**** BEGIN LOGGING AT Sat Aug 04 02:59:59 2018
Aug 04 03:27:03 <mangix> sucks that ac gives poor performance
Aug 04 03:27:33 <DonkeyHotei> on which?
Aug 04 03:28:02 <mangix> everything
Aug 04 03:28:46 <DonkeyHotei> more performance than n
Aug 04 03:31:08 <salcedo> errr this 5ghz radio doesn't have DFS
Aug 04 03:31:23 <salcedo> neighbors need to get off my frickin lawn
Aug 04 03:32:04 <DonkeyHotei> intel?
Aug 04 03:32:08 <salcedo> ar750s
Aug 04 03:32:17 <salcedo> new gl-inet thingy
Aug 04 03:32:36 <DonkeyHotei> ath10k?
Aug 04 03:32:44 <salcedo> nah i think it's 9k
Aug 04 03:32:52 <DonkeyHotei> oh.
Aug 04 03:33:02 <salcedo> it shows it has radar detection but no DFS
Aug 04 03:34:51 <salcedo> hrm or maybe it's both?
Aug 04 03:46:21 <salcedo> aye
Aug 04 03:46:27 <salcedo> qca9887
Aug 04 03:49:22 <DonkeyHotei> that's ath10k
Aug 04 06:17:31 <xcom169> Protocol: Dual-Stack Lite (RFC6333) RX: 0 B (0 Pkts.) TX: 0 B (0 Pkts.) IPv4: 192.0.0.2/32
Aug 04 06:17:43 <xcom169> Why it's not calculating the ipv4 traffic?
Aug 04 06:21:59 <SwedeMike> xcom169: you need to provide more information.
Aug 04 06:22:58 <xcom169> SwedeMike: Sure, which commands I need to run?
Aug 04 06:23:35 <SwedeMike> xcom169: do you use ds-lite for your IPv4 internet connectivity?
Aug 04 06:24:13 <xcom169> SwedeMike: yes, my isp UPC is giving native ipv6, and ipv4 with ds-lite only
Aug 04 06:25:19 <SwedeMike> ok, and where did the above text come from regarding the traffic counters?
Aug 04 06:28:04 <xcom169> I saw it on Luci
Aug 04 06:28:10 <xcom169> Luci's dashboard
Aug 04 06:29:14 <xcom169> SwedeMike: I have 2 interfaces, ds-wan, and ds-wan6_4
Aug 04 06:29:48 <xcom169> ds-wan shows Zero, but ds-wan6_4 shows a lot of traffic
Aug 04 06:30:03 <xcom169> ds-wan6_4 Link encap:UNSPEC  HWaddr 2A-02-AB-82-00-80-00-04-00-00-00-00-00-00-00-00             inet addr:192.0.0.2  P-t-P:192.0.0.1  Mask:255.255.255.255           inet6 addr: fe80::584c:8fff:fe0f:73ae/64 Scope:Link           UP POINTOPOINT RUNNING NOARP  MTU:1280  Metric:1           RX packets:6356086 errors:0 dropped:0 overruns:0 frame:0           TX packets:3623362 errors:0 dropped:0 overruns:0 carrier:0           collision
Aug 04 06:47:17 <xcom169> hello all
Aug 04 06:48:28 <xcom169> In you Simple AdBlock, do you have currently Download errors?
Aug 04 07:41:38 <m4t> tmn505: gotta run, but some comments/suggestions were added to the pr (https://github.com/openwrt/openwrt/pull/1237)
Aug 04 07:48:42 <bela_> Anyone here use Adblock?
Aug 04 07:52:20 <bela_> I think adblock simple is broken in 18.06
Aug 04 07:52:51 <bela_> 1) libustream* is not mandatory for that
Aug 04 07:53:05 <bela_> 2) It block totally  normal websites, like www.index.hu
Aug 04 08:38:46 <ldir> jow: tried your fw3 dscp commit https://pastebin.com/aJzTBJNH - works a treat :-)
Aug 04 08:49:01 <f00b4r0> let's see if sgruszka's patches fix my mt7620 woes
Aug 04 11:30:45 <Hauke> jow: since some days I am seeing often a HTTP error 504 when cloing from the openwrt git, did something change?
