**** BEGIN LOGGING AT Wed Aug 01 03:00:00 2018 Aug 01 03:41:51 philipp64: that link doesn't resolve, but i see what you are referring to Aug 01 04:25:13 philipp64: works for me Aug 01 04:28:07 russell--: awesome. just tried an x86_64 build with the same recipe… so apparently -fno-strict-aliasing doesn't break the x86_64 build. Aug 01 06:15:09 rspro on ath79 :D [ 5] 0.0-60.0 sec 6.37 GBytes 913 Mbits/sec Aug 01 06:15:26 was this before enabling the flow offload: [ 4] 0.0-60.0 sec 3.75 GBytes 538 Mbits/sec Aug 01 06:15:32 * m4t amazed Aug 01 06:16:33 m4t: where are those stats from ? Aug 01 06:17:04 m4 can you add -l 512 -P 10 to the iperf3 cmdline Aug 01 06:17:05 iperf. i set up rspro's wan to be my normal lan, then am connected to rspro lan via a usb3 ethernet Aug 01 06:17:11 and use iperf3 not iperf please Aug 01 06:17:32 ok i'll install it Aug 01 06:17:44 it didn't build w/ipv6 disabled :/ Aug 01 06:20:47 you're not supposed to be disabling v6 :\ Aug 01 06:20:58 i don't use it anywhere Aug 01 06:21:06 * m4t always removes stuff he doesn't use Aug 01 06:21:34 blogic: https://paste.ee/p/eJxF4#Q8GO2DdJbuOic08PryQxyq1gf5TyamRq Aug 01 06:22:45 that's weird though. i just ran it again with top running on the rspro, and it said 35% idle, 65% sirq Aug 01 06:23:14 (anywhere on my lan i mean, i'm obv connected to irc via ipv6 lol) Aug 01 06:23:15 philipp64: imho simply bump perl Aug 01 06:23:28 you kinda should be using it by this point. being stuck dealing with NAT is just ugh Aug 01 06:23:44 philipp64: in contrast to other languages *cough* Lua *cough* a minor version update usually does not mean incompatible syntax and removed features Aug 01 06:24:04 Dagger: I am a big fan of IPv6 but its only fun when native Aug 01 06:24:25 Dagger: I have to disable the ipv6 tunnel here whenever the gf wants to watch netflix on the ipad, because its seen as "vpn" Aug 01 06:24:42 heh, yeah my isp doesn't offer native Aug 01 06:24:55 i used to have an he.net tunnel..back in like, 2000 or 2002 Aug 01 06:24:57 m4t: that is the max throughput of your gmac Aug 01 06:25:03 blogic: oh ok Aug 01 06:25:11 m4t: or your usb3 Aug 01 06:25:12 * m4t wonders if it's the asix Aug 01 06:25:14 yeah.. Aug 01 06:25:16 uhm Aug 01 06:25:25 * m4t thinks Aug 01 06:25:36 still 100kfps Aug 01 06:25:40 no native does suck a bit, but not building v6 support into your router firmware is just going to make it harder to deploy when your ISP finally provides it Aug 01 06:26:01 jow: FWIW there's a DNS proxy somewhere on github specifically for filtering out Netflix's AAAA records Aug 01 06:26:08 make menuconfig && make -j16 takes no time at all lol Aug 01 06:26:17 Dagger: are they bad ? Aug 01 06:28:48 Netflix publish AAAAs, and their software uses them, but then their servers refuse to serve video over v6 (at least to some clients) Aug 01 06:29:57 ok, not had that problem yet Aug 01 06:31:17 Dagger: yeah I know, but thats basically solving ipv6 problems I only have because I've gone out and got myself ipv6 Aug 01 06:31:42 not doing it is far simpler and all my media consumption stuff just works Aug 01 06:32:09 (I know how to workaround etc. etc., but this just gives a little bit of context why people still avoid ipv6 in this day and age) Aug 01 06:32:48 blogic: https://paste.ee/p/TpLuZ#Tf9vVbZhuCTynKtQA3L7zpn2AQhccxbu is direct to the switch without going through rspro Aug 01 06:33:00 not the rspro switch...my regular gige switch Aug 01 06:34:03 ok, so 100kfps