danitool has quit [Ping timeout: 480 seconds]
Grische has quit [Remote host closed the connection]
guerby_ has quit [Remote host closed the connection]
guerby_ has joined #openwrt-devel
fda has joined #openwrt-devel
Tapper has joined #openwrt-devel
fda has quit []
fda has joined #openwrt-devel
fda has quit [Quit: ZNC - https://znc.in]
fda has joined #openwrt-devel
rua has joined #openwrt-devel
nick[m]1234 has quit [Ping timeout: 480 seconds]
whatevs111[m] has quit [Ping timeout: 480 seconds]
pavlix has quit [Ping timeout: 480 seconds]
t4h4[m] has quit [Ping timeout: 480 seconds]
JuniorJPDJ has quit [Ping timeout: 480 seconds]
hexagonwin[m] has quit [Ping timeout: 480 seconds]
Q__ has quit [Ping timeout: 480 seconds]
MatrixTravelerbot[m] has quit [Ping timeout: 480 seconds]
will[m]1 has quit [Ping timeout: 480 seconds]
Movedtomkg20001mkg20001io[m] has quit [Ping timeout: 480 seconds]
tohojo has quit [Ping timeout: 480 seconds]
schmars[m] has quit [Ping timeout: 480 seconds]
olmari has quit [Ping timeout: 480 seconds]
lipnitsk has quit [Ping timeout: 480 seconds]
John[m]12345678910 has quit [Ping timeout: 480 seconds]
evils[m]1 has quit [Ping timeout: 480 seconds]
gnustomp[m] has quit [Ping timeout: 480 seconds]
aparcar[m] has quit [Ping timeout: 480 seconds]
MatMaul[m] has quit [Ping timeout: 480 seconds]
decke[m] has quit [Ping timeout: 480 seconds]
fpsusername[m] has quit [Ping timeout: 480 seconds]
fieryeagle954[m] has quit [Ping timeout: 480 seconds]
bluse-blue[m] has quit [Ping timeout: 480 seconds]
znullptr[m] has quit [Ping timeout: 480 seconds]
mkg20001 has quit [Ping timeout: 480 seconds]
domon has quit [Ping timeout: 480 seconds]
lykos153 has quit [Ping timeout: 480 seconds]
fda has quit [Quit: ZNC - https://znc.in]
fda has joined #openwrt-devel
nick[m]1234 has joined #openwrt-devel
Tapper has quit [Quit: Tapper]
<neggles> hurricos: oh my gof
<neggles> uh, s/gof/god, but, *they included the code for their password generator*
<Grommish> neggles: How else they gonna find it again?
<neggles> im just surprised they didn't cut it out of the u-boot source
<neggles> i mean looking at these sources, it would *definitely* be a GPL violation if they did, but i'm still surprised they didn't try it
Grommish has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
<neggles> though they have only given me a sha1 hash, finding a collision / crack for that could be fun
guerby_ has quit [Remote host closed the connection]
guerby_ has joined #openwrt-devel
shoragan has quit [Ping timeout: 480 seconds]
shoragan has joined #openwrt-devel
swalker has joined #openwrt-devel
dansan has joined #openwrt-devel
odin has joined #openwrt-devel
odin has quit []
Grommish has joined #openwrt-devel
philipp64 has joined #openwrt-devel
Slimey has quit [Remote host closed the connection]
valku has quit [Quit: valku]
csharper2005 has joined #openwrt-devel
<csharper2005> Hi guys! How to make kernel choose mtd with rootfs depending on the bootloader arg? Is it possible without additional kernel patches?
<dwfreed> on platforms where A/B booting is common, openwrt has a patch to consume the root device provided via the bootloader's kernel command line
<dwfreed> see eg the nbg6817
Movedtomkg20001mkg20001io[m] has joined #openwrt-devel
<csharper2005> dwfreed: thanks! I'll take a look
Slimey has joined #openwrt-devel
ekathva has joined #openwrt-devel
Grommish_ has joined #openwrt-devel
Grommish has quit [Ping timeout: 480 seconds]
csharper2005 has quit [Remote host closed the connection]
srslypascal is now known as Guest475
srslypascal has joined #openwrt-devel
Guest475 has quit [Ping timeout: 480 seconds]
csharper2005 has joined #openwrt-devel
PaulFertser_ has quit [Ping timeout: 480 seconds]
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
will[m]1 has joined #openwrt-devel
danitool has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
borek has joined #openwrt-devel
rua has quit []
borek1 has joined #openwrt-devel
borek has quit [Ping timeout: 480 seconds]
borek1 is now known as borek
Tapper has joined #openwrt-devel
shibboleth has joined #openwrt-devel
<rsalvaterra> OMFG. OpenConnect seems to have some gained support for the ArrayNetworks SSL VPN protocol while I wasn't looking…! Thanks, dwmw2_gone! :)
* rsalvaterra needs this crap for $dayjob…
robimarko has joined #openwrt-devel
<rsalvaterra> I have no words to express my hatred of wiki instructions using UCI commands. https://openwrt.org/docs/guide-user/services/vpn/openconnect/client
<dwfreed> what's with the renaming especially
<grift> rsalvaterra: what is wrong with that, and what alternative do you see?
