<stintel> like how to properly integrate all of this in OpenWrt
<stintel> I'll publish my local /60 instead, see if that sticks
<neggles> iirc thread doesn't have to only use ULA though?
<neggles> thread border routers can request a DHCPv6 PD and hand that out to devices
<stintel> yep
<stintel> but right now it's ULA
<stintel> I might need to ask for a /56 instead of a /60 :P
<neggles> shouldn't a /64 be plenty for a single thread network? :P
<stintel> I'm running out of /64s in this /60
<neggles> ah wait, misread, yeah you should definitely ask for a /56
<neggles> handing out /56s to even home users seems to be what most ISPs are converging on
<stintel> yeah this is just 2 tunnels with BGP to local hacker space
<neggles> ah
<stintel> they asked if /60 was enough, back then it was for sure, so I said yes
<neggles> (/56s is partly because the way cisco/arista/juniper lay out their TCAM, you can fit more /56s than /60s in the same space)
<dwfreed> oh, I do have a /56 from cox
<stintel> all bird2 here :)
<dwfreed> I think I had to set something special to make that work, though
<stintel> ugh something is leaking ULAs though
<neggles> a lot of arista's current gen stuff is bird2 under the hood
<neggles> but the FIB is still in... sortahardware
<dwfreed> ah, yeah, specifically specifying reqprefix 56, otherwise it would just get a /64
<neggles> yeah most routers / DHCPv6 clients seem to default to just asking for a /64
<neggles> my ISP gives me an address for on-link and a /56, or a /48 if you ask for it
<dwfreed> well, the way it works on openwrt, odhcp6c doesn't specify a prefix length (just the option for getting a prefix), and then my ISP just says "here's a delegated /64"
<neggles> yeah, same here
<neggles> have to send prefix length hint
<dwfreed> but if you set option reqprefix 56, then it'll hand out a /56 just fine
<dwfreed> (and that's the largest I can get)
<stintel> uff where are these fucking ULAs on my workstation coming from
<dwfreed> stintel: leftovers that haven't expired yet?
<stintel> dwfreed: I just removed 10
<stintel> now they're back
<neggles> something doing silly autoconfig shenanigans?
<stintel> those thread thingies are in different VLANs ... not sure wtf is going on
<stintel> and especially since there are like 10
<dwfreed> tcpdump -nvvvei eth0 icmp6
<neggles> is it pcap time?
<neggles> it *is* pcap time
<stintel> yeah I was wiresharking
<neggles> come to think of it, these could be fun to put openwrt on at some point (especially with thread support) https://wiki.pine64.org/wiki/Ox64
<stintel> aaand there kwin_x11 did an amber heard
<neggles> the upstream/linux support in general still needs an awful lot of work, but
<stintel> inet6 2001:67c:21bc:20::10/128 scope global dynamic noprefixroute
<stintel> valid_lft 6318sec preferred_lft 2718sec
<stintel> inet6 2001:67c:21bc:20:52bf:b29c:6e1e:7c70/64 scope global dynamic noprefixroute
<stintel> valid_lft 2591922sec preferred_lft 604722sec
<stintel> fuck
<stintel> wrng paste buffer
<stintel> ip -6 addr | grep fd | awk '{ print $2 }' | xargs -n1 ip -6 addr del dev vlan16
<stintel> aaand they're back
<stintel> they're coming from a neighbor solicitation from my workstation itself
* stintel stabs NetworkMangler
<neggles> I've long since given up on keeping the kshjillion ipv6 ULAs away personally :P
<dwfreed> stintel: neighbor solicits don't make ULAs; they're intended to verify an address isn't already in use (DAD)
<stintel> dwfreed: it's NM
<dwfreed> NetworkMangler could be doing its own ULAs, though
<stintel> NM restored the ULAs it previously received when I was struggling with these sunxi devices
<stintel> back then they were on vlan16
<stintel> NM restored the addresses, which then triggered NS
<neggles> heh was going to say "sounds like something's cached them and is putting them back"
<neggles> NM strikes again
<stintel> sudo nmcli connection up ovs-interface-vlan16
<stintel> done
<stintel> fucking ULAs
Tapper has quit [Ping timeout: 480 seconds]
<neggles> IPv6: it's a good thing we have so many addresses, 'cause you're going to need at least five
<stintel> :P
<stintel> looks like the /60 route sticks
<stintel> gonna let this scp run just for testing
<stintel> 20 packets transmitted, 19 received, 5% packet loss, time 19030ms
<stintel> rtt min/avg/max/mdev = 126.242/308.896/526.042/118.421 ms
<stintel> I know it's not the intended use, but still, with matter being so huge that some mfgs are unable to add support to some existing devices because not enough storage ...
