**** BEGIN LOGGING AT Thu Oct 29 02:59:57 2020 Oct 29 04:38:46 sounds like a round-to-next 64k block Oct 29 07:21:57 ynezz: thanks, this happens almost all the time when there is a power cut Oct 29 07:22:20 ynezz: in my experience, fsck is not always enough because then the builder complains about missing files Oct 29 07:22:41 I just clean up each builder datafile in /builder/ Oct 29 07:54:39 ynezz: I suggest to remove the barrier=0 mount option to avoid these manual fixups, any opinion? Oct 29 07:55:07 this option was probably added to speed up builds, I have no idea how much it helps Oct 29 07:55:50 if build speed is important, I think we have space to build everything in a ramdisk :) Oct 29 07:57:06 just put dl/ on disk Oct 29 09:04:48 the builds are barely I/O bound, it's always CPU Oct 29 09:08:50 I've removed that barrier=0, thanks for the hint Oct 29 09:09:47 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_sunxi.html has been updated. (0% images and 99.7% packages reproducible in our current test framework.) Oct 29 09:16:08 Good morning, guys! Oct 29 09:17:43 I'm creating an hostapd-basic-openssl variant and things are looking good. Oct 29 09:17:56 811781 bytes (hostapd-full) Oct 29 09:17:56 531573 bytes (hostapd-basic-openssl) Oct 29 09:20:58 I personally have no use (yet?) for a wpad-basic-openssl, but I can also add it, if anyone wants it. Oct 29 09:27:31 morning rsalvaterra Oct 29 09:28:24 Hey, nitroshift, how's it going? :) Oct 29 09:28:45 rsalvaterra, lazy day at the office today Oct 29 09:29:28 might as well continue cracking at the mwlwifi firmware blob Oct 29 09:29:43 Heh. My bedroom is the office since early March this year. :P Oct 29 09:30:02 rsalvaterra, i love you too. NOT! :)) Oct 29 09:30:43 Oh, dear…! Mwlwifi is a sad story. Oct 29 09:31:13 rsalvaterra, tell me about it :( Oct 29 09:53:09 what about mwlwifi? Oct 29 09:54:53 i want to submit a small patch and ideally i'd like to include the boot log as documentation for the change. where would i put the log in the mail? Oct 29 09:57:07 below the tear line above the diffstate Oct 29 09:57:16 so a patch has Oct 29 09:57:23 prefix: subject Oct 29 09:57:28 description Oct 29 09:57:35 Signed-off-by: ... Oct 29 09:57:43 --- ( the tear line) Oct 29 09:57:52 (place your log here) Oct 29 09:58:00 thanks Oct 29 09:58:17 diff --git a/package/bas .... (the diffstat) Oct 29 09:59:44 mangix, mwlwifi *can* be improved if we know what lies inside that firmware blob Oct 29 10:00:18 somehow reverse engineer it Oct 29 10:00:19 ynezz: mounting with barrier=0 == "please, eat my data". ;) Oct 29 10:30:17 Hmm… I'm seeing lots of netifd complains on my log… https://paste.debian.net/1169061/ Oct 29 10:30:31 sh: acs_survey: out of range Oct 29 10:30:31 sh: out of range Oct 29 11:14:48 Done. :P https://patchwork.ozlabs.org/project/openwrt/patch/20201029111005.128590-1-rsalvaterra@gmail.com/ Oct 29 12:11:37 been wondering... if 'openwrt' had to choose a tool for their repos/automation/scripts in 2020 - would people still pick something beside make/shell ? Oct 29 12:17:59 certainly meson Oct 29 12:18:02 * ynezz hides Oct 29 12:20:20 Oh, ffs. Oct 29 12:26:43 mangix good luck with the mwlwifi blob! Oct 29 13:11:50 >KGB-1< https://tests.reproducible-builds.org/openwrt/openwrt_omap.html has been updated. (0% images and 100.0% packages reproducible in our current test framework.) Oct 29 17:41:16 >KGB-1< https://tests.reproducible-builds.org/openwrt/openwrt_lantiq.html has been updated. (98.2% images and 100.0% packages reproducible in our current test framework.) Oct 29 17:43:16 dangole: Hey! :) Oct 29 17:58:07 Acinonyx: how is your OSPF fu? Oct 29 18:46:24 dangole: ping Oct 29 18:54:52 rsalvaterra: pong Oct 29 18:55:54 dangole: Ok, I have the patch ready to send. To be honest, like you said, it doesn't make much sense to show size comparisons. Oct 29 18:56:23 rsalvaterra: just skip that then, by now it's more about the number of hostapd variants we are building... Oct 29 18:56:47 but i'm about to nuke hostapd-hs20 (because it's covered by hostapd/wpad-full now)... Oct 29 18:56:50 Yeah, and it's a lot of them… that submenu is getting crazier by the day. Oct 29 19:03:43 rsalvaterra: at least they have all been banned into that menu, before they were on the loose in the Network menu... Oct 29 19:03:59 dangole: Alright, squashed and sent. Oct 29 19:04:16 LOL! Banished like the daemons they are? :P Oct 29 20:55:46 touch ~/test && ls -alZ ~/test Oct 29 20:56:00 never mind, wrong chan Oct 29 20:59:49 dangole: I just tested a build with wpad-basic (internal