MatMaul[m] has joined #openwrt-devel
decke[m] has joined #openwrt-devel
<rsalvaterra> grift: Text files. Just show what the damn /etc/config/<file> should contain.
<rsalvaterra> Text files don't bite. And they're WAY more readable.
<jow> those wiki instructions are designed to be copy and pastable
<jow> text snippets are unsuitable for that
<rsalvaterra> jow: I have opinions about copy/pasting commands from web pages. :P
<rsalvaterra> And text snippets can also be copied into the text editor.
<jow> another problem with text snippets is that there's many ways to use them wrongly
<jow> accidentally truncating trailing quotes, using wrong newlines, pasting at the wrong place, pasting into the wrong file
<jow> introducing sublte syntax errors
<jow> any you'd be surprised how many users do not have a text editor available and are unable to operate vi
<grift> the only thing i find a bit annoying is the use of variables, i dont see the point and it makes it harder to read
<rsalvaterra> jow: Ugh… you're not wrong.
<rsalvaterra> grift: Yeah, don't get me started on that. :)
<csharper2005> jow: let's add mcedit by default :)
<stintel> well can you blame normal users for not knowing vi?
<jow> I don't blame them, just stating that this is one of the reasons why copy-pasteable commands make sense
<stintel> and this will only get worse
<jow> instead of having a half long page intro on how to paste (and exit!) vi with each text snippet
* stintel hates uci and /etc/config/* equally
* jow hates the usual freak zoo of native application configs commonly found on a normal linux system
<rsalvaterra> stintel: You don't need to go that far. I am able to use vi, but I don't like it. Coming from a long DOS background, I feel more comfortable with an edit-style editor, like nano.
<rsalvaterra> jow: Violent agreement there. /etc is a jungle.
<jow> is it key value style? does it follow shell quoting semantics? is it xml-like? or nested curly brace style? commands are hashmarks? or slashes? oh no comments at all? significant white space? application config parser smart enough to trim trailing spaces?
<grift> i am usually not a big fan of abstraction, but in the case of uci it makes sense in my view
<f00b4r0> rsalvaterra: afaict I see here pretty much word for word the suggested firewall config for wireguard https://openwrt.org/docs/guide-user/services/vpn/wireguard/client and it makes no sense to me either
* f00b4r0 suspects a heavy dose of copy-pasta has spread around
<jow> anyone here has experience with commercial wireguard vpn providers?
<f00b4r0> nope. My use is self-contained
<jow> do they usually provide a wireguard.conf you're supposed to use or are they just giving you the private key or are you supposed to give them your public key?
<rsalvaterra> jow: None. However, dangole seems to like Mullvad. I don't know if he personally used it.
<jow> was considering adding some kind of config import to the luci wireguard settings but I have no idea how users usually receive their client credentials
<rsalvaterra> jow: Most providers I've seen have some way to generate the credentals (IP address and keys) at connection time, in order not to keep a static address.
<jow> f00b4r0: would love to see some non-openwrt examples
<rsalvaterra> And that doesn't integrate at all with OpenWrt.
<jow> so you have some kind of client application wrapping the wireguard setup?
<rsalvaterra> Yes, pretty much.
<jow> okay, so most seems to provide some sort of wg.conf generator
Borromini has joined #openwrt-devel
<Tapper> Any one know what this is from my logs on a r7800 running latest master Kernel 5.15?
<rsalvaterra> dwmw2_gone: Just a heads up, while building OpenConnect 9.01 in OpenWrt, with OpenSSL, I had to enable deprecated API support, otherwise the build would fail.
<Tapper> Thu May 12 09:30:12 2022 kern.warn kernel: [ 4509.081665] ath10k_pci 0000:01:00.0: Invalid peer id 6 or peer stats buffer, peer: 00000000 sta: 00000000
<stintel> Tapper: it's qca, replace with mtk ;)
* stintel hides
<Tapper> lol I am thinking about it.
<rsalvaterra> dwmw2_gone: openssl.c:59:46: error: implicit declaration of function 'SSLeay_version'; did you mean 'SSL_version'?
<Tapper> Got my 6260 as a AP and it's quite good
<robimarko> Probably one of the stupid FW prints
<robimarko> ath10k has had a ton of useless prints enabled
<Tapper> OK
<robimarko> Basically, if it works ignore
<Tapper> I was looking at the llogs because my wifi keeps dropping a streem for 3 seconds every 2 or 3 minets
<Tapper> *mMinutes
<Tapper> Mostley voice streems like when using clubhouse or teamtalk
<Tapper> on my android phone
Lynx- has joined #openwrt-devel
<Lynx-> The attended sysupgrade seems to fail since at least yesterday? "revision": "r19315-d4053d2e8e", "profile": "linksys,e8450-ubi", "target": "mediatek/mt7622","version": "22.03-SNAPSHOT", leads to error: Server response: The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.