<stintel> firmware updates need to happen
<stintel> and if that can cause >500ms latencies ... hmmm
<stintel> maybe I should add a another thread client and ping that while scping to the other
<stintel> anyway, I think that's enough for today, possibly for this week, and maybe even for this year
<dwfreed> keep in mind the maximum packet size on 802.15.4 is 127 octets
<dwfreed> at the same time, a peak latency of only 500ms really isn't *that* bad
<dwfreed> especially if you're transferring a file across at the same time
<stintel> now if only I could figure out how to trick the Eve app into thinking I have a crapple thread network
<stintel> so that I can upgrade my eve energy and eve motion sensor, get the thread key, and try joining them in my thread network
<stintel> maybe I can run homebridge
<stintel> it doesn't have thread support but I'm seeing mdns requests for homekit stuff
<stintel> hmmmz
<stintel> also seeing _meshcop._udp.local. queries but no responses
<stintel> maybe I don't know how mdns works
<stintel> Installing Homebridge and user plugins, please wait...
<stintel> on container startup *facepalm*
dansan has quit [Ping timeout: 480 seconds]
<neggles> stintel: homekit has Matter support now and iirc Matter runs on Thread as well as wifi
<neggles> homepods, appletvs can be / are matter routers/gateways
<stintel> I know, but I don't have homepod nor apple tv
<stintel> and I apparently need to one of those before the Eve app will allow me to upgrade the firmware of my eve energy or eve motion sensor to the firmware that supports matter
<stintel> they're working on an android app also, so this firmware should drop the homekit requirement
<stintel> just a bit annoying that I need some homekit/thread device to do the firmware upgrade
<stintel> good thing I didn't go bonkers and ordered 20 eve energy with black friday deal
<stintel> openwrt-sunxi-cortexa53-olimex_a64-olinuxino-emmc-squashfs-sdcard.img.gz 100% 15MB 10.3KB/s 25:00
<stintel> at least something is working
<neggles> stintel: heh, "you can upgrade to the Matter firmware and use matter with Samsung SmartThings, Google Home, and HomeKit!"
<neggles> further down the page: "ur gonna need an appletv or a homepod mini and an iphone to update to this firmware"
<stintel> I have an ipad with the required version of ipados and have the testflight version of the eve app
<stintel> but no homepod or appletv
<neggles> > All Eve products that are currently available in retail work with Apple Home only. So in order to update them to Matter, you have to use the Eve app for HomeKit.
<neggles> hmm
<neggles> someone will probably work out how to DIY a homekit-compatible thread border router sooner rather than later
<stintel> I guess I can get a homepod or appletv :)
<stintel> unlinked the motion sensor a few times already and rejoined but keeps connecting via bluetooth
<stintel> oh well
<neggles> homepod mini is quite cool and good as a speaker, but if you don't trust apple with a microphone in your house the appleTV
<neggles> more exxy though but not too bad
<neggles> these guys claim to have worked it out https://hoobs.com/pro/ but *ew*
<neggles> US$400, not available until march, and it's almost definitely an RK3566 in there
<stintel> and proprietary iirc
<neggles> yeah
<neggles> no thanks
<stintel> yeah nope
<stintel> then rather get an apple tv 4k and see if I can tcpdump and figure out what's missing
<stintel> or just use it for the upgrade
<neggles> if you've got an apple store nearby that's a viable option
<neggles> 14 day "yes" return policy and all
<stintel> and then hope the code / qr I will get after upgrading the firmware is usuable to join an openthread network
<stintel> hmmm I can "upgrade" my home in the home app
<stintel> "The accessories in My Home can now take advantage of improved performance and reliability."
<stintel> whatever that means :P
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/25/builds/198) of `sunxi/cortexa7` failed.
<stintel> hmmm the eve energy strip seems to be updating now
<stintel> or is that just "polling"
<stintel> that thing just connected to the wifi I was connected to at that time
<stintel> typical apple philosophy "we know better than user"
<stintel> no message telling me that the device will connect to my wifi, no question to ask if this was actually the network I wanted to connect it to
<stintel> https://support.apple.com/en-us/HT207057 > Set up iPad as a home hub
<stintel> hmmmm
danitool has quit [Remote host closed the connection]
<stintel> iPad is not supported as a home hub on the new, more reliable, and more efficient Home architecture, which is available as a separate upgrade in the Home app.
<stintel> grmbl
<stintel> and I just did that upgrade probably
<stintel> fml
<stintel> ok support for that was dropped in ipados 16
<aiyion> stintel: good morning. Trains have been taken, birthdays been celebrated, time to test the patch.