is the max of the ethernet inside the rspro Aug 01 06:34:05 nice Aug 01 06:34:33 i'm going to see what i get with sqm Aug 01 06:35:42 btw, to clarify, the rspro builds to have ipv6 enabled at the moment, while i ready things for a PR. not doing any fancy local configs while i test this stuff. Aug 01 07:34:33 anyone here with a netgear r6100 ? Aug 01 07:34:43 I consider ordering one myself to check those ath10k issues Aug 01 07:35:14 but since its quite some crappy hw for the price I would avoid it if possible Aug 01 07:53:03 morning Aug 01 07:53:20 ok can someone verify if the xiomi 3C RL is suppoorted or not... its not clear looking at the wiki Aug 01 07:53:56 Hello all .. I am facing a strange issue in my OpenWRT CC .. Whenever I create an SSID (an then press save & Apply), i get cfg numbers for that wifi interface with difference of 2 .. like cfg04xxxx cfg06xxxx etc .. But, then I only save SSIDs (not pressing save & apply, clicking Save button only), then i get CFG numbers with increment of 1 , like cfg08 cfg09 .. It is causing problem .. Do anyone know where are CFG numbers genera Aug 01 07:54:01 jow: ?? Aug 01 07:54:53 obaid: yeah, the is uci which creates the section hashes based on the contents Aug 01 07:55:07 so if you change the options within a section, it will recalculate the hashes Aug 01 07:56:35 jow: Where can I check how CFGs are created?? I am having problem when I "Save" WiFi sections (and not using "Save & Apply") Aug 01 07:57:19 obaid: https://git.openwrt.org/?p=project/uci.git;a=blob;f=list.c;h=25aec56fb31afde8bcc1526fe070ad7574c4b4cc;hb=5d2bf09ec594d97eb9284b8c721dbfe10b81a6f6#l150 Aug 01 07:57:21 jow: So, i want to check (and apply condition) to fix this issue Aug 01 07:58:22 obaid: this is unfixable in uci Aug 01 07:58:42 it needs to be fixed in luci but the situation is extremely complex Aug 01 07:58:55 the easiest workaround is simply naming the sections Aug 01 07:58:57 jow: I am not having this issue in openWRT AA Aug 01 07:59:21 yeah because AA doesn't mix in ubus runtime info which might go out of sync with the underlying configs Aug 01 08:00:13 in any case it needs to be fixed in luci, not uci Aug 01 08:00:26 but then CC is a dead end (as is luci 0.10 and luci 0.11) Aug 01 08:00:41 a simple workaround might be to simply name the sections Aug 01 08:00:49 e.g. "radio0net1", "radio0net2" Aug 01 08:01:03 then uci's recalculated section hashes will be no issue anymore Aug 01 08:05:24 jow: The issue is not appearing with "Save & Apply" because luci recalculates it ?? Aug 01 08:05:31 yes Aug 01 08:05:55 save & apply will commit the changes which will let uci recalculate and finalize the hashes, then the form is rerendered from the ocmmitted config state Aug 01 08:06:04 on save, the hashes are not recaclulated iirc Aug 01 08:08:01 jow: Ok .. and how can I create a named section?? any reference link ? Aug 01 08:08:23 jow: via TypedSection ?? Aug 01 08:09:52 obaid: you need to override s:create() iirc and make it add a named uci section instead of an anonymous one Aug 01 08:58:50 so the release notes say "Devices using the mt76 wireless driver may experience stability issues on 2.4GHz" Aug 01 08:58:58 does that include device resets? Aug 01 09:15:32 so I think I managed to fix usbip, it should appear in the repos during the next one or two days Aug 01 09:15:57 the failing patch was adressed by picking the appropriate fixes from feed master to openwrt-18.06 Aug 01 09:16:20 the build-in-sdk