shebang, I would be surprised if you don't find some other places where /bin/sh is expected to be bash Sep 27 18:29:38 yeah probably. I'll have a look this week if I can fix those bits Sep 27 18:51:18 JaMa: works for me now Sep 27 18:52:52 still down for me, but the build on bonaire is still running (still in qtwebengine) Sep 27 18:55:10 PureTryOut[m]: All our components are build with latest Yocto all the time, so that means modern glibc, systemd etc ;) Sep 27 18:55:26 New GCC (7) currently for example Sep 27 18:55:37 We're happy to accept patches like Tofe said Sep 27 18:55:53 And in terms of tagging it was quite some work and we're a small team with very limited resources Sep 27 18:56:05 But it's something we might be able to pick up or automate somehow Sep 27 19:27:40 Tofe: OK connected the front fan of the case, removed bay with 2 large HDD's that were not used and blocking airflow, moved some cables around. PSU is old model, so not modular, so can't disconnect the cables :P Sep 27 19:27:47 Now will have some look in BIOS Sep 27 19:28:20 I see you've been playing a bit :p Sep 27 19:29:22 Yeah LOL Sep 27 19:29:27 Better flow won't hurt Sep 27 19:30:19 Tofe: This one, so easy to take apart ;) http://www.tomshardware.co.uk/mini-itx-small-form-factor-pc-case,review-32111-4.html Sep 27 19:31:20 yes indeed Sep 27 20:51:40 Herrie: did you kill some of the unstable jobs? I cannot access http://jenkins.nas-admin.org but from the shell I see that now it's building luneos-testing and there should be rsync+cleanup from unstable before testing was started Sep 27 20:52:08 Herrie: if you don't run rsync and cleanup soon for unstable, you'll run out of memory, because unstable tmpfs already takes some 20G Sep 27 20:55:34 heh, thanks to opera integrated vpn I can access http://jenkins.nas-admin.org through there :) Sep 27 20:56:27 queued rsync+cleanup now Sep 27 20:57:56 JaMa: No I didn't kill anything Sep 27 20:58:00 I did kick off a testing Sep 27 20:58:15 But I would expect it to finish the whole unstable chain before starting with testing? Sep 27 20:58:25 Or that's a wrong assumption? Sep 27 20:58:25 ok, strange Sep 27 20:58:40 Because I see the unstable rsync and cleanup scheduled now :S Sep 27 20:59:03 yes, it would finish the queue, before adding new entries to queue from the pipeline Sep 27 20:59:23 I've added them just a minute ago Sep 27 20:59:46 in this case all unstable jobs were added manually without using the pipeline Sep 27 20:59:53 because I didn't want to build them all Sep 27 21:00:11 JaMa: Ah OK Sep 27 21:00:27 but when 2 pipelines are triggered at the same time, they will take turns adding new item to the queue Sep 27 21:00:35 Ah ok Sep 27 21:00:43 JaMa: How much tmpfs we have in general? Sep 27 21:00:49 Sorry for late reply, toying with builder Sep 27 21:00:54 To see why it shuts down all the time Sep 27 21:01:13 which sometimes is unwanted behavior (unlike old workflows which were blocking the queue until they were completely finished) Sep 27 21:01:30 I mean my local builder Sep 27 21:02:18 Herrie|Laptop: each (stable, unstable, testing) has tmp-glibc mount with 80G of tmpfs Sep 27 21:02:35 but there is only 98G of ram total on that builder Sep 27 21:02:45 Seems it's now rsyncing Sep 27 21:02:50 Your unstable Sep 27 21:03:04 good, I see it as well in opera Sep 27 21:03:46 Weird that you have issues connecting Sep 27 21:03:49 With me it works fine Sep 27 21:03:53 total used free shared buff/cache available Sep 27 21:03:55 And I'm on UPC too, though in NL Sep 27 21:03:57 Mem: 98 0 2 35 94 61 Sep 27 21:04:00 Swap: 1 0 1 Sep 27 21:04:36 it's probably some routing issue from CR affecting only some routes, so the VPN to work and VPN provided by opera works Sep 27 21:04:46 Luckily I no longer use UPC Sep 27 21:05:06 but I've seen some article now that someone dig through UPC optical cable where I live Sep 27 21:05:30 so maybe my provider was using trafic from UPC and we were affected as well Sep 27 21:06:01 Ah OK Sep 27 21:06:06 I can't complain here Sep 27 21:06:54 I'm getting 272 Mbps downstream on my laptop via WiFi :P Sep 27 21:07:06 And 28 Mbps upstream ;) Sep 27 21:07:10 Not bad on 300/30 ;) Sep 27 21:07:15 funny that the article doesn't show anywhere what day and time it was issued, so I cannot say if it's from few hours ago (when the issues started here) or some very old one just found by google search :) Sep 27 21:07:28 nice :) Sep 27 21:07:47 Desktop has more proper wifi card so should get better speeds Sep 27 21:08:00 And this is via good old coax with some Docsis ;) Sep 27 21:08:08 v3 I guess Sep 27 21:11:31 I have 100/100Mbps, currently giving me 93,9 down, 57,6 up, with 7.0 ms ping Sep 27 21:11:58 I never got that close to bought 400Mbps from UPC in old flat Sep 27 21:12:26 and because they screwed up moving the services when I was moving to new flat I never want to have contract with them Sep 27 21:14:57 Can imagine... I had 7 "engineers" coming and replacing boxes, putting amplifiers "free of charge" over the years till I got fed up and asked them to finally send a proper line engineer. He was inside for 2 mins, saw a dip in signal 10m from my apartment, got out to main box, was