<stintel> and needed 16.1 iirc to be able to run the eve preview app
<stintel> aiyion: morning :)
goliath has quit [Quit: SIGSEGV]
<stintel> I'm gonna head to bed though
<aiyion> On the train the build wouldn't compile but hang on the musl stage -.-'
<aiyion> good choice.
<aiyion> will do once I've upgrade to your staging tree.
<stintel> had enough of the apple wiseassery for today :)
minimal has joined #openwrt-devel
<aiyion> stintel: compilation sucessful, sysupgrade as well. 3/3 reboots without problems. Will leave this rebooting till morning and share the stats, but I'd say this is ready to merge.
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/62/builds/199) of `pistachio/generic` failed.
nkko has joined #openwrt-devel
<nkko> I'm having issues with BCM963138 routers, they seem to be close to bcm53xx, but not? There is mainline support for them, but I just cannot get it to boot (vendor kernel works, I was also able to build a neer cfe release)
<nkko> I've checked the ELF sections, etc., ran diff on the vendor source, I just cannot figure out what could be causing it
<nkko> *newer
<nkko> I don't think it'd be related to the device tree, though, surely (cmplzma is probably stripping the appended dtb, anyway)
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/71/builds/198) of `bcm63xx/generic` failed.
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/50/builds/199) of `apm821xx/sata` failed.
gromero__ has quit [Ping timeout: 480 seconds]
nkko has quit [Quit: leaving]
minimal has quit [Quit: Leaving]
schwicht_ has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
tSYS has quit [Quit: *squeak*]
tSYS has joined #openwrt-devel
<owrt-2203-builds> Build [#195](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/49/builds/195) of `mediatek/mt7622` failed.
srslypascal is now known as Guest109
srslypascal has joined #openwrt-devel
<champtar> ynezz: thanks for the links, I added a comment to the ticket, Turris Omnia is broken for me with latest master / 5.15 with a pretty basic config
Guest109 has quit [Ping timeout: 480 seconds]
<owrt-snap-builds> Build [#749](https://buildbot.openwrt.org/master/images/#builders/54/builds/749) of `ramips/mt7620` failed.
<owrt-2203-builds> Build [#192](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/60/builds/192) of `layerscape/armv8_64b` failed.
<owrt-2203-builds> Build [#201](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/41/builds/201) of `apm821xx/nand` failed.
<owrt-2203-builds> Build [#209](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/11/builds/209) of `at91/sama5` failed.
<owrt-2203-builds> Build [#197](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/69/builds/197) of `gemini/generic` failed.
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/32/builds/200) of `realtek/rtl838x` failed.
eigma has quit [Remote host closed the connection]
philipp64 has quit [Read error: Connection reset by peer]
philipp64 has joined #openwrt-devel
jlsalvador has quit [Remote host closed the connection]
jlsalvador has joined #openwrt-devel
srslypascal has quit [Quit: Leaving]
cbeznea has joined #openwrt-devel
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/46/builds/200) of `omap/generic` failed.
valku has quit [Quit: valku]
dangole_ has quit [Ping timeout: 480 seconds]
srslypascal has joined #openwrt-devel
csrf has joined #openwrt-devel
<owrt-2203-builds> Build [#197](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/59/builds/197) of `x86/64` failed.
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/39/builds/199) of `layerscape/armv7` failed.
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/74/builds/199) of `mpc85xx/p2020` failed.
torv is now known as Guest121
torv has joined #openwrt-devel
Guest121 has quit [Ping timeout: 480 seconds]
<Znevna> LuCI has no support for setting phyXtpt as a LED trigger, does it?
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/52/builds/198) of `bcm27xx/bcm2709` failed.
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/23/builds/199) of `ath79/nand` failed.
csrf has quit [Remote host closed the connection]
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/66/builds/199) of `zynq/generic` failed.
<owrt-2203-builds> Build [#255](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/1/builds/255) of `bcm27xx/bcm2711` failed.
<Znevna> did I miss any changes that broke Channel Analysis ?
<Znevna> x,x
felix has quit [Ping timeout: 480 seconds]
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/72/builds/198) of `ath79/generic` failed.
<owrt-2203-builds> Build [#197](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/70/builds/197) of `sunxi/cortexa53` failed.
<dhewg> Znevna: iwinfo got quite some patches, which the analysis uses, but that worked as I tested the changes a few weeks ago
chuck48 has joined #openwrt-devel
<dhewg> now I see js errors on that page using the ff debugger console
<dhewg> Uncaught (in promise) ReferenceError: diff is not defined
<dhewg> which seems to be a bug in luci commit 75dcb097547257d654e9b7f7d58578b9210968e5
rmilecki has quit [Ping timeout: 480 seconds]
<dhewg> this seems to fix it https://dpaste.com/ET93AU5T5
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/68/builds/198) of `mediatek/mt7623` failed.