<KGB-1> https://tests.reproducible-builds.org/openwrt/openwrt_sunxi.html has been updated. (0% images and 99.9% packages reproducible in our current test framework.)
<rsalvaterra> Damn, OpenConnect is chubby…! Increased my image size by 856 kiB. Oh, well.
tohojo has joined #openwrt-devel
enyc has quit [Ping timeout: 480 seconds]
goliath has joined #openwrt-devel
rua has joined #openwrt-devel
<grift> well-rounded
enyc has joined #openwrt-devel
<rsalvaterra> Heh… depends on quite a number of libraries, not entirely unexpected.
ekathva has quit [Remote host closed the connection]
torv has quit [Remote host closed the connection]
shibboleth has quit [Write error: connection closed]
torv has joined #openwrt-devel
Borromini has quit [Quit: Lost terminal]
shibboleth has joined #openwrt-devel
<neggles> tbh i don't know why the mini-nano isn't included by default
<neggles> it's friggin' tiny
<robimarko> That would be great
<robimarko> I am used to vi now, but its still not the easiest to use
<neggles> the whole obsession with "NO!!! embedded system space limited! no user-friendly editor! only vi! ONLY VI!" just seems completely ass-backwards these days
<neggles> is vi *actually* smaller than the base `nano` openwrt package???
<robimarko> Doesnt nano currently pull-in ncurses?
<neggles> i don't know tbh
<robimarko> Yeah, it pulls in libncurses
<neggles> i've gotten vaguely used to vi insofar as i can use it for incredibly basic stuff but i still routinely find myself having to go esc-esc-:Q! and start again
<robimarko> Same here
<neggles> one time i managed to get it to paste a block of text 4096 times
<robimarko> Its just not pleaseant to use
<neggles> instead of typing 4096
<neggles> still have no idea how i did that.
<owrt-snap-builds> Build [#550](https://buildbot.openwrt.org/master/images/#builders/51/builds/550) of `ramips/rt3883` completed successfully.
<neggles> or how I copied the chunk it pasted in the first place
<neggles> i guess libncurses is 145k so that's less fun
<robimarko> Yeah, and nano itself is 65k
<neggles> could probably make a statically linked nano with a bare-minium-required-features libncurses
<neggles> does pico need ncurses?
<neggles> ah no pico package because weirdness with licensing fairo
<Lynx-> auc output: Running: 22.03-SNAPSHOT r19315-d4053d2e8e on mediatek/mt7622 (linksys,e8450-ubi)
<Lynx-> No data available (61) <--- what gives?
lykos153 has joined #openwrt-devel
<f00b4r0> i couldn't live without vi. Then again, $editor wars :)
<grift> the status quo seems fine. you generally dont need an editor anyway, and if you cannot deal with vi then nano is one opkg install away. and you wont hear emacs users complain because they have tramp
<grift> or do like stintel and tunnel the whole thing trough ssh
<stintel> ?
<grift> werent you the person that accesses a web interface binding sockets to 127.0.0.1 from a local browser?
<stintel> wtf?
<grift> i guess mistaken identity
<stintel> you must have me confused with someone else yeah
<stintel> I usually edit /etc/config/whatever with vi over ssh
<stintel> I use LuCI mostly for checking status or the realtime graphs
<stintel> and to manually create a config backup
<hanetzer1> vi is nice.
hanetzer1 is now known as hanetzer
<rsalvaterra> neggles: There's a mini-nano…? o_O
<rsalvaterra> Repo, please.
Tapper has quit [Ping timeout: 480 seconds]
<rsalvaterra> Oh, still depends on libncurses…
<hurricos> neggles: is it pure sha1, or how is it salted?
<neggles> hurricos: pure sha1
<neggles> literally just sha1(text)
<neggles> e597301a1d89ff3f6d318dbf4dba0a5abc5ecbea
<hurricos> stintel: some companies (mine) still require regular support folks to do the vi tutorial during training
<hurricos> the dream is alive and well
<hurricos> I'm an emacs user and I still use vi
<neggles> rsalvaterra: i am currently halfassedly attempting to build the smallest static nano i can
<stintel> when I first used Linux I used nano, but I learned vi because it's the one thing that is usually available everywhere
<stintel> and right now I can't imagine using anything else
<neggles> stintel: this is my entire point - *isn't that the only reason anyone learns vi?!*
<hurricos> Yes! It's a blessing it comes in busybox. I just wish mode changing out of insert didn't take so long
Tapper has joined #openwrt-devel
<neggles> also... i must point out, we're not exactly 'normal users' :P
<neggles> how big is vi even
<hurricos> Lynx-: that server is down. Try asu.aparcar.org in /etc/config/attendedsysupgrade
<stintel> ach normal users shouldn't be allowed to use the internet :P
<hurricos> neggles: it's a busybox app
<grift> arguably normal users rely on LuCI
<neggles> yeah but cutting it out of busybox must save some KiB
<hurricos> I can't imagine why you'd want to do that though
<hurricos> err
<hurricos> OK, I can imagine
<neggles> if I can cut vi out of busybox and add in a tiny static nano and only bump total used space by some tens of KiB
<neggles> i call that a win
<robimarko> that would be interesting
<hurricos> busybox.net is at this point unresponsive ;(
<robimarko> No idea, how much space does disabling vi even saves
<neggles> don't even need 90% of nano's features
<neggles> literally just give me 'the four keyboard shortcuts you need are displayed at the bottom of the window' and 'basic editing works without having to read a single manpage'
<hurricos> just do your editing with sed
<robimarko> How is that easier?