Aug 04 11:40:43 <jow> Hauke: no, but maybe it is overloaded - will investigate
Aug 04 11:40:51 <jow> Hauke: consider using ssh transport to avoid it
Aug 04 11:41:35 <jow> rotanid: sure, will pick it to 17.01 as well
Aug 04 11:41:59 <guifipedro[m]> for when is scheduled 18.06.1 or even 18.06.2 ?
Aug 04 11:42:14 <guifipedro[m]> I probably need to fix something in a luci app and I would know  the timing
Aug 04 11:42:40 <jow> guifipedro[m]: 1.5-3 weeks
Aug 04 11:43:00 <jow> non builtin packages can be updated at any time and do not need coordination with the rleease schedule
Aug 04 11:46:57 <Hauke> jow: this was from the Intel network, it could be that it is caused by our proxy
Aug 04 11:47:10 <Hauke> if you do not see anything specail I will ask internally
Aug 04 13:05:17 <f00b4r0> well, 20GB later it would seem that these patches do improve the situation on mt7620
Aug 04 13:12:43 <DonkeyHotei> f00b4r0: which? because i'm in contact with somebody trying to track down a regression in master
Aug 04 13:12:54 <f00b4r0> DonkeyHotei: https://bugzilla.kernel.org/show_bug.cgi?id=82751
Aug 04 13:12:58 <f00b4r0> the 3 listed at the bottom
Aug 04 13:13:06 <f00b4r0> from Stanislaw Gruszka
Aug 04 13:13:25 <f00b4r0> needed a bit of massaging to apply on top of backport 2017-11-01
Aug 04 13:14:59 <f00b4r0> i still see the error message, but the connection holds regardless. I've been unable to crash it over the past 3.5 hours, whereas it used to die in a matter of minutes before
Aug 04 13:20:52 <DonkeyHotei> you should add that as a comment on that page
Aug 04 13:21:27 <f00b4r0> i need to register
Aug 04 13:21:29 <f00b4r0> first
Aug 04 13:22:57 <f00b4r0> in the meantime I have emailed sgruszka
Aug 04 13:34:22 <DonkeyHotei> or you could comment on his github, idk
Aug 04 14:34:23 <DonkeyHotei> within the past 90 minutes, two different people in #openwrt have asked about 18.06 images for two different ar71xx devices with 4MB flash that don't have downloads in -generic nor -tiny
Aug 04 14:40:30 <karlp> tell them the same thing we told ar2317 users with 4mb flash in 17.xx?
Aug 04 14:46:37 <DonkeyHotei> karlp: no more images?
Aug 04 14:51:34 <Borromini> build logs should show if image is too big or if something else is the matter no?
Aug 04 14:51:43 <Borromini> if it's the latter it might be easily fixed.
Aug 04 14:53:20 <DonkeyHotei> TP-Link WR703n and Dlink DIR 615 C1
Aug 04 14:54:07 <DonkeyHotei> and i misread the timestamp -- the tp-link was asked about almost 3hrs ago
Aug 04 14:59:37 <salcedo> i wonder how big dnscrypt2 is
Aug 04 15:00:09 <salcedo> it relies on golang - but once it's compiled it doesn't need golang anymore
Aug 04 15:03:19 <f00b4r0> of course. All go executables are statically linked. That's a language "feature"
Aug 04 15:17:40 <Borromini> DonkeyHotei: the 615 E4 revision is/was too big already according to the forum so i reckon C1 won't fit anymore either.
Aug 04 16:00:33 <salcedo> hrm wonder why dropbear isn't bumped to 2017.76
Aug 04 16:01:03 <Borromini> salcedo: you mean the pr sitting in github?
Aug 04 16:01:12 <salcedo> Borromini: i'm not aware of it. link please?