<Znevna> nice
<Znevna> that fixes it :D
<Znevna> thank you
<dhewg> but no idea if that's the proper fix
<dhewg> do you mind opening a luci issue about it?
<Znevna> I'll try x.x I'm new to this git thing
<dhewg> an issue is good enough, no pr
<aiyion> stintel: gm. Tested another 17 times, so in total 20/20 boot procedures without problems on your current staging master.
<Znevna> k done
<owrt-2203-builds> Build [#206](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/10/builds/206) of `realtek/rtl930x` failed.
Borromini has joined #openwrt-devel
chuck4888 has joined #openwrt-devel
Tapper has joined #openwrt-devel
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/36/builds/200) of `sunxi/cortexa8` failed.
chuck48 has quit [Ping timeout: 480 seconds]
robimarko has joined #openwrt-devel
danitool has joined #openwrt-devel
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/51/builds/198) of `x86/legacy` failed.
<owrt-2203-builds> Build [#209](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/15/builds/209) of `at91/sam9x` failed.
<owrt-2203-builds> Build [#202](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/17/builds/202) of `ipq40xx/generic` failed.
xes has quit [Quit: bye..]
<Borromini> do the buildbots clean the build environment on every run? I suppose so?
<dhewg> which is prolly sdk related again
<dhewg> I never used the sdk, but it's just a headache if buildbot/ci fails when everything works locally, it's a reoccuring issue
<robimarko> Well, there have been weird failures recently
srslypascal is now known as Guest138
srslypascal has joined #openwrt-devel
Guest138 has quit [Ping timeout: 480 seconds]
srslypascal is now known as Guest139
srslypascal has joined #openwrt-devel
tomn has quit [Quit: leaving]
tomn has joined #openwrt-devel
srslypascal has quit [Remote host closed the connection]
srslypascal has joined #openwrt-devel
Guest139 has quit [Ping timeout: 480 seconds]
robimarko has quit [Remote host closed the connection]
robimarko has joined #openwrt-devel
<tmn505> WRT build error; AFAICS we don't build gzip in buildroot but it tries to run staging_dir/host/bin/gzip, is the 47df168 change bogus? < oliv3r[m]
<dhewg> include/prereq-build.mk checks for gzip, and hence creating a shortcut in staging/host/bin
<dhewg> and I guess that doesn't work with the sdk
<tmn505> hmm, I don't have that symlink in my staging_dir.
<tmn505> ok, that's recent addition and I didn't rebase my tree yet
<Znevna> jkolp6
<Znevna> whops, sorry
<tmn505> so f595beb should be backported to to 22.03
<tmn505> hauke: ^^
<dhewg> oh haha, yeah that makes sense
<stintel> aiyion: done
<stintel> oops I added the wrong closes tag
<stintel> rectified and dupe closed
<stintel> cool. already 2 issue report since the template is in place
<f00b4r0> yay :)
<stintel> and 2 feature requests in the forum
<f00b4r0> looks like it's working then :)
<stintel> yep
felix has joined #openwrt-devel
<stintel> and it looks like my OpenWrt Thread "client" is still accessible
<stintel> yay
<Borromini> stintel: this? 'I am reporting an issue for OpenWrt, not an unsupported fork.'
<stintel> Borromini: what do you mean?
<stintel> that particular element is only a *very* small part of the issue template
<Borromini> ok :)
<owrt-2203-builds> Build [#197](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/40/builds/197) of `mvebu/cortexa53` failed.
<Borromini> but yes, reports like https://github.com/openwrt/openwrt/issues/11588
<stintel> Borromini: looks like a report with enough info ?
<Habbie> 1 task done
<Habbie> :)
<stintel> I mostly hope the page *before* reporting an actual issue makes people think twice about reporting a bug, and reach out to the community instead
<stintel> yeah
<Habbie> it's even worse in pdns, where some of the checkboxes are not required (that's a PR thing, though)
<stintel> we discussed this in the PR, it's being worked on to make checkboxes not tasks unconditionally
<stintel> by gh
<Habbie> ah, on the github side?
<Habbie> ok nice
<stintel> hold on
<stintel> there ya go
<Habbie> ta
<stintel> yw
felix has quit []
goliath has joined #openwrt-devel
<stintel> anyone know how to contact shibboleth ?
gromero has joined #openwrt-devel
gromero has quit []
gromero has joined #openwrt-devel
<Borromini> stintel: what i meant was 'is this an example of the result', so I suppose yes :)
<Borromini> shibboleth joins IRC now and then AFAIK
<stintel> if they join and I'm not here, ask him to pastebin logread right after a failed sysupgrade (one that didn't keep settings)
<stintel> I thought this was about f2fs threshold but it's totally unrelated
<stintel> the device has 8MB flash so there's no way f2fs is used
dangole_ has joined #openwrt-devel
<hauke> tmn505: thnaks for the hint, I reverted the change again
<Borromini> oftc dumped memoserv right?