<neggles> it is not :P
minimal has joined #openwrt-devel
<neggles> the problem with just statically linking ncurses into nano is, i bet more than half the library is functions it's not using
<hurricos> hot take
<hurricos> remove vi
<hurricos> use emacs tramp to edit
<hurricos> 0kb :^)
<neggles> emacs is not better!
<hurricos> emacs will ship the remote system an awk script to base64 your data lol
<hurricos> no awk? I think it drops down to like hexdump
<robimarko> If I am right, then disabling vi saves whopping 12.43kB
<neggles> hurricos: that doesn't help me when i'm using a uart console m8
<hurricos> 12.43kB you'd have if you used emacs^tm
<neggles> from a windows box
<hurricos> true
<hurricos> > windows box < O_o
<neggles> I am playing the part of a random end user
<neggles> who has screwed up their config
<neggles> or just wants to not deal with pasting eleventy billion uci commands to add a dozen bridge vlans or whatever
<grift> in that scenario vi isnt that much less user friendly than nano
<neggles> sure is
<neggles> i open nano, i type things, i hit the keyboard shortcut it shows me at the bottom to save and exit
<robimarko> If you can squeeze nano into lets be generous 13k
<robimarko> I doubt thats possible
<neggles> vi: i open vi. i try to type something and nothing happens, i end up pressing D twice and deleting the line, oh shit. now what do i do? I keep randomly typing things, hoping to find some sort of help or menu. it has been twenty minutes, and i cannot even work out how to exit without saving. i powercycle the device.
<grift> lol
<robimarko> I just love how it moves to start of the line when I go into insert mode
<neggles> "just look up how to use it" *my router is broken, I have no internet*
<robimarko> Which I always forget before already being at the place I would like to edit
<lemmi> robimarko: then don't press the shift key
<neggles> i just love how delete works even when you aren't in insert mode
<neggles> </s>
<grift> i think your leaving out an important aspect
<grift> such a person wouldnt even know `vi file`
<robimarko> lemmi: thanks, old habits die hard
<grift> if that person knows how to call `vi` (or nano for that matter) then that person knows the basics
<robimarko> Why?
gladiac has quit [Read error: Connection reset by peer]
<neggles> ive come across a lot of systems that have `alias edit=$EDITOR`
<robimarko> Just calling vi file doesnt mean you know how to use it
<robimarko> nano is kind of idiotproof compared to that
<neggles> and you assume they haven't gone and searched 'edit file openwrt' and seen 'you type vi'
<grift> well you know of its existence
<neggles> then stopped reading
<neggles> maybe they know it exists because they've seen people talk about it, but they've never used it before
<neggles> mostly i just think it's really dumb that the default text editor for every embedded system is completely impossible to use without a manpage.
<robimarko> Which is not present on OpenWrt
<neggles> ^
gladiac has joined #openwrt-devel
<neggles> i'm not saying vi is bad btw, i can see how with some practice it could be very cool and good. but the same things that make it so powerful make it utterly useless to someone who doesn't already know how it works
<f00b4r0> and here we have another $editor flame. Internet history keeps repeating ;)
<neggles> eh
<neggles> i just said i'm not saying vi is bad
<lemmi> neggles: you can say that about almost any tool
<f00b4r0> ^
<neggles> i have never once read any manual at all for nano
<neggles> im sure it can do all kinds of stuff i've never used
<robimarko> f00b4r0: well, cause defaults seem to be set in stone
<f00b4r0> the first time I encountered nano, I was completely bummed. It's all a matter of habits.
<neggles> but i have not needed to read it to be able to use it to edit text
<f00b4r0> as for the user who wants to paste a gazillion lines to restore some file, $editor is not the answer.
<f00b4r0> cat is.
<neggles> what about the user who wants to edit one line because they typed 'eht0' and didn't double check before `service network restart`
<robimarko> Know you are entering into power user level
<f00b4r0> robimarko: the minute you went to command line, you did :)
<grift> thats why we use uci
<robimarko> Yeah, but its not like you have an option
<f00b4r0> ^
<f00b4r0> neggles: sed! :)
<grift> you do a uci commit network && reload_config
<neggles> sigh
<neggles> uci won't stop you typing completely invalid values for things.