crypto) and I don't have WPA3/OWE…? Is this expected? Oct 29 21:01:40 wpad-basic is just WPA2 with 802.11w Oct 29 21:02:30 Well, that's definitely confusing, as the other -basic variants have WPA3. Oct 29 21:03:37 you guys are making sure all the help text on those menuconfig items are up to date too right? :) Oct 29 21:03:38 I believed the feature set was the same across the same basic/full variants. Oct 29 21:04:34 karlp: By the looks of it, I think I'm not done yet with the WirelessAPD menu. :P Oct 29 21:07:26 rsalvaterra: oh. then it should probably work with wpad-basic-openssl as well. it definitely cannot work without any SSL library linked with it. Oct 29 21:09:41 dangole: Yeah, it does. Are you saying the hostapd internal crypto implementation isn't enough to support WPA3/OWE, only WPA2? Oct 29 21:11:22 rsalvaterra: exactly. Oct 29 21:11:51 Ah, that's a whole other story. :) Oct 29 21:37:29 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_bcm47xx.html has been updated. (100.0% images and 100.0% packages reproducible in our current test framework.) Oct 29 21:44:51 rsalvaterra: :+1: Oct 29 21:46:37 karlp: I must have a really short attention span… I just noticed hostapd is still using select() instead of epoll() (there's a config option for that). :P Oct 29 21:47:57 I'm sure there's no problem enabling it… epoll() is supported since Linux 2.5.something… Oct 29 21:59:02 does anyone have good OSPF (Quagga/FRR) few that could look at a config and tell me how I’m being stupid? Oct 29 23:06:34 Hi is there a work around for this bug? https://bugs.openwrt.org/index.php?do=details&task_id=3344 Oct 29 23:15:16 OK found it. https://github.com/openwrt/openwrt/commit/34a96529041d4e9502c490c66f8af0154187c6d2 Oct 30 00:34:26 rsalvaterra: how many connections are you planning on running with hostapd for it to even matter? :) Oct 30 00:35:29 build #595 of x86/geode is complete: Success [build successful] Build details are at http://buildbot.openwrt.org/master/images/builders/x86%2Fgeode/builds/595 Oct 30 00:36:10 karlp: Me, personally, not many… but the support is there, why not use it? :) Oct 30 00:36:48 Besides, I know of public hotspot deployments which might benefit. Oct 30 00:42:27 i guess this is more about open filehandles, which is kinda relevant if you have many things connecting to the control socket simultanously Oct 30 00:48:06 dangole: I'm looking at the CONFIG_WPS_NFC setting now… Is there any reason it's available on -wolfssl-full but not -openssl-full? Oct 30 00:48:54 Actually, it's in both -mesh and -full. Oct 30 00:50:45 rsalvaterra: i've included that long ago when i packaged wpad-mesh because (at least back then) it didn't build without it as some crypto functions used by SAE would only be built with CONFIG_NFC enabled as well. not even sure if this is still true. Oct 30 00:51:43 Ah, nice to know! I'm trying to clean things up a bit, so I'll test SAE with/without this too. Oct 30 01:01:46 rsalvaterra: i've merged your wpad-basic-openssl patch Oct 30 01:02:53 Thanks! Oct 30 01:34:55 build #680 of octeontx/generic is complete: Failure [failed pkgbuild] Build details are at http://buildbot.openwrt.org/master/images/builders/octeontx%2Fgeneric/builds/680 blamelist: Rui Salvaterra , Chuanhong Guo , Daniel Golle , Paul Spooren , Philip Prindeville Oct 30 01:34:55 , John Thomson , Jo-Philipp Wich Oct 30 01:46:28 build #616 of ramips/rt288x is complete: Success [build successful] Build details are at http://buildbot.openwrt.org/master/images/builders/ramips%2Frt288x/builds/616 Oct 30 01:58:27 build #675 of oxnas/ox820 is complete: Failure [failed pkgbuild] Build details are at http://buildbot.openwrt.org/master/images/builders/oxnas%2Fox820/builds/675 blamelist: Rui Salvaterra , Chuanhong Guo , Daniel Golle , Paul Spooren , Philip Prindeville @redfish-solutions.com>, John Thomson , Jo-Philipp Wich Oct 30 02:09:10 build #536 of mpc85xx/p2020 is complete: Failure [failed pkgbuild] Build details are at http://buildbot.openwrt.org/master/images/builders/mpc85xx%2Fp2020/builds/536 blamelist: Rui Salvaterra , Chuanhong Guo , Daniel Golle , Paul Spooren , Philip Prindeville @redfish-solutions.com>, John Thomson , Jo-Philipp Wich Oct 30 02:10:38 build #666 of at91/sam9x is complete: Failure [failed pkgbuild] Build details are at http://buildbot.openwrt.org/master/images/builders/at91%2Fsam9x/builds/666 blamelist: Rui Salvaterra , Chuanhong Guo , Daniel Golle , Paul Spooren , Philip Prindeville @redfish-solutions.com>, John Thomson , Jo-Philipp Wich **** ENDING LOGGING AT Fri Oct 30 02:59:57 2020