issue will be addressed by properly shipping the usbip tools kernel sources with the sdk Aug 01 09:29:07 hmm, if wifi is unusable on this AP the device is kinda useless Aug 01 10:03:48 2.4gig has been flaky since ~forever on ralink parts, mt76 included Aug 01 11:05:54 karlp: ok but does that include spurious device resets? Because I have an AP that would reboot every few hours that's been stable overnight after I disabled both radios (2.4 and 5)... Which is a bit puzzling ;P Aug 01 11:06:12 m4t: I updated a bit the ubntrs branch (now useless initramfs is not created) and added ubntrsx to show an example of combined dts. The rspro dts is based on a paste You posted here. Regarding added packages/kmods, it's usually what submitter decides based on his/hers preferences. Aug 01 11:07:11 tmn505: not really. it is the bare minimum to get the hardware working Aug 01 11:07:51 oh Aug 01 11:07:55 i was actually about to send a PR to you Aug 01 11:08:02 i split everything and made a rspro.dtsi Aug 01 11:09:50 i'll push it to github in a minute, i need to make some commit messages Aug 01 11:10:11 karlp: even more so considering this happens with _no_ client associated Aug 01 11:10:52 mkresin: I should stated moderate, nobody would like samba by default even if it fits. Aug 01 11:11:18 m4t: thanks Aug 01 11:11:30 yeah! Aug 01 11:11:39 i ran some nat routing benchmarks with flow offload Aug 01 11:11:46 pretty much hit wirespeed with large packets Aug 01 11:13:09 that's nice, unfortunately rs won't benefit from it, as driver for the switch is nonexistent. Aug 01 11:13:36 oh, yeah 100mbit Aug 01 11:14:27 i tested the urescue in redboot too Aug 01 11:14:29 tftp recovery Aug 01 11:14:38 i had to tweak the variables in generic-ubnt.mk Aug 01 11:15:53 which one was it? Aug 01 11:16:32 well when i switched to reuse vars i wasn't aware of how they were turned into the header of the factory image Aug 01 11:16:46 so it said RSx ar7100 instead of RSPRO ar7100pro Aug 01 11:17:20 yes those are difrent for both boards. Aug 01 11:20:09 have you tested wifi much? Aug 01 11:20:34 i don't have a good setup to test throughput, and it's pretty congested where i live too Aug 01 11:24:06 nope, not yet. I'll have to first update all the boards to stable. Aug 01 11:25:04 oh Aug 01 11:30:49 OutBackDingo: do you already have the 3C model? because the 3G does not cost more and has better specs Aug 01 11:31:17 jow: Thank you for the hint to save as named section .. I made changes in model/network.lua and passed a name in section() function .. It is working fine now .. Aug 01 11:32:00 obaid: how do you generate the names? Aug 01 11:32:06 I consider doing the same in master Aug 01 11:33:12 jow: I used some bash commands to get the number of created instances of Radio0 and Radio1 .. and then I generated a name as "radio0net1" etc .. and passed this value in section() function .. Aug 01 11:36:39 ok Aug 01 11:38:44 jow: It might now be the most efficient way, but it is working :D :D Aug 01 11:43:51 not* Aug 01 11:50:58 m4t: if You'll have something ready ping me, I'll be away for few hours. Aug 01 11:51:05 k Aug 01 11:59:34 Hi all, I am playing with an old version of OpenWRT on a Zsun at the moment (great little device!). Anyone know of another device in a similar form factor e.g. as tiny and able to be powered directly from USB plug? Aug 01 12:00:21 you bricked it? Aug 01 12:01:05 lol no, not yet. They are getting harder to get and wondering if there are any