<grift> uci commit should catch that
<f00b4r0> neggles: and it won't stop you easily correcting them too
<neggles> this whole thing started with 'uci bad' though
<neggles> and in many ways it is
<neggles> the friggin [@blah] syntax is confusing as hell
* stintel likes Cisco like CLI
<neggles> i like vyos/juniper-ish cli tbh
<neggles> yes it's basically just json with less pedantry
dedeckeh has joined #openwrt-devel
<neggles> but both cisco and juniper/vyos/edgeos/whatever cli have one major thing that uci doesn't
<neggles> `?`
<neggles> 'show me what i can put here / suggest what the next word should be from the ones that it could be'
<neggles> (sometimes it's 'mash tab twice' not 'press ?', details)
<grift> i want to believe that nano solves all these issues but i suspect that even if you type eht0 with nano and restart the network, you might run into issues and maybe even end up power cycling
<grift> but i think its clear by now that nano is too big +/- 200KiB vs. 12KiB
<neggles> ive got it down to 130 gzipped
<neggles> without even trying
<robimarko> Nobody is talking about shipping the current nano with libncurses
<neggles> god no
<neggles> way too big
<neggles> considering gutting huge chunks of it, folding in the few bits of ncurses it actually *needs*, idk, that's probably beyond my programming capabilities
<neggles> core bits and pieces of nano as a busybox addon would work. or even... is it possible to make a vi addon/plugin that adds some more intuitive input modes? kinda surprised nobody's done that?
<neggles> grift: its more that if i just want to edit a bit of a file, or copy-paste a bit of config and change it slightly (like adding bridge vlans), i don't have to work out exactly how uci expects me to refer to that config entry, and i don't have to work out how tf vi works, i can just... move the cursor and type, with backspace and delete working the
<neggles> way they do in every graphical editor / thing that is not vi (or emacs? i do not actually know jack about emacs)
<grift> i wouldnt mind nano (obviously) if you want to maintain a fork then be my guest. even if you use vi, nano, sed, emacs i still encourage that you run uci commit before actually reloading a service
<Lynx-> hurricos thank you
<Lynx-> hurricos OK to keep it that url?
<neggles> interesting.
<neggles> 8.6KiB gzip'd for mipsel
<neggles> grift: i do generally run uci commit, but i'm yet to have it actually catch any of my boneheaded mistake
<neggles> tbh it's not really about me, more about when i'm helping friends because i made the mistake of telling them to put openwrt on something 😆
<neggles> and a general dissatisfaction with `vi` being The One Editor You're Getting, Son
<grift> that will still apply anyway except then it will be s/vi/nano/
<neggles> yes, but nano is possible to use by just opening it and reading what it says on screen
<neggles> i do not have to work out that i need to press a key to be able to type text - though i can delete it without doing so?! - then press escape and type !wq to save and quit
<grift> its :wq
<neggles> ...yeah
<neggles> that kinda reinforces my point though.
<neggles> i used vi twice today.
Lynx-- has joined #openwrt-devel
Lynx- is now known as Guest502
Lynx-- is now known as Lynx-
Guest502 has quit [Read error: Connection reset by peer]
<neggles> but nano? you just type like it's notepad.exe, and it shows you right there on screen "ctrl-x to exit" - then it asks if you wanna save Y/N, so you press y and hit enter
<neggles> *anyone* can work it out *from within the program, without looking anything up*
<neggles> i should just port msdos's edit.exe 😝
<grift> i do understand your point but it doesnt solve checking for valid syntax, its too expensive unless you fork it and strip 80 percent of it and if uci is too complicated then maybe its better to address that issue instead?
<neggles> goalposts
<neggles> specifically, moving them
<neggles> "it would be nice to have a very small editor that is self-explanatory, so we can be free of the 'you can have any editor you like as long as it's vi' in embedded systems"
<neggles> making uci more user friendly is also a good idea, but i suspect you can't really do that without losing the 'u' in uci
<neggles> idk maybe at some point i'll fork nano and turn it into `femto` or something
<grift> i just don't know if its a good idea to encourage users to edit config bypassing uci
<lemmi> isn't there already pico somewhere?
<neggles> yeah but its wrapped up inside alpine
<neggles> and the licensing is weird
<neggles> (alpine the mail client, not the distro)
<lemmi> ah
<lemmi> i was just wondering for a bit
<neggles> yeah me too :P
<neggles> maybe it's not too hard to extract? but i wouldn't be shocked if it had even more dependencies
<neggles> maybe the best solution here is just 'an old version of nano from before it grew so many features' - i think some of the older ones can even be built without ncurses
<f00b4r0> alpine-pico pulls libtinfo6 on debian
<neggles> nano/libncurses tries to pull libtinfo as well, but you can tell it not to
<f00b4r0> nano-tiny adds libncurses to that
<neggles> took me a few tries. i had to pass `PKG_CONFIG=/bin/false` so it wouldn't pick up my host pkg-config and try to link tinfo
<f00b4r0> anyhow, I'm a vi(m) user, so I'm happy already :)
<neggles> grift: it's less 'encouraging users to edit config manually' and more 'making it possible for them to edit files directly when that's what's necessary / the easiest thing to walk them through remotely'
<Lynx-> Is it unreasonable to assert that mostly those needing to configure OpenWrt via command line will either be familiar with vi already or able to learn how to use it pretty quickly?
<neggles> yes!
<f00b4r0> no
<f00b4r0> :)
<neggles> well. okay. yes, "mostly"
<neggles> what about the other people?
