**** BEGIN LOGGING AT Mon May 17 02:59:57 2021 May 17 04:19:31 Anyone good with using swconfig that might be able to lend their knowledge? May 17 10:32:35 Build [#94](https://buildbot.openwrt.org/master/images/#builders/9/builds/94) of `lantiq/ase` completed successfully. May 17 10:37:20 what's the best way of disabling a procd/init.d service when building an image? first thought was having a file within /etc/uci-defualts/ doing `/etc/init.d/X disable` however that would be reverted when factory resetting May 17 10:44:33 by reverted, you mean, "restored" right? May 17 10:44:57 only real problem with disable in a uci defaults file is that it still starts up and then gets disabled on that first boot. May 17 10:46:58 karlp: i want the service being disabled by default and only manually being, not "accidentally" being enabled and then starting during boot May 17 10:48:04 well, "factory reset" will restore uci-defautls files, as they're part of your "image" May 17 10:48:16 in other words: for my use case i never want the service being enabled, so it never starts at boot May 17 10:48:17 so youcan decide for yourself whether a uci defaults file is a suitable solution. May 17 10:48:36 I was just trying to clarify that you understood how the factory reset worked :) May 17 10:48:38 karlp: that's what i said (re factory reset), hence not liking the idea May 17 10:49:17 well, are you hoping to have an image that you have the service disabled, and then it can be manually enabled, and then they factory reset and it stays enabled? May 17 10:49:23 because that's not what factory reset would imply. May 17 10:49:39 if the service is disabled in your image, factory reset goes back to that.... May 17 10:49:40 putting empty files a la /etc/rc.d/SXXY into the image (e.g. via files/) sounds pretty hack'ish, either May 17 10:50:06 karlp: i want to manually start (not enable) the service whenever i need it May 17 10:50:08 hackish/clear, easy to follow solution... May 17 10:50:23 but it should never be started without me explicitly starting it May 17 10:50:48 hence, it should always be in disabled state May 17 10:50:52 remove teh START= line from the init script? May 17 10:51:24 oh, ok, indeed that's the part triggering the symlink creation when building the image May 17 10:51:38 that could work indeed, thanks May 17 10:52:04 involves modifying a package file, but i could live with that May 17 10:53:36 now's when you want bitbake/oe's "layers" :) May 17 10:54:38 karlp: we have openwrt.git/files/ which could do the job, however that's a) part of the environment and b) changes within the package file need to be tracked May 17 10:54:55 but it will do the job May 17 13:40:03 Hello humans. Does any1 know how to glue collectd-exec together with collectd-rrdtool to generate RRD stats with luci-app-statistics? I'm somehow stuck. My collectd-exec binary runs and prints valid values to stdout, but collectd does not create any *.rrd file. What am I missing? May 17 13:55:22 nbd: I have some time to debug the precal stuff on the unifi 6 lite today if you have time May 17 13:55:42 yes, that would be great May 17 13:55:47 please start by testing latest mt76 master May 17 13:55:48 nbd: I just tested a brand new build after you disabled precal and everything is running at full speed again May 17 13:55:54 great May 17 13:56:08 over 400 download on a 40mhz channel, very nice May 17 13:56:09 i'll make another patch for you to test on top of that master version then May 17 13:56:50 ok May 17 13:57:14 https://termbin.com/rnc8 May 17 13:57:50 ok May 17 14:58:05 nbd: I've seen the disappearing signal weirdness on my MT7603, once again… I don't know if I should open an issue on mt76, since I have no idea on how to reproduce it. :/ May 17 14:59:14 It just… happens. May 17 15:01:12 nbd: openwrt master from today, with CONFIG_SRC_TREE_OVERRIDE=y, and then clone from https://github.com/openwrt/mt76.git, I applied the patch you provided. Download is 30-50mbit, upload is normal. May 17 15:02:11 netprince: okay, thanks. i'll report that back to mtk May 17 15:03:03 rsalvaterra: how often does it happen? May 17 15:03:33 ndb: Hard to say. It happened to me twice in the last week, or so. May 17 15:05:32 nbd: Note that at the MAC level everything seems peachy. No errors in dmesg, all VIFs are there, iw reports everything ok. However, on my phone's analyser, the signal is gone (an no 2.4 GHz-only device is able to connect). May 17 15:05:36 *and no May 17 15:05:43 ok, let me know if you want me to test anything else May 17 16:30:58 Build [#98](https://buildbot.openwrt.org/master/images/#builders/64/builds/98) of `realtek/generic` completed successfully. May 17 20:40:11 dnsmasq in 19.07 is affected by CVE-2021-3448: https://security-tracker.debian.org/tracker/CVE-2021-3448 I think OpenWrt is not affected in normal configurations. It looks harder to backport the fix and the big Linux distributions didn't do it. Is someone working on a backport? May 17 20:40:27 This is fixed in master and 21.02 May 17 20:41:34 manawyrm: I am planning to tag 21.02.0-rc2 tomorrow and 19.07.8 in some days, do you want to merge some kernel driver changes in the package feed? May 17 20:43:29 hrrm fwiw ath10k fun and Bt-HHv5a (lantiq/xrx200) seem problematic in 21.02 ....... May 17 20:43:38 i'm testing non-ct ath10k firmware on one box... May 17 20:43:50 https://bugs.openwrt.org/index.php?do=details&task_id=3635 etc. and other bugs aiui ;-( May 17 20:44:10 I don't know what OpenWRT policy is on ''release critical bugs'' for want of better term (borrowed from debian world) May 17 20:44:40 manawyrm: sorry I wanted to highlight mangix May 17 20:51:32 nbd: are you working on backporting the mac80211 updates to 19.07 ? May 17 21:58:39 Hauke: no, i don't have the time for it at the moment May 17 22:10:45 is there an easy way of using (lib)ubus for starting and stopping procd-services? May 18 01:50:22 Anyone around who knows the Makefile rules well? Is there a reason we can't compile PROFILE and PROFILE_SANITIZED in rules.mk? May 18 02:13:08 philipp64 I don't know any thing about the make file, but I wish we could speed up the building of OpenWrt! May 18 02:13:36 There is a lot of times when my CPU is not being used to it's fullest. **** ENDING LOGGING AT Tue May 18 02:59:56 2021