<stintel> dwfreed: can you answer that? ^ :)
<stintel> dangole_: that aliases solution for the bpi-r2, what's upstream's take about that?
<stintel> it's also a problem for wifi on some devices because we use option path 'platform/soc/1c10000.mmc/mmc_host/mmc0/mmc0:0001/mmc0:0001:1'
<stintel> so on my a64-olinuxino I have 2 wireless devices in /etc/config/wireless
dangole_ is now known as dangole
<stintel> although even with alias it still didn't work, I guess because the rtl8723bs driver is crap
<stintel> reminds me of some raspberry pi
<dangole> stintel: is it the same 'platform/soc/1c10000.mmc' which offers more than one mmc host?
<stintel> option path 'platform/soc/1c10000.mmc/mmc_host/mmc0/mmc0:0001/mmc0:0001:1'
<stintel> option path 'platform/soc/1c10000.mmc/mmc_host/mmc1/mmc1:0001/mmc1:0001:1'
<dangole> dangole: ok, so that's different from mediatek SoCs, then it's up to the driver in which order they show up...
<dangole> dangole: because it's just a single call to drivers probe() function from Linux kernel...
<stintel> ah so the aliases won't solve this
gladiac has quit [Quit: k thx bye]
robimarko_ has joined #openwrt-devel
gromero has quit [Remote host closed the connection]
gromero has joined #openwrt-devel
schwicht has joined #openwrt-devel
felix has joined #openwrt-devel
robimarko has quit [Ping timeout: 480 seconds]
<opty> what do you think about iwinfo build on other systems using https://termbin.com/1j8m and e.g. 'make CFLAGS="-DNO_UCI -DNO_UBUS -DNO_UBOX -DNO_LUA -DUSE_STDBOOL_H -I/usr/include/libnl3 -D_GNU_SOURCE" FPIC=-fPIC BACKENDS=nl80211 SOVERSION=20210430'? (no essid, encryption and expected throughput yet, though)
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/54/builds/200) of `x86/generic` failed.
goliath has quit [Quit: SIGSEGV]
danitool has quit [Remote host closed the connection]
<stintel> aaand fedora work laptop shit the bed again
<robimarko_> What did poor Fedora do?
<Znevna> this channel is bleeding
<Znevna> oh x.x i was two pages up when only 'failed' was on the screen
gromero has quit [Remote host closed the connection]
gromero has joined #openwrt-devel
<stintel> robimarko_: hard lockup the machine
<stintel> it's been doing that a few times lately :(
<oliv3r[m]> How can I test if 'port_mdb_del' works properly?
<stintel> hmmmm my ot-cli-{f,m}td firmwares don't seem to have the joiner command :/
<Strykar> stintel: work chooses the distro for you
nkko has joined #openwrt-devel
<robimarko_> I am quite happy with Fedora
<nkko> I managed to join the general chan instead of the dev chan
<robimarko_> Been using it for years on both my workstations and notebooks
<nkko> So, has anyone gotten mainline linux to run on the bcm963138?
<nkko> I've been going through the cfe code, and I think I know what's going on now
<stintel> Strykar: unfortunately
<owrt-2203-builds> Build [#197](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/31/builds/197) of `bcm27xx/bcm2710` failed.
<owrt-2203-builds> Build [#206](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/6/builds/206) of `mpc85xx/p1020` failed.
gromero_ has joined #openwrt-devel
gromero has quit [Remote host closed the connection]
goliath has joined #openwrt-devel
ruck has joined #openwrt-devel
Tapper has quit [Ping timeout: 480 seconds]
minimal has joined #openwrt-devel
<aiyion> stintel: nice thanks. My time is limited today, but maybe I'll find another hour later this evening.
ruck has quit [Quit: Page closed]
valku has joined #openwrt-devel
<Znevna> I'm gonna attempt to play with github, don't kill me.
<Borromini> that's what it's for. Real men use git send-email }:-)
<gromero_> lol
<Znevna> meh
<gromero_> that's a good one, but I'd like to see one specific to Github :P
<Znevna> If I find a good one I'll share ^^
cbeznea has quit [Quit: Leaving.]
<Borromini> :P
<mrnuke> robimarko_: How come developers use fedora, and everyone else hates it?
<nkko> mrnuke: bleeding edge?