<lemmi> they reset or sell the device
<Lynx-> Like those who have no business trying to configure via command line anyway?
<lemmi> because what the hell is ssh or seria
<neggles> way to gatekeep.
<Lynx-> Well I am just thinking that mostly those with the skills/capability to configure via command line will be fine to use vi
<Lynx-> how often is someone well able to configure via commandline but stumped by vi?
<neggles> i've walked people who didn't even know what a command line *was* through SSHing into things and editing files with nano, or plugging a cisco console cable into something and opening putty
<Lynx-> I mean let's face it, OpenWrt is for total nerds
<neggles> but vi? hahahahahahahahaha
<rsalvaterra> neggles: https://github.com/mavaji/dos-editor ;)
<f00b4r0> because "start with 'i' and end with ':wq'" is too hard to convey? ;P
<Lynx-> I'll wager the vast majority of OpenWrt users are well on the autistic spectrum, and with such enhanced systemizing capability, learning how to use vi should be a doddle.
<rsalvaterra> f00b4r0: :wq? Not :x? :)
<f00b4r0> :)
<neggles> it's not as simple as start with i though. and it's far too easy for them to mistap something and get it all outta wack.
<stintel> ZZ
<neggles> Lynx-: and it will stay as a nice thing for total nerds as long as that's the attitude we have...
<neggles> s/nice/niche
<Lynx-> but there's LuCi
<Lynx-> LuCi is for neurotypical people; commandline is for nerds
<neggles> i'm sure you've seen some of the absolute lunacy luci generates sometimes.
<Lynx-> I mean if you are accessing your router via a command line, I think you are a nerd
<grift> so fix luci?
<Lynx-> If you use IRC you are definitely a nerd
<grift> instead of blaming vi for everything that is wrong in this world
<neggles> okay i am done having this pointless circular conversation, because i have made multiple points about how it's not about *me, personally* but about making things easier to walk people through remotely, and friendlier to newcomers, and it's not just an openwrt thing, nor does it have anything to do with my personal dislike of vi.
<neggles> sorry i brought it up and kept it going so long.
<Lynx-> yes I agree with grift - LuCi should be enhanced rather than switching out vi
<neggles> but this whole attitude of 'if you can SSH into something you should know how to use vi' is just... yet another of the many reasons why this year will always be the year of linux on the desktop. etc.
<grift> i dont have this urge to please
<Lynx-> I think you can also just samba share / and then use notepad in windows?
<neggles> preeeeeeeeetty sure even ksmbd is bigger than nano.
<Lynx-> :)
<neggles> and notepad is a big fan of CRLF :P
<Lynx-> OK I have an idea for you neggles
<grift> didnt they put samba in the kernel?
<Lynx-> I got a sweet idea
<neggles> that's ksmbd and it's not small
<Lynx-> this is really the dog's bollocks so are you ready?
<neggles> is it using scp
<neggles> it better not be using scp
<Lynx-> nope
<Lynx-> this is actually really good
<neggles> i already hate it hit me
<Lynx-> someone should write a text editor interface for LuCi to allow editing files
<Lynx-> via web
<neggles> that is neat, but doesn't help the uart console case
<Lynx-> :)
<grift> you already can edit rclocal and crontab that way i think
<neggles> also the luci infrastructure for that already exists yeah
<Lynx-> yep grift - so just extension
<grift> come on, no "normal" user is ever going to use uart
<grift> not even if you guide them
<Lynx-> haha yes
<Lynx-> this is not just nerd
<neggles> eh
<Lynx-> but uber nerd
<Lynx-> like no just on the spectrum, but far gone
<neggles> i've walked J Random Office Worker through plugging a micro-usb cable into the console port on a sophos firewall
<neggles> not the most pleasant thing on the planet but saved me a 4hr round trip drive
<neggles> (the secret is that PuTTY is installed in our standard laptop image)
Tapper has quit [Ping timeout: 480 seconds]
<neggles> but really my ulterior motive here is 'make vendors start putting something that is not vi into their OS images so i don't have to keep 16 different static nano binaries around'
<neggles> and that's probably not happening
<Lynx-> is it possible you could be converted
<Lynx-> do you feel the rush I feel when I use vi like a kind of wow this is so cool type irrational feeling?
<neggles> i am also a raging autist and i have decided I Don't Like Thing, so no
<Lynx-> hmm yes I get that
<Lynx-> I mean I'm the same for other things
<neggles> maybe can make a vi plugin so that you don't have to flip back and forth between insert mode and whatever not-insert-mode is just to get notepad/nano/whatever-style 'i can just type stuff and both backspace and forward delete work'
<Lynx-> the one thing I find totally stupid in vi is the find and replace
<Lynx-> it's obscenely complicated just to do a simple find and replace
<Lynx-> quite honestly I just use notepad for that
<neggles> this is actually quite good https://github.com/antirez/kilo
<neggles> looks like vi but shows you the three keyboard shortcuts on startup, and you can just type stuff.l
<neggles> Lynx-: tbh for find and replace... i do just end up using sed.
<neggles> or copying the file and editing in vscode or whatever.
