**** BEGIN LOGGING AT Sat Oct 27 03:00:00 2018 Oct 27 07:23:01 stintel: have to say I've not seen this problem...yet. All my static leases are outside the dynamic range (but in subnet) and so far all ok. Time for some logging? add 'log-dhcp' to the end of /etc/dnsmasq.conf Oct 27 08:06:08 stintel: is this since a particular time/dnsmasq version? do you have dhcp-authoritative ? I'm trying to work out why I don't see this problem. Oct 27 08:21:50 * ldir does a sysupgrade brb Oct 27 08:29:16 See :-) Oct 27 09:02:12 setting password to MAC address: good idea or bad idea? Oct 27 09:02:56 $PASSWORD=(cat /sys/class/net/wlan0/address); Oct 27 09:02:56 uci set wireless.radio0.key=$PASSWORD; Oct 27 09:02:57 uci commit; Oct 27 09:02:58 Password is now 12:34:56:ab:cd:ef Oct 27 09:06:53 from a security perspective it's bad because the MAC address is visible to other devices Oct 27 09:07:24 Oh boy, how did I miss that one Oct 27 09:09:47 use a password manager like for example keepassxc, which has a random password generator, and create a long random password Oct 27 09:38:27 https://www.xkcd.com/936/ Oct 27 09:39:23 random passwords are fun, until you need to enter one and you don't have your password manager with you Oct 27 09:40:08 or you could make a passphrase with diceware Oct 27 09:44:49 stintel: how's your wifi doing? stable and no crashes? Oct 27 09:47:20 huaracheguarache: booted my DAP-2695 at my parents' place 13d ago and not a single issue since then Oct 27 09:48:41 do you still have the archer c7? Oct 27 09:51:34 huaracheguarache: never had one Oct 27 09:52:29 right, sorry. I just had a look at the issue you filed on the ath10k-ct github and see that you used the Ubiquiti Unifi AP AC Pro Oct 27 09:52:33 damn insomnia Oct 27 09:53:19 that's my other AP back at home Oct 27 09:53:25 I have a DAP-2695 there as well Oct 27 09:53:42 do you use -ct ath on both? Oct 27 09:54:36 huaracheguarache: since the 11w slowness was fixed, yes Oct 27 09:55:01 the UAP AC Pro has been running for almost 25 days no issues Oct 27 09:55:23 that's nice Oct 27 09:55:37 but only 1 client connected currently Oct 27 09:55:45 as I am not there :) Oct 27 09:55:49 hehe Oct 27 09:56:11 just my ESP8266 Oct 27 09:56:29 which is weird, my OnePlus One should be connected too Oct 27 09:57:31 actually on the DAP-2695 in Sofia I see this: [14858.010190] ath10k_pci 0000:00:00.0: SWBA overrun on vdev 0, skipped old beacon Oct 27 09:57:52 yeah, I've seen that on my r7800 as well Oct 27 09:58:08 usually things slow to a crawl after that message Oct 27 09:58:24 hmm Oct 27 09:58:34 it's weird because that AP usually doesn't get to serve many clients Oct 27 09:59:44 I'm testing some beta firmware that ben has made to assist in debugging some of the issues the firmware has Oct 27 11:46:12 can uci option identifiers have a hyphen (-) ? Oct 27 13:16:18 ldir: option authoritative '1' Oct 27 13:17:05 ldir: I'll enable logging (option logdhcp '1') Oct 27 17:40:26 Is there any documentation available on the uci validation capability? Oct 27 18:00:14 hi, how do I compile a package using the OpenWrt SDK? Oct 27 18:03:15 "./scripts/feeds update && ./scripts/feeds install zerotier" does not find that package Oct 27 18:03:43 but I get a lot of mesaages ala ".find.bin: loadlocale.c:129: _nl_intern_locale_data: Assertion `cnt < (sizeof (_nl_value_type_LC_TIME) / sizeof (_nl_value_type_LC_TIME[0]))' failed." Oct 27 18:24:35 Why do kmods only build modules and not kernel-builtins? Oct 27 18:25:44 I mean, '*' means build a module while 'M' means build an opkg. It seems that '*' should mean to build it into the kernel while 'M' means build a module. Either that add a third option to Kbuild... Oct 27 18:25:57 anybody already in lisbon? Oct 27 18:41:11 mwarning: prepend all feeds commands with LC_ALL=C. Oct 27 18:47:06 ah, that helped :) Oct 27 19:02:51 exit Oct 27 21:51:31 nbd: Look, I don't mean to be a jerk, but wtf is this patch? +KBUILD_LDFLAGS_MODULE = -T $(srctree)/scripts/module-common.lds $(if $(CONFIG_PROFILING),,-s) Oct 27 21:52:08 This has taken me the past two hours to discover the cause of my modules STILL being stripped despite having the option turned off in my kernel config Oct 27 21:52:13 This crap has got to come out Oct 27 21:52:36 Sorry, I don't mean to be a jerk Oct 27 21:53:52 Also, the title is wrong. This strips more than "unneeded" as it also breaks kmemleak. You never strip with -s, you only use --strip-unneeded or similar Oct 28 00:11:21 * mangix grabs popcorn Oct 28 00:12:12 lol Oct 28 00:21:57 dansan: a hint, don't be so emotional. nobody is interested. **** ENDING LOGGING AT Sun Oct 28 02:59:59 2018