similar sized devices as an alternative. Aug 01 12:01:34 iirc the reason support for it was never merged was that openwrt leaves wifi off by default, and that device has no ethernet Aug 01 12:03:29 Right, that makes sense. Pity as wifi is quite capable on it being a AR9331 Aug 01 12:10:25 18.06 for the zsun would be lovely, yes. I don't see anyone working on it though. Aug 01 12:10:54 takimata: working on it would require the original problem to be solved Aug 01 12:13:42 J1mX: What about EAGET A50? Same device. Aug 01 12:13:58 same device, same problem. Aug 01 12:14:36 really: no ethernet, same problem. there's a generic problem to solve there. Aug 01 12:15:27 how to secure wifi after initial installation. Aug 01 12:16:41 takimata: Can it act as a usb client/ usb ethernet port Aug 01 12:17:58 Rene__: Thanks, I wasn't aware of the EAGET A50. That opens up some more supply :) Aug 01 12:22:16 Rene__: it can be a USB host, if that's what you're asking. but the problem is really more generic than that, and it boils down to "how to create a non-guessable, yet 'knowable' password" Aug 01 12:23:47 I don't even see a problem to get a solution into upstream. there's really no difference between "having wifi disabled" and "having wifi with a pre-set password disabled" for all other devices. Aug 01 12:23:50 takimata: nope, I think the cpu can also act as a usb client. So make it a usb network adapter. Aug 01 12:24:11 the issue is how to create this password. Aug 01 12:24:56 usb network is the "LAN" side with a dhcp server Aug 01 12:25:18 and with the web interface you can configure your wifi Aug 01 12:25:51 Rene__: but that's a very specific fix for this device then. what to do with devices without USB, or usb host capability? Aug 01 12:25:56 tmn505: oh i can't do a PR because yours is based on lede and mine is a fork of openwrt/openwrt :/ here is is https://github.com/tofurky/openwrt/tree/routerstation Aug 01 12:26:14 or act as a usb serial device and run a console behind it. Aug 01 12:26:30 Rene__: if we're talking device specific fixes, we can also just use the sdcard to store a wifi config. Aug 01 12:27:41 takimata: I say it can, but that seems also not standard. Aug 01 12:29:02 takimata: do you have examples of devices which only have wifi and no other "outside" connections that wifi? Aug 01 12:29:26 I seem to remember one or two others, but not off the top of my head right now. Aug 01 12:29:47 (in part because my brain is hitting temp limits and throttling.) Aug 01 12:47:11 I just tried out 18.06.0. I installed the map package so I could use LW4o6 tunnel for IPv4 connectivity over IPv6 WAN. Control plane seems to work, but as opposed from earlier 17.01 and 15.05, the LW4o6 (IPv4-in-IPv6 proto 42 packets) now have a destination header 4 (fragmentation header) on all packets. Is there a kernel knob (sysctl?) to turn this off? Aug 01 13:02:15 blogic: any updated ETA on the linksys e1700 test? Aug 01 13:02:46 DonkeyHotei: some point in the future Aug 01 13:03:00 honestly cant tell, too many things to do right now i am afraid Aug 01 13:04:01 in that case, if it is broken on 18.06, it will have to stay so Aug 01 13:04:52 i'll prepare a commit of the dt property removal with the change you suggested Aug 01 13:05:04 (for master only) Aug 01 13:40:58 gninrom Aug 01 13:43:09 which file is better for wrap unit, is it combined-ext4.img.gz or root-ext4.img.gz? Aug 01 13:43:20 combined Aug 01 13:43:27 the root is just the rootfs