Aug 04 16:01:30 <salcedo> ed25519 is broken because dropbear expects the client to append @libssh.org
Aug 04 16:05:19 <Borromini> weird
Aug 04 16:05:26 <Borromini> i just pulled it this morning
Aug 04 16:05:31 <Borromini> s/pulled/grabbed/
Aug 04 16:06:09 <Borromini> nevermind wrong repo
Aug 04 16:06:12 <Borromini> https://github.com/openwrt/openwrt/pull/915
Aug 04 16:06:26 <Borromini> you'll need https://github.com/openwrt/openwrt/pull/786 first
Aug 04 16:06:45 <Borromini> i've been running dropbear 2018.76 for months on multiple devices, with 18.06
Aug 04 16:09:57 <Borromini> salcedo: sorry. i just saw you were referring to 2017.76... and this is obviously newer, so no wonder you didn't find any PR
Aug 04 16:10:08 <Borromini> that being said, 2018.76 works just fine.
Aug 04 18:06:03 <mangix> fascinating
Aug 04 18:06:50 <mangix> so if roaming is enabled, someone can crack the AP password without the handshake
Aug 04 18:06:57 <mangix> I must test this
Aug 04 18:08:18 <DonkeyHotei> ?
Aug 04 18:10:00 <mangix> https://hashcat.net/forum/thread-7717.html
Aug 04 19:22:25 <m4t> mkresin: i think i understand the difference between ath79 DEFAULT_PACKAGES and rs/rspro DEVICE_PACKAGES. DEFAULT_PACKAGES are =y by default if building *only* the specific target, but can be unset or set to =m. DEVICE_PACKAGES are what are unconditionally placed in the rootfs.bin when CONFIG_TARGET_PER_DEVICE_ROOTFS=y and cannot be set to anything except =y or =m. sound right?
Aug 04 19:26:34 <m4t> if not, then there might be a bug ;(
Aug 04 22:13:13 <karlp> iirc, config_target_per_device can't be used with mnuconfig though
Aug 04 22:16:10 <karlp> you have to use them with config stubs,
Aug 04 22:37:47 <huaracheguarache> has anybody here tried out blogic's ath10k commits in his staging tree?
Aug 04 22:40:04 <huaracheguarache> i get the following error when i try to download the mac80211 package: The requested URL returned error: 404 Not Found
Aug 04 22:42:44 <karlp> Can't locate Time/HiRes.pm wat? when did that come in?
Aug 04 22:42:58 <karlp> is this more fallout from gtime time blah incompat?
Aug 04 22:43:10 <stintel> huaracheguarache: indeed, the tarbal is not available
Aug 04 22:43:16 <karlp> jow: yeah, I don't have that perl module.
Aug 04 22:43:24 <karlp> but I had time the whole time.
Aug 04 22:43:28 <stintel> huaracheguarache: I am running it on both my APs for ~27h
Aug 04 22:43:35 <stintel> and neither of them has pooped itself
Aug 04 22:43:45 <stintel> so it's looking promising so fare
Aug 04 22:43:47 <stintel> far*
Aug 04 22:43:51 <huaracheguarache> and usually they do for you?
Aug 04 22:43:54 <karlp> jow: what do you mean about time being "rarely installed" ?
Aug 04 22:44:16 <ntd> stintel, i'm getting the impression from the last couple of days of messages in here that ath10k-ct (qca988*) isn't all that?
Aug 04 22:44:36 * karlp backs up one commit.
Aug 04 22:45:01 <karlp> ntd: the generally common discussion here is that wireless sucks overall, in general.
Aug 04 22:45:08 <stintel> huaracheguarache: since I have my Gemini PDA (mediatek), in combination with my huawei phone, intel iwl 8265 laptop and samsung tablet, it crashed within ~24h
Aug 04 22:45:17 <stintel> see flyspray #333
Aug 04 22:45:30 <stintel> ntd: it wasn't for me, when I tried
Aug 04 22:45:37 * huaracheguarache takes a look
Aug 04 22:46:11 <stintel> stable, yes, but slow af and because of that barely unusable
Aug 04 22:46:29 <stintel> barely usable
Aug 04 22:46:40 <stintel> sorry, had some drinks tonight
Aug 04 22:46:46 <ntd> and ath isn't playing nice re driver development?
Aug 04 22:47:24 <stintel> qca is dead silent on bugzilla.kernel.org
Aug 04 22:47:51 <stintel> I'll be doing tests with -ct again
Aug 04 22:47:57 <huaracheguarache> color me surprised
Aug 04 22:48:26 <huaracheguarache> stintel: are you running 802.11w?