<mrnuke> nkko: That would be fedora rawhide... I'm not brave enough to step there
<stintel> I'm on Fedora 36, didn't upgrade to 37 yet to avoid bleeding edge breakage
<stintel> I tried some sysrq at some point but didn't respond
<stintel> if I can trigger a panic then check pstore maybe I can report it but ... meh
<Habbie> i've pondered fedora for my desktop - two upgrade days a year, very fresh software the 364 other days
gladiac has joined #openwrt-devel
<mrnuke> stintel: One of my machines I recently updated was on fedora 30. Now I have a mix of F34, F35 and F37. :p
<mrnuke> once the kernel gets stable enough on a given machine, It's easy sailing with any non-ubuntu distro :)
gladiac is now known as Guest163
gladiac has joined #openwrt-devel
<stintel> yeah I suspect it's the kernel doing something silly
<stintel> oh well
chuck4888 has quit [Remote host closed the connection]
<robimarko_> It has to be the kernel
<robimarko_> I like Fedora as its usually up to date but actually rarely breaks
Guest163 has quit [Ping timeout: 480 seconds]
<stintel> it was breaking all the time when colleagues convinced me to try it back in 2007
<stintel> and every so many years I try it again, and it hasn't changed
<opty> f35 xD
<stintel> I'll put it on the same instability level as SLED/SLES :P
<robimarko_> stintel: Honestly, so far I think I had one break with Fedora in like 5+ years
<stintel> sigh, thread. now I need a new CLI tool for commissioning a new device
<robimarko_> And I am the guy that usually even updates to beta versions
<stintel> I guess it's the hardware then (thinkpad x1 carbon g9 iirc)
* Borromini is happy enough with Debian testing
<robimarko_> Well, I am using it on T490, Huawei Matebok 16 and on my workstation which is 5950X
<f00b4r0> linux desktop. /me rolls eyes xD
* f00b4r0 hides
<stintel> I don't remember the last time my workstation crashed
<robimarko_> Neither
<robimarko_> And I did some crazy stuff
<stintel> my work macbook ... acting up all the time
<stintel> yeah I do crazy shit on my workstation too :P
<f00b4r0> stintel: it can tell you hate it :}
<stintel> and this is Gentoo
<Znevna> my storage server is missing a capacitor from the CPU, still doesn't crash.
<stintel> thread community on github does seem very active
<f00b4r0> so thread is a replacement to matter, itself an alternative to homekit, did I get this right?
<stintel> no
* f00b4r0 is a bit confused by the iot space
<f00b4r0> more than a bit then :)
<Borromini> you're in the same boat as me then :P
<stintel> matter works over (on top of) Thread, WiFi, Ethernet
<stintel> well, or just over IP
<stintel> so all your IoT devices will have an IP, and you use the matter standard to tell e.g. your lamp: switch to color X, over IP
<stintel> I kind of like that idea, you can use ULA and you don't have to worry for the Internet to be able to access your IoT devices, but it does allow simple monitoring with ICMP "hey is my lamp still on the network"
<stintel> and OpenThread should fix a lot of the shortcomings of Zigbee and Zwave
<stintel> and iiuc we will finally be able to have a redundant home automation controller
<stintel> something that simply isn't possible with zwave or zigbee
<stintel> which makes it ... super mega annoying
<Borromini> i haven't got any IoT stuff in here but my brother pointedly observed our screens (when we install them) will need some form of control, and that's often domotics
<Borromini> so i guess we'll need to have *some* form of that stuff anywya
<Borromini> * anyway
<stintel> I've 20 zigbee lamps, 2 zwave lamps, 6 zwaves smart wall plugs, 13 esp32s, an esp-based smart wall plug, and a few raspberry pis doing IoT stuff
<stintel> and some more zigbee lamps to be added
<stintel> but if I can get redundant thread network with openwrt border routers and then use matter from whatever device with the right tools ...
<stintel> I will replace *everything*
<Borromini> :)
<f00b4r0> e-waste :P
<stintel> nah, donate to local hacker lab
<stintel> or my parents
<stintel> or a friend
<Borromini> i guess he'll find someone to relieve him from that stuff
<f00b4r0> i kinda like the idea that I can still turn all my lamps on after an EMP :)
<stintel> f00b4r0: just flip the switch
<f00b4r0> simply by getting my ass up to the switch
<stintel> but if EMP happens ... I guess we've other things to worry about :P
<f00b4r0> fair enough :)
<stintel> but I like that my back row of spots switches off and my front row of spots switches to yellow / low brightness whenever I start watching some movie or series
<f00b4r0> I simply turn off the lights
<f00b4r0> i guess I have simpler needs ;}
<Borromini> we have a dimmer B-)
<stintel> or that all my lamps switch off 15 minutes after me or my partner leaves the apartment (not if one of us is still here of course)
<stintel> "hmmm, I think I forgot to unplug the iron"
<stintel> vpn, click, not a problem
<stintel> etc
<f00b4r0> to me all that iot / automation screems of more points of failures, maintenance and security nightmares. There's only so much time I'm willing to dedicate to controlling lamps :)
<f00b4r0> screams*
<f00b4r0> but I guess I'm just being old ;P
<stintel> :P
<stintel> some consider me old too :P
<f00b4r0> teenagers? :D
<Borromini> :P
<stintel> hehe
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/62/builds/200) of `pistachio/generic` completed successfully.