without kernel or loader Aug 01 13:44:16 jow: thank you :-) Aug 01 13:45:39 still same procedure. Unpack and by dd write to CF card, right? Aug 01 14:01:58 Hi, I need to rescue a few devices where the sysupgrade was interrupted Aug 01 14:02:10 I get this u-boot menu: https://paste.ubuntu.com/p/dzZGz2vwzp/ Aug 01 14:03:02 but under "Please choose the operation" there is no delay to select tftp or terminal Aug 01 14:03:20 is there a way to enter a selection? Aug 01 14:16:41 is it possible to set a cookie to rember login on next time? after browser ask password again Aug 01 14:18:01 login where? Aug 01 14:18:48 Luci Aug 01 14:21:01 hsp: edit /etc/config/luci and increase the session timeout Aug 01 14:21:59 option sessiontime '3600' Aug 01 14:22:08 ^^ current Aug 01 14:24:44 jow, 5000 or what? Aug 01 14:25:56 DonkeyHotei: yes cost isnt the issue, its about avaiability Aug 01 14:26:16 the 3G is at least as available Aug 01 14:26:39 DonkeyHotei: so i do actually have a 3C model R3L Aug 01 14:26:44 hsp: 3600 is one hour, 86400 would be one day, 2592000 roughly one month Aug 01 14:26:52 DonkeyHotei: not where i am its not Aug 01 14:27:06 i even tried to get a xiomi mini or nano Aug 01 14:27:11 no joy Aug 01 14:27:39 does it use the same firmware as the Router 3 ? Aug 01 14:27:45 hsp: however I just noticed that its a session cookie now, so it will expire when closing the browser Aug 01 14:28:21 and after closing the Browser ist not possib Aug 01 14:28:27 and after closing the Browser ist not possible? Aug 01 14:30:45 DonkeyHotei: no idea, we just got htis thing how can i tell ? Aug 01 14:31:30 hsp: a workaround is to setup basic authentication: Aug 01 14:31:32 hsp: echo '/:root:$p$root' > /etc/httpd.conf Aug 01 14:31:40 are you able to download the original firmware from xiaomi? Aug 01 14:31:51 hsp: then your browser will prompt you for username and password which you can then remember Aug 01 14:32:09 hsp: luci will then use thiese credentials for login and you do not need to perform an extra login anymore Aug 01 14:32:32 hsp: only downside is that logout of luci will not work anymore, you'll always be logged in Aug 01 14:32:55 to disable that again, delete /etc/httpd.conf Aug 01 14:33:09 and restart the server using /etc/init.d/uhttpd restart Aug 01 14:38:43 DonkeyHotei: one sec Aug 01 14:41:08 jow, big thanks. works :) Aug 01 14:42:01 DonkeyHotei: kinda hard to tell, looks like a diffferent version to me http://en.miui.com/thread-379049-1-1.html Aug 01 14:54:48 DonkeyHotei: seems the 3C is a Mediatek MT7628N Aug 01 14:54:58 hey i want to reduce host compile time, by only building my qemu host helper for the target arch, instead of all. Yet i can't figure out a trick to check/reinitiate the build if the $ARCH changed, so the needed helper can be build? Any ideas? Aug 01 14:55:40 OutBackDingo: then it probably doesn't have existing support in openwrt and would need investigation Aug 01 14:55:57 what package needs host qemu? Aug 01 14:55:59 the 3G is MediaTek MT7621AT Aug 01 14:56:08 my samba4 package Aug 01 14:56:19 the 3 is 7620 iirc Aug 01 14:56:31 the 3G is 7621, yes Aug 01 14:59:22 DonkeyHotei: interestingly the xioomi nano is supported by openwrt and is also MediaTek MT7628 , the wiki also states the OEM firmware is a openwrt fork/derivitive.... now im curtious Aug 01 15:00:37 i wonder if the nano xiomi fioirmware is the same as the 3C ... time to hunt some info Aug 01 17:23:01 I enabled /dev/mem to try to debug