Aug 04 22:48:31 <stintel> but I'm traveling tomorrow and it probably will be in ~10d that I will have enough time do some decent testing
Aug 04 22:48:32 <stintel> yes
Aug 04 22:48:36 <huaracheguarache> ok
Aug 04 22:48:59 <stintel> 11w is a requirement for me
Aug 04 22:49:22 <huaracheguarache> i've never tried it
Aug 04 22:49:41 <stintel> without it every script kiddie can render my network unusable
Aug 04 22:49:54 <huaracheguarache> so it has happened?
Aug 04 22:50:25 <stintel> I've tested it myself
Aug 04 22:50:39 <stintel> it's so easy that it's simply unacceptable to have a network without 11w
Aug 04 22:51:23 <ntd> iirc stintel made a herculean effort in getting decent 11w support in LEDE
Aug 04 22:51:33 <stintel> nah
Aug 04 22:51:40 <stintel> I just submitted a few patches
Aug 04 22:51:43 <pkgadd> huaracheguarache: ieee80211=optional is easy, and I'd strongly recommend it, making it mandatory unfortunately isn't (way too many clients that still can't do it)
Aug 04 22:51:59 <ntd> stintel, iirc you also submitted some patches to networkmanager?
Aug 04 22:52:08 <huaracheguarache> pkgadd: do i need full wpad?
Aug 04 22:52:13 <ntd> yes
Aug 04 22:52:15 <huaracheguarache> ok
Aug 04 22:52:25 <pkgadd> huaracheguarache: I think so, but... I never install wpad-mini
Aug 04 22:52:29 <huaracheguarache> i'll try it out
Aug 04 22:52:46 <ntd> stintel, how did that go, were they accepted and in which version were they included?
Aug 04 22:52:53 <stintel> ntd: I laid the groundwork for that
Aug 04 22:53:02 <stintel> but I didn't finish those patchs into acceptable state
Aug 04 22:53:39 <stintel> ntd: but hostapd/wpa_supplicant and networkmanager have now support for it, afaik
Aug 04 22:54:06 <huaracheguarache> apparently many things are optional in the new wpa3 standard
Aug 04 22:54:19 <ntd> in which version of NM was it introduced?
Aug 04 22:56:17 <stintel> ntd: 1.8
Aug 04 22:57:11 <stintel> but it needs support on the supplicant side as well, and I don't think there's been a new release after those commits
Aug 04 22:57:26 <stintel> so probably only when distros backported it
Aug 04 23:01:21 <stintel> http://w1.fi/cgit/hostap/commit/?id=3cdb4ac074f76accf24a51d143db545afad2c90b
Aug 04 23:02:18 <stintel> yep, 2.6 is current release and is from 2016
Aug 04 23:03:09 <Monkeh> Well that's painful and sucky
Aug 04 23:03:28 <ntd> and ubuntu xenial was released with a networkmanager that wanted to do STA MAC randomization, but a wpasupplicant that couldn't (2.4)
Aug 04 23:03:41 <stintel> ubuntu ... no comment
Aug 04 23:04:10 <stintel> feel free to +1 https://www.spinics.net/lists/hostap/msg04449.html
Aug 04 23:05:14 <stintel> I just did
Aug 04 23:27:55 <jow> karlp: never had gnu time on any system
Aug 04 23:28:06 <jow> will take a look at time::hires tomorrow
Aug 04 23:28:18 <jow> if it were for me, we could rip out that entire time crap completely
Aug 05 00:52:10 <karlp> finewith me.
Aug 05 00:52:19 <karlp> without tooling that shows the results, ti's just noise
Aug 05 02:16:58 <Mangix> Stintel: roaming is also problematic if you use that :P
Aug 05 02:17:45 <Mangix> Oh btw. The 802.11w crashing happens for me with all ath10k devices with all firmwares.
Aug 05 02:18:20 <Mangix> I've also tested QCA6174. The firmware crashes.
Aug 05 02:19:11 <Mangix> ath10k-ct has a workaround where instead of going down with the firmware it will
Aug 05 02:19:30 <Mangix> restart it, but still annoying.
**** ENDING LOGGING AT Sun Aug 05 02:59:59 2018