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/25/builds/199) of `sunxi/cortexa7` completed successfully.
<Znevna> yay green
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/71/builds/199) of `bcm63xx/generic` completed successfully.
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/50/builds/200) of `apm821xx/sata` completed successfully.
zkrx has quit []
zkrx has joined #openwrt-devel
dansan has joined #openwrt-devel
<SlimeyX> brr
<Znevna> mkay hope I didn't brick my ax53u x.x
<Znevna> it hangs just after pcie detectin
<Znevna> what did i fsck up x.x
<Znevna> ok, weird
<Znevna> official firmware was hanging too
<Borromini> Znevna: still messing with it?
<Borromini> luckily you got multiple heh :P
<owrt-2203-builds> Build [#196](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/49/builds/196) of `mediatek/mt7622` completed successfully.
<Znevna> yes apparently the last flash trashed it
<Znevna> my fault probably
<Znevna> official firmware was hanging at mt7621-pci 1e140000.pcie: Failed to get gpio for PCIe1
<f00b4r0> hmm maybe I shouldn't have crossgraded this raspi to vanilla devuan. It keeps resetting every now and then now
* f00b4r0 disables frequency scaling for a try
<Borromini> f00b4r0: from raspbian?
<f00b4r0> Borromini: raspbian -> debian -> devuan (I have documented the process)
<Borromini> :)
<f00b4r0> i've had success stopping at debian, although I didn't update the kernel to the current 6.0
<f00b4r0> I've also seen cpufreq causing similar havoc on raspi zeroes, so that'd be my primary suspect right now :)
<Borromini> ok :)
<Borromini> I'm on Armbian for my Odroid, happy enough with it.
danitool has joined #openwrt-devel
<f00b4r0> yeah armbian is mostly ok
<f00b4r0> i just got fed up with raspbian (lack of) security policy
<Borromini> :-/
<f00b4r0> and I wanted to give a shot at running a raspi without systemd :) (and I know I could do that with openwrt too, it might end there ;)
<Borromini> :D
Tapper has joined #openwrt-devel
<minimal> f00b4r0: Alpine has OpenRC rather than Systemd ;-)
<Borromini> i've made my peace with systemd, grudgingly. I found networkd more robust than Debian's ifupdown framework :(
<Znevna> looks like something trashed nvram x.x
<f00b4r0> well I nearly bricked my device when systemd upgraded: I had the watchdog setup and apparently the configuration changed, there was no warning, and the device ended up rebooting 6-7s after coming online.
<f00b4r0> I had to be "creative" to recover from that, considering it's 200km from here
<f00b4r0> and I don't have console
<Borromini> :-/
<f00b4r0> i don't have enough foul words to express my contempt for this piece of software (and the people who brought it to life) ;-P
<slh> Borromini: there's one very annoying catch with networkd (I use it - and prefer it over ifupdown, but...), upgrading systemd over ssh tends to kill the active ssh session - which is 'great design(TM)'...
<Borromini> slh: i haven't noticed that personally.
<slh> I'm seeing that quite regularly (Debian/unstable), as I'm upgrading quite a few (local-) systems via ssh
<Borromini> i take your word for it.
<Borromini> maybe a regression?
<slh> I've been seeing it for a couple of years now (and I hopped onto networkd rather early on)
<f00b4r0> i think that's "known". I remember seeing something like that on a website that collects all the reasons why systemd is an abomination :)
<Borromini> xD
<Borromini> i'm surprised Poettering didn't DDOS it yet ;)
<aiyion> stintel, blocktrron: I've handed in a patch that ensures a unique and reboot persistent mac for the NanoPi R1 with eMMC and makes it more probable for the variant with just SD-cards.
<owrt-2203-builds> Build [#210](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/11/builds/210) of `at91/sama5` completed successfully.
stokito has joined #openwrt-devel
gladiac has quit [Quit: k thx bye]
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/69/builds/198) of `gemini/generic` completed successfully.
stokito has quit [Quit: Page closed]
stokito has joined #openwrt-devel
stokito has left #openwrt-devel [#openwrt-devel]
stokito has joined #openwrt-devel
stokito has quit [Quit: Leaving.]