my MTD/SPI boot problem Aug 01 17:23:07 but the mere act of doing that made it work :| Aug 01 17:23:34 bloody heisenbugs Aug 01 17:27:04 ^-this. Aug 01 18:46:52 nice, ath10k is even more broken now Aug 01 18:47:01 stintel: Cool. Aug 01 18:48:18 so it was already worse that when it breaks, it stays visible and devices keep trying to connect to it Aug 01 18:48:23 now it's not even fixed by running "wifi" Aug 01 18:49:27 stintel: in the release? Aug 01 18:56:27 Anyone know Eric Leuhrsen's IRC handle? Aug 01 18:59:01 philipp64: I think he's not present in irc at all Aug 01 18:59:52 however in my expierence he's responsive on github if you @mention him there, or in the forum. Aug 01 19:01:16 okay, thanks…. trying to get dhcpd (isc-dhcp) to come after the relevant DHCP interfaces are up, as it's a problem on some slower platforms (issue 6382). but the reload and service_trigger stuff is a little unclear. Aug 01 19:02:03 jow: sorry I missed having a fix in time for 18.06.1 for Perl's CVE. Aug 01 19:05:58 philipp64: packages are continuously built from the openwrt-18.06 branch in the feed Aug 01 19:06:18 philipp64: if you cherry-pick your fixes there, they'll get built and made available in the 18.06 opkg repositories Aug 01 19:06:29 that applies to all non-builtin and non kernel-related packages Aug 01 19:24:24 m4t: I dropped few comments on Your commits in routerstation branch. Aug 01 20:09:27 tmn505: hmm i tested and it seemed like DEVICE_PACKAGES := overrode the ones in the Makefile, which is why i did it that way Aug 01 20:09:38 i did have all the -pkg before Aug 01 20:10:11 the only way it had any effect was if i also enabled CONFIG_TARGET_PER_DEVICE_ROOTFS=y Aug 01 20:13:40 hey guys. can someone quickly tell me what numbering a backport needs under the new patch numbering scheme? Aug 01 20:14:49 re: backlog, is a 18.06.1 release expected in short order? Aug 01 20:16:19 m4t: ok, I'll re-checkit. Aug 01 20:16:19 having the latest -stable kernel in there would be nice; I saw it landed on master but not on the 18.06 branch. Aug 01 20:16:35 yeah it kinda threw me off at first too Aug 01 20:17:06 Sagi: it's being worked on, there was an ext4 showstopper bug that got fixed in master for 4.14, but not fully for 4.9 Aug 01 20:18:01 ah, I saw some ext4 changes. Thanks! Aug 01 20:18:33 i'm running 4.14.59 and 4.9.116 myself on 18.06, but i wasn't bitten by the ext4 bug Aug 01 20:18:46 fs bugs suck Aug 01 20:19:06 what can i say :P Aug 01 20:25:57 any devs around? Aug 01 20:27:55 while I'm here typing instead of lurking it might be fitting to express my gratitude for all the hard work that has gone into 18.06. I enjoy its fruits every day Aug 01 20:28:25 congrats on the release, thank you very much from a happy user. Aug 01 20:29:44 =) Aug 01 20:30:05 very happy with it as well, although i've been running 18.06 builds for a while Aug 01 20:34:47 jow: so basically everything in /packages but not in /openwrt. Aug 01 20:35:13 philipp64: roughly speaking, yes Aug 01 20:35:35 so… do I "git cherry-pick" directly into the branch and then "git push -u …." ? Aug 01 20:36:03 because github only understands merge requests and rebases, right? Aug 01 20:36:22 in other words, I need to do this from the CLI and not via GH? Aug 01 20:36:54 yeah, I usually do it from the cli (using git cherry-pick -x hash) Aug 01 20:37:15 the -x will automaticly add a (cherry picked from commit xxx) note Aug 01 20:37:27 jow: quick question - if i am backporting