<neggles> but I suppose that's also kind of the point? if i'm doing anything more complicated than typing a few lines, i'm going to pull the file into something vastly more powerful
<neggles> heh. minor issue with kilo, can't quit it from inside a vscode terminal becsuse it wants ctrl-Q, which vscode intercepts. 🤦
rua has quit [Ping timeout: 480 seconds]
<grift> emacs
<neggles> emacs isn't an editor
<grift> true
<neggles> emacs is
<neggles> i don't even know what to call it
<grift> lisp framework
<grift> but emacs can do editing (theres not much it cannot do)
<neggles> the bastard lovechild of an editor, busybox, and an entire programming language?
<KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_omap.html has been updated. (11.1% images and 99.9% packages reproducible in our current test framework.)
<neggles> emacs is 1970s python don't @ me
<grift> plus something like vscode is probably just not an option, in linux mesa breaks every other week so you need something that works both in aa gui environment (for the weeks when mesa works) and at the same time from a tty (for the weeks where mesa doesnt work)
<neggles> if you have a working browser
<neggles> you have vscode-server :D
<grift> emacs has browser
<neggles> mate you can use emacs as your entire userspace if you try hard enough
<neggles> didn't someone write a friggin emacs hypervisor
rua has joined #openwrt-devel
<neggles> anyway. point being if i'm doing something on-target it's going to be basic and simple, usually whatever's necessary to get network back up or change one tiny thing. anything more complex than that i'll do off-host and copy
<grift> yes and you should really use uci or luci for that, and if there are issues with that then its probably really best to address those
<neggles> okay, so, with uci
<neggles> is it possible to have more than one `system.` node? if so, is there a remotely legitimate usecase for it?
<Zero_Chaos> I have a bug that I don't know how to properly report. something locks up and uci calls timeout when booting with >3 radios configured
<Zero_Chaos> Can someone advise me on how/what to do to make that into a useful bug report?
<neggles> Zero_Chaos: that's... odd. can you still get into it / dump logs?
<neggles> actually i gotta go to sleep. but, one way or another you'll need to grab some logs - i am sure someone else here can help :)
<Zero_Chaos> neggles: I have full access but the logs are just a few timeout message and nothing useful as to why it's timing out
<Zero_Chaos> neggles: enjoy that sleep though, this bug will be here tomorrow
<neggles> if you can pastebin `logread` and `dmesg` + your /etc/config/wireless that would be helpful (sanitized of course)
<neggles> is it multiple radios, or multiple ssids? having 4 radios shouldn't be massively breaking stuff...
<neggles> bad neggles! don't get dragged into troubleshooting at half past midnight! :P good luck
<Zero_Chaos> 4 radios, 8 ssid per. works fine with 3 radios, fails with 4 radios and less ssid
<Zero_Chaos> neggles: go to bed, the answers will also be here tomorrow
srslypascal has quit [Remote host closed the connection]
srslypascal has joined #openwrt-devel
<grift> i can't really be helpful here other than point to https://openwrt.org/bugs
<grift> my gut feeling tells me look into ubus, but thats not helpful
<Zero_Chaos> grift: yeah, I'm all for opening a bug, I was just trying to see if I could gather something more useful so folks can actually guess.
<Zero_Chaos> and yeah, my gut feeling is something related to ubus as well, since that's what uci talks to?
<grift> yes probably
Tapper has joined #openwrt-devel
srslypascal is now known as Guest506
srslypascal has joined #openwrt-devel
srslypascal is now known as Guest507
srslypascal has joined #openwrt-devel
Guest506 has quit [Ping timeout: 480 seconds]
Guest507 has quit [Ping timeout: 480 seconds]
<Tapper> I just want to say that I am with neggles on the points about openwrt and it's edditer
srslypascal has quit [Ping timeout: 480 seconds]
srslypascal has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
t4h4[m] has joined #openwrt-devel
Tapper has quit [Quit: Tapper]
hexagonwin[m] has joined #openwrt-devel
pavlix has joined #openwrt-devel
csharper2005 has quit [Remote host closed the connection]
JuniorJPDJ has joined #openwrt-devel
csharper2005 has joined #openwrt-devel
borek has quit [Ping timeout: 480 seconds]
Larhzu has quit [Quit: leaving]
<csharper2005> Hi again, all! Can I consider this patch as "accepted" in Linux or not yet? https://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux.git/commit/?h=mtd/next&id=4213e556fe2a18782e91a87fec70d98c48f1ee41
<csharper2005> Should I backport it to target/linux/generic? My current version of the patch is in ramips dir.
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
ekathva has joined #openwrt-devel
<f00b4r0> csharper2005: I'm surprised it even got accepted with that license identifier
<robimarko> Why?
<robimarko> Its GPLv2 or later?
<f00b4r0> because linux is 2.0-only
<f00b4r0> unless things changed recently and I wasn't paying attention, of course
<robimarko> I have stuff merged with GPLv2 or later
<robimarko> As long as its GPLv2 compatible its fine
<f00b4r0> suggests nobody pays attention anymore.