<owrt-2203-builds> Build [#202](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/41/builds/202) of `apm821xx/nand` completed successfully.
stokito has joined #openwrt-devel
<Znevna> I .. give up
<Znevna> with serial connected wireless doesn't show up
stokito has quit [Quit: Leaving.]
rsalvaterra has quit []
rsalvaterra has joined #openwrt-devel
<owrt-2203-builds> Build [#201](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/46/builds/201) of `omap/generic` completed successfully.
<karlp> stintel: I admire your faith and optimize in matter.
<karlp> I'm not at all convinced it's well founded yet, but they're all certianly talking the talk :)
cmonroe has joined #openwrt-devel
<Borromini> Znevna: is this master with 5.15?
<Znevna> yeah but it's not related to kernel
<Znevna> I've removed the cables, it booted fine
<Znevna> beats me
cmonroe_ has quit [Ping timeout: 480 seconds]
<Znevna> I have to get a better serial adapter :P
<Borromini> o_O
xes has joined #openwrt-devel
<Znevna> is all this normal? https://paste.debian.net/plainh/b0dcc21f
<Znevna> upgrade without keeping data
Tapper has quit [Quit: Tapper]
<stintel> karlp: :)
<stintel> karlp: thread looks decent at least
<stintel> neither are open standards but ... 802.11 is kind of the same, no ?
<Habbie> they are not open standards?!
<dwfreed> they are freely obtainable standards, but you cannot redistribute them
<Habbie> ah
<stintel> can you freely download the 802.11/thread spec?
<dwfreed> nor can you make a product that uses it without paying
<robimarko_> stintel: 802.11 you can download from IEEE
<Habbie> uses it, or advertises its support?
<stintel> robimarko_: everything?
<stintel> robimarko_: I have access because meta, but I didn't think I could, without paying
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/74/builds/200) of `mpc85xx/p2020` completed successfully.
<nkko> so, has anyone gotten mainline to run on the bcm963138?
<robimarko_> You can get the latest versions for free
<nkko> I have access to broadcom's cfe code and I've been going through it
<nkko> I think I know how it mostly works now, with regards to loading the kernel image
<dwfreed> stintel: free as in beer, but you do have to tell them who you are
<robimarko_> Yeah, you have to have an account
yolo has joined #openwrt-devel
<robimarko_> But no subscription or anything, same goes for 802.3 and some other standards families
<dwfreed> Habbie: "Membership in Thread Group is necessary to implement, practice, and ship Thread technology and Thread Group specifications."
<yolo> watched 2022.11 yocto summit esp its SPDX and SBOM, saying it's possible EU etc will mandate it for open source projects, openwrt might need that sometime
<Habbie> dwfreed, ack, thanks
<hauke> yolo: Is this about the EU Cyber Resilience Act?
<owrt-2203-builds> Build [#200](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/39/builds/200) of `layerscape/armv7` completed successfully.
<owrt-2203-builds> Build [#198](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/59/builds/198) of `x86/64` completed successfully.
<f00b4r0> Borromini: turns out random reboots were due to lack of syslogd. Now I have more incentive to see if I could use openwrt's so I can stop killing my sdcard :)
robimarko_ has quit [Quit: Leaving]
<Borromini> f00b4r0: glad you found the culprit :)
<f00b4r0> was easy enough. After cpufreq didn't change anything I had a hunch the system might be unhappy without a way out for all those log messages
<f00b4r0> probably overflowing the kernel buffer
<Borromini> :)
<Borromini> i've had times where I thought to myself 'nope can't be it' and kept digging stubbornly... only to find out hours later it *was* the first thing i thought of but found too obvious >_>
<f00b4r0> yeah, I suppose old age taught me to pay more attention to the little voice ;)
<f00b4r0> and be a bit less stubborn that I used to be
<Borromini> i'm curious as to what you define as old age ;)
<f00b4r0> I'm kidding
* Borromini just passed 40
<f00b4r0> although this day next month I'll be 42
<Borromini> see ;)
<f00b4r0> well this day -1 since we just past midnight
<Borromini> congrats!
<f00b4r0> heh
<Borromini> oh next month
<Borromini> sorry
<f00b4r0> ;)
<slh> beware of the soon-to-be-neckbeard staring at you in the bathroom mirror ;)
<f00b4r0> rofl
<Borromini> :P
<f00b4r0> anyway, me not getting any younger I need more sleep. gnight all :)
<Borromini> night
csrf has joined #openwrt-devel
<yolo> hauke: not sure about that
SamantazFox has quit [Quit: Bye]
<owrt-2203-builds> Build [#199](https://buildbot.openwrt.org/openwrt-22.03/images/#builders/52/builds/199) of `bcm27xx/bcm2709` completed successfully.
gromero_ has quit [Ping timeout: 480 seconds]