a patch from 4.14 to 4.9, should i prefix it with 0xx? Aug 01 20:37:33 Borromini: sec Aug 01 20:38:11 ty Aug 01 20:38:24 Borromini: https://git.openwrt.org/?p=openwrt/openwrt.git;a=blob;f=target/linux/generic/PATCHES;h=86ced468048680cccbf7e8b4cdffd0a7607b2162;hb=b9bd3285c7c6fbc60374b05ac2da9be41ce75651 Aug 01 20:38:44 jow: okay, thanks… I've been putting off doing it because I thought it was more complicated than that… Aug 01 20:38:55 jow: yeah i saw that on the wiki as well. so i take that as a yes :) Aug 01 20:39:05 Borromini: so in this case 0xx Aug 01 20:39:18 since its clearly upstream (present in 4.14) and not awaiting merge Aug 01 20:39:28 jow: yeah. i just saw koen turned it into backport-4.9/500-ext4-fix-check-to-prevent-initializing-reserved-inod.patch though Aug 01 20:39:45 works too Aug 01 20:39:51 with the new patch dir structure Aug 01 20:40:15 ah yes FS related Aug 01 20:40:21 koen isn't hanging out on IRC is he Aug 01 20:40:32 not sure offhand Aug 01 20:40:43 if he does, he likely does with the handle xback Aug 01 20:40:53 wouldn't be fair to me to take just his patch (and stintel's) and just add that to my bumps Aug 01 20:41:00 thanks :) Aug 01 20:42:34 jow: thanks for reverts Aug 01 20:42:45 just came back from the pool Aug 01 20:44:08 you have a pool? :( :( Aug 01 20:44:18 how hot is the german hinterland right now? :) Aug 01 20:48:12 its cold tonight Aug 01 20:48:15 26 Aug 01 20:48:23 down from 36 yesterday Aug 01 20:48:32 o_O Aug 01 20:48:40 where are you situated jow? Aug 01 20:48:45 leipzig Aug 01 20:48:48 oh Aug 01 20:49:09 let me look that up, i can locate munich, frankfurt and berlin... but leipzig i don't know which Land that is :^) Aug 01 20:49:18 middle between berlin and munich Aug 01 20:49:29 more or less Aug 01 20:49:48 ok close to czech republic? Aug 01 20:49:55 It fairly regularly gets up to 36 in the summer where I live (North Carolina, US) but everyplace here has air conditioning, which makes it bearable. Aug 01 20:49:56 in Sachsen Aug 01 20:49:56 nah, middle? i would have said nearer to berlin Aug 01 20:50:19 mamarley: more and more people are resorting to A/C here as well Aug 01 20:50:22 mamarley: AC isn't really common here... and 35° today, >20 at night Aug 01 20:50:31 i had to move my mattress to the living room to keep it bearable Aug 01 20:50:39 summers keep getting hotter. Aug 01 20:50:39 Borromin1: yeah, one could say that. About an hour Aug 01 20:50:54 jow: how are summers in general? hot? Aug 01 20:51:01 dry and warm Aug 01 20:51:08 but not as extreme as this year Aug 01 20:51:11 was in berlin half june and it was pretty hot Aug 01 20:51:13 when i was in leipzig ~4 weeks ago it was pretty warm though, too ^^ Aug 01 20:51:17 :-/ Aug 01 20:51:32 A/C isn't common in private homes in .de, offices and stores to some extent, but not homes Aug 01 20:51:45 I've neither have AC at home, nor at work Aug 01 20:51:46 i am so happy my employer has A/C Aug 01 20:51:50 and both are under the roof Aug 01 20:51:55 oh god. Aug 01 20:51:58 hehe Aug 01 20:52:01 jow: ouch :P Aug 01 20:52:11 have no a/c either, but not under the roof both places Aug 01 20:52:25 my apartment is south facing. luckily my balcony shields the living room from the sun, but not so for my bedroom Aug 01 20:52:43 my apartment is still like 27° C because the building has heated up so much the past few weeks Aug 01 20:52:51 dark bricks are a gift from hell. Aug 01 20:53:24 im using a big fan each night to