<f00b4r0> the legal implications are bound to be "interesting"
<robimarko> Why?
<robimarko> Its GPLv2 compatible
<csharper2005> f00b4r0: I've seen a lot of examples with that license during the patch preparation
<robimarko> 2007, things appear to have changed since than
<f00b4r0> gpl3 is not gpl2 compatible. Anyone who elects for the "or later" breaks shit
<f00b4r0> no they really haven't, because well, nobody can relicense the kernel
<f00b4r0> i see, so individual files can have the or later
<f00b4r0> interesting
<robimarko> Heck, we even have GPL/BSD dual licensing in bindings/DTS for example
<f00b4r0> yeah. The copyrightability of DTSes has always left me wondering tho
<f00b4r0> since it's really a description
<robimarko> Its probably not enforceable
<f00b4r0> but IANAL :)
<f00b4r0> that's exactly my guess. And the reason why I never bothered licensing the first few DTSes I wrote
<f00b4r0> then again I never used "or later" on any code I wrote. I've always found that to be a serious risk, because if the yet nonexisting future license version does something you're not happy with, you're very much screwed.
Larhzu has joined #openwrt-devel
Tapper has joined #openwrt-devel
<f00b4r0> (incidentally I think that was Linus' argument back then)
<robimarko> I understand that, its a personal choice
<robimarko> BTW, I think that checkpatch requires DTS-es to have SPDX identifier as well
<f00b4r0> i think so too
<hauke> DTS files are mostly BSD to allow inteagration into proprietary boot loaders too
<hauke> if you do GPLv2 or later, the user can choose which license he wants to use for this file
<hauke> if you want to use it inside of Linux GPLv2 it set, but if you copy this to a GPLv3 only kernel it is fine too
<hauke> ath9k is also BSD licensed to allow code sharing with FreeBSD
<f00b4r0> hauke: or later is extremely (i would say dangerously) broad, as exemplified by Linus' argument. If the kernel had been licensed under "or later", I suspect things would be very different today.
PaulFertser has joined #openwrt-devel
<robimarko> Ughh, anything I am trying to upstream requires dt-bindings to be converted to YAML first
<robimarko> I am getting better at it, but its taking me more time to do that and get that ACK-ed than the code itself
<f00b4r0> heh
valku has joined #openwrt-devel
<robimarko> Lets just say that bindings guys have more power than actual code maintainers
<f00b4r0> ;)
whatevs111[m] has joined #openwrt-devel
valku has quit [Remote host closed the connection]
MatrixTravelerbot[m] has joined #openwrt-devel
danitool has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
<KGB-1> https://tests.reproducible-builds.org/openwrt/openwrt_lantiq.html has been updated. (96.2% images and 99.9% packages reproducible in our current test framework.)
<csharper2005> my new patch conflicts with generic/435-mtd-add-routerbootpart-parser-config.patch... generic dir is below than backports dir. how to resolve this? Is it necessary yo rework "435-mtd-add-routerbootpart-parser-config.patch" in a separate commit?
SamantazFox is now known as Guest531
SamantazFox has joined #openwrt-devel
torv has quit [Remote host closed the connection]
torv has joined #openwrt-devel
Guest531 has quit [Ping timeout: 480 seconds]
Q__ has joined #openwrt-devel
<csharper2005> ping $(scripts/getmaintainers.pl generic/pending-5.15) :))
dedeckeh has quit [Remote host closed the connection]
lemoer has joined #openwrt-devel
aparcar[m] has joined #openwrt-devel
mattytap has joined #openwrt-devel
cmonroe has quit [Quit: Textual IRC Client: www.textualapp.com]
mattytap has quit [Quit: Leaving]
shibboleth has quit [Quit: shibboleth]
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
<xdarklight> hauke: thanks for reviewing my patches. I might get back to you on the weekend with a few more questions as I think that I don't fully understand what the hardware does (yet)
John[m]12345678910 has joined #openwrt-devel
olmari has joined #openwrt-devel
schmars[m] has joined #openwrt-devel
<hauke> xdarklight: no problem
ekathva has quit [Remote host closed the connection]
csharper2005 has quit [Remote host closed the connection]
cmonroe has joined #openwrt-devel
robimarko has quit [Quit: Leaving]
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
lipnitsk has joined #openwrt-devel
cmonroe has quit [Ping timeout: 480 seconds]
cmonroe has joined #openwrt-devel
domon has joined #openwrt-devel
<KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_bcm47xx.html has been updated. (100.0% images and 99.9% packages reproducible in our current test framework.)
rua has quit [Ping timeout: 480 seconds]
borek has joined #openwrt-devel
rua has joined #openwrt-devel
dansan has quit []
dansan has joined #openwrt-devel
borek has quit [Ping timeout: 480 seconds]
Tapper has quit [Quit: Tapper]
cmonroe has quit [Read error: Connection reset by peer]
cmonroe has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
fpsusername[m] has joined #openwrt-devel
mangix has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
mangix has joined #openwrt-devel
<neggles> I hope the SFC’s Vizio case goes well
<neggles> if it does that should make getting useful tarballs from vendors significantly easier