a corroded connector. He replaced it and since then everything flawless Sep 27 21:15:28 Alternative was 8Mbps/2Mbps ADSL :P Sep 27 21:15:31 So had little choice Sep 27 21:15:52 Trying to get fiber retrofitted in our building soon Sep 27 21:16:26 But the structure of our building complex means we need to run a lot of fiber ;) 4 bunches per tower instead of regular 1 in most building... So quite costly... Sep 27 21:16:43 https://www.a1m.cz/ has optical cable to our building and ethernet to each flat, so I had good option :) Sep 27 21:17:38 with UPC cable modem I never got under 20ms ping Sep 27 21:17:53 JaMa: Nice :) Sep 27 21:18:19 We have old antenna from back in 70's that we're not using, so trying to see if we can retrofit fiber in there Sep 27 21:18:28 But lots of people might have closed it off etc Sep 27 21:18:29 and with all the ajax everywhere I appreciate good ping more than big bandwidth nowadays Sep 27 21:18:31 So challenge ;) Sep 27 21:18:58 My ping is 7-15 usually Sep 27 21:19:04 So not that bad Sep 27 21:20:40 yes, that's good, for work VPN I have 250+ ms just to ping DNS Sep 27 21:20:56 LOL Sep 27 21:21:12 so I'm using static entries in /etc/hosts and sometimes clicking a lot in jenkins UI is quite painful Sep 27 21:21:59 you can abort 10 jobs before the queue is refreshed and then you can realize that your browser was clicking on some other jobs you didn't want to abort and so on :) Sep 27 21:24:24 Not good ;) Sep 27 21:24:41 You'd expect they could sort some better infra @ SVL :P Sep 27 21:25:51 or better webOS :) Sep 27 21:29:58 JaMa: That too Sep 27 21:30:09 Though I haven't played with it on a TV yet Sep 27 21:30:17 My Panasonic is still holding up pretty OK Sep 27 21:30:22 But it's 7 years old by now Sep 27 21:30:31 So might be getting EOL for Plasma soon :P Sep 27 21:30:33 my borrowed Samsung is also good enough for wife :) Sep 27 21:32:04 This one is "only" 42", so when it dies we'll probably get a nice LG webOS one that's a bit bigger :P Sep 27 21:32:05 but webOS on TVs is pretty good (especially compared with what LG used before) Sep 27 21:32:19 unfortunatelly very different from what was dumped as openwebos Sep 27 21:32:22 Well everything 2010 era was complete and utter garbage Sep 27 21:32:30 On TV's Sep 27 21:32:48 yes, that's what I bought to my father :) Sep 27 21:33:10 but it was big, so he is still happy with it Sep 27 21:33:17 It's a shame that LG doesn't contribute anything in the open anymore to webOS :( Sep 27 21:33:53 look at the 65inch LG Signature W (wallpaper) TVs, there is a lot of youtube unboxing videos (it almost looks like LG is giving them to youtubers for free) Sep 27 21:34:21 it's very impressive OLED, I wish I could explain why I would need one for development :) Sep 27 21:35:38 Bit outside of the budget I'd like to spend for a TV :P Sep 27 21:36:11 Ah it's LOT cheaper in Germany :P Sep 27 21:36:57 persuade your wife to became famous youtuber then :) Sep 27 21:36:57 Went from 7799 on Google shopping to 6100 on local site to 5700 EUR in Germany :P Sep 27 21:37:21 That's a quick 2000 earned for a minute of Googling LOL Sep 27 21:37:37 I've seen somewhere around 8K USD Sep 27 21:38:08 This one right? https://geizhals.eu/lg-electronics-signature-oled-65w7v-a1591880.html Sep 27 21:38:12 but until it drops for 1/10 price and gets to 75inch I'll stay with old samsung :) Sep 27 21:38:56 This plasma was just above 1000 EUR back in the day Sep 27 21:39:02 I think Sep 27 21:39:05 So that was OK Sep 27 21:39:12 yes, the same Sep 27 21:39:29 I would have a hard time spending over 3000 on a TV even when it's 65" :P Sep 27 21:39:44 Then again, maybe I get some nice bonus @ work :P Sep 27 21:39:44 same here Sep 27 21:40:10 but people buy iphones (or any new phones in general) for stupidly high prices as well Sep 27 21:40:28 We already have kinda 13 and 14th month and will probably get 30% annual bonus :P Sep 27 21:40:50 So I guess I could potentially stretch the budget a bit :P Sep 27 21:41:22 daugher kicked wife's phone yesterday, so we have Samsung J5 2016 with broken display (not the glass though) Sep 27 21:41:37 so I'm glad I didn't bought here something more expensive than that last year Sep 27 21:41:42 Yeah I'm happy with my good old HP phones :P Sep 27 21:41:51 Don't feel to sorry if 1 dies :P Sep 27 21:42:01 Got a bunch of USD 20 Veers :P Sep 27 21:42:16 and this wallpaper oled holding on wall just with magnets isn't really children-proof as well Sep 27 21:42:27 Yeah LOL Sep 27 21:42:44 I anyway mainly use phone for calls, SMS, email & IRC Sep 27 21:42:51 & music, alarm Sep 27 21:42:55 So not that demanding Sep 27 21:43:00 WhatsApp would be nice, but well :P Sep 27 21:43:03 I don't use phones at all if I can help it :) Sep 27 21:43:16 Same here, but on the go it's nice ;) Sep 27 21:43:30 Spend an hour commuting by bike, so music on the road is nice :P Sep 27 21:43:31 if I need to tell something to my (grand) parents, I ask wife to call them :) Sep 27 21:43:59 On daycare days 1,5 hrs in total, so :) Sep 28 00:17:55 FWIW, the previous generation webOS TVs are a *lot* cheaper - and you can still get a 65” display. **** ENDING LOGGING AT Thu Sep 28 03:00:01 2017