push colder air inside :D Aug 01 20:53:46 but maybe fans are sold out atm Aug 01 20:53:57 pkgadd: norway's had a pretty warm summer too, and, as you can imagine not, a lot of people have ac here either Aug 01 20:54:17 living room was 30° this morning and during the night I couldn't really open the windowes due to a thunderstorm Aug 01 20:54:29 which made the temperatures drop from something like 31 to 28 degrees Aug 01 20:54:43 while simulatenously keeping me awake Aug 01 20:54:44 yep, this summer is a tad.. special... dark bricks, black roof tiles - it's fun... Aug 01 20:55:21 my brother has floor heating/cooling Aug 01 20:55:27 i wish i had a heat exchange pump Aug 01 20:55:32 it's effing 20°C in his house... Aug 01 20:55:35 ^^ Aug 01 20:55:38 my brother has one :P Aug 01 20:55:39 *shudder* Aug 01 20:55:40 cooling during summer and heating during winter Aug 01 20:56:12 the great thing about it being in the floor is it doesn't give you any headaches or running noses like traditional A/C Aug 01 20:56:54 I'm out for today, gn8 Aug 01 20:58:04 rotanid: i work in a store that sells fans in norway and pretty much everyone is sold out Aug 01 20:58:38 night jow Aug 01 20:59:10 huaracheguarache: weather like this makes everyone run down the store. same for ordering A/C i reckon. Aug 01 20:59:42 and nobody will get it quickly enough because A/C guys probably have a backlog as is already, it being summertime and whatnot Aug 01 20:59:48 i'm out as well, goodnight Aug 01 21:01:22 cya Aug 01 21:09:53 Anyone having issues with the RaspberryPI_3B on 18.06.0 and the Thermal sensor? Aug 01 21:10:43 there is nothing in /sys/class/thermal, I've installed hwmon-core and hwmon-bcm2835 but this did not help Aug 01 21:11:03 LEDE 17.01.4 worked fine Aug 01 21:12:58 In 17.01.4 there was no need for installing hwmon-bcm2835 but still it worked Aug 01 21:13:55 tmn505: i gotta run for a bit but i'll add the changes aside from the DEVICE_PACKAGES. i'm fine with just using DEVICE_PACKAGES += for everything too instead of := Aug 01 21:18:04 I noticed commit "da529d54107b1c1fe281274c723ea4789838c54b" which was not in the 17.01.x tag, did anyone actually test this? Aug 01 21:36:44 m4t: *facepalm* You made me realise that You're not adding to packages just re-defining them. Please disregard my comments about DEVICE_PACKAGES, You're completely right, please keep it the way it is now. Aug 01 22:02:53 No one tried the RPI3b images I suppose? Aug 01 22:23:25 Any one know if the r71xx snapshots are fixt? **** BEGIN LOGGING AT Wed Aug 01 22:26:51 2018 Aug 01 22:34:27 jow: ping Aug 02 01:37:02 so i guess 18.08 is cut now right ? Aug 02 01:38:12 it's tagged 18.06.0 Aug 02 01:38:37 the bugfixes will be 18.06.1 Aug 02 01:49:11 cool Aug 02 01:49:47 DonkeyHotei: orders a usb kitso i can console this xiomi, and seeing if thhe nano image is the same for the 3C Aug 02 01:50:38 might wanna compare a gpl tarball for the 3c with one for the nano Aug 02 01:51:10 DonkeyHotei: gpl tar ball ? Aug 02 01:51:32 all i can barely find is xiomi images Aug 02 01:55:22 DonkeyHotei: hrmmm https://xiaomi-mi.com/wifi-routers/xiaomi-mi-wifi-router-3c-white/ Aug 02 01:56:00 Operating system OpenWRT MiWiFi ROM Aug 02 02:33:45 DonkeyHotei: interesting i found this also https://forum.lede-project.org/t/support-for-xiaomi-miwifi-3c/11643 Aug 02 02:48:13 this xiomi definitly seems to be LEDE/OpenWRT basedhttp://192.168.31.1/cgi-bin/luci **** ENDING LOGGING AT Thu Aug 02 03:00:00 2018