<rsalvaterra>
I'm pretty sure I never updated mine.
<rsalvaterra>
And I have my Omnia since late 2016.
<ynezz>
seems like 21.02-SNAPSHOT has this new commit, so probably it's causing regressions 089efd61e9a6 netifd: update to the latest version
<rsalvaterra>
Oh, wait, there are more variables… I'm running the 5.10 kernel, compiled with gcc 11 and binutils 2.37.
<ynezz>
21.02 runs on 5.4.132
<ynezz>
I assume, that kernel is not the culprit here, as the versions are same in both tests
<rsalvaterra>
Sure, netifd could be a likely culprit.
<ynezz>
nbd: ^
<rsalvaterra>
But in that case, it didn't fail on my machine.
<ynezz>
maybe it just doesn't work with default setup?
<rsalvaterra>
That's another likely possibility.
<ynezz>
I don't do any config changes, I test pristine images with default config
goliath has joined #openwrt-devel
<rsalvaterra>
It does smell like some network setup failure.
<rsalvaterra>
Do you have remote serial access to it?
aleksander has joined #openwrt-devel
<ynezz>
hm, seems like for some reason the wireguard link is broken on that device, but I can still run any command from inside the CI job
<ynezz>
what do you've in mind?
<ynezz>
from here it's quite clear, I'm not able to ping the other end with IP 192.168.1.2 from the DUT with default LAN IP 192.168.1.1
<ynezz>
link is up
<ynezz>
well, at least `ifstatus lan | jsonfilter -qe "@.up"` test success
<ynezz>
s/success/succeeds/
<rsalvaterra>
ynez: Nothing special, just doing a brctl show in order to see if the default bridge is correctly set up.
<rsalvaterra>
(Since that's what the last netifd commits seem to touch.)
<rsalvaterra>
*ynezz
<slh>
the nvmem might warrant a closer look as well
<ynezz>
slh: what do you mean exactly?
danitool has joined #openwrt-devel
<russell-->
"cat: can't open '/sys/class/ieee80211/phy0/macaddress': No such file or directory
<russell-->
"
<russell-->
in bootlog
Tapper has quit [Ping timeout: 480 seconds]
Tapper has joined #openwrt-devel
<slh>
ynezz: eventually fallout from https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=2ca8e424b93cbf359a8028acbbe9acabdb9365a9
<ynezz>
slh: unlikely 1) nvmem changes were not backported into 21.02 2) the tests run fine yesterday on 60fad8f82b
aleasto has joined #openwrt-devel
embargo has joined #openwrt-devel
<Pepes>
ynezz: I noticed the same on Turris Omnia right now, running the latest commit from 21:02 with netifd.
<Pepes>
It would be nice to do revert and then fix it properly
rejoicetreat has joined #openwrt-devel
decke has joined #openwrt-devel
FPSUserename has joined #openwrt-devel
<FPSUserename>
I'm back
aleksander has quit [Quit: Leaving]
<rsalvaterra>
hauke: Thanks for porting the gc-sections patch fix to 5.4. I assumed it was so bleeding edge it wouldn't matter… :)
Tapper has quit [Ping timeout: 480 seconds]
rejoicetreat has quit [Ping timeout: 480 seconds]
Tapper has joined #openwrt-devel
rejoicetreat has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
goliath has joined #openwrt-devel
goliath has quit []
aleksander has joined #openwrt-devel
valku has joined #openwrt-devel
aiyion has quit [Remote host closed the connection]
aiyion has joined #openwrt-devel
Slimey has joined #openwrt-devel
goliath has joined #openwrt-devel
Tusker has quit [Quit: Time wasted on IRC: 14 hours 31 minutes 21 seconds]
rsalvaterra_ has joined #openwrt-devel
rsalvaterra has quit [Ping timeout: 480 seconds]
rsalvaterra_ has quit []
rsalvaterra has joined #openwrt-devel
Rentong has joined #openwrt-devel
Rentong has quit [Remote host closed the connection]
Rentong has joined #openwrt-devel
Rentong has quit [Remote host closed the connection]
Rentong has joined #openwrt-devel
Rentong has quit [Remote host closed the connection]
rejoicetreat has quit [Ping timeout: 480 seconds]
Rentong has joined #openwrt-devel
Rentong has quit [Remote host closed the connection]
Rentong has joined #openwrt-devel
Rentong has quit [Ping timeout: 480 seconds]
FPSUserename has quit [Remote host closed the connection]
Rentong has joined #openwrt-devel
Rentong has quit [Remote host closed the connection]
rsalvaterra_ has joined #openwrt-devel
rsalvaterra has quit [Ping timeout: 480 seconds]
decke has quit [Quit: Leaving.]
rsalvaterra_ has quit []
rsalvaterra has joined #openwrt-devel
<rsalvaterra>
Hm… Is it just my dad's laptop, or do RTL8821AE cards really get angry with interrupt remapping in x2APIC mode?
<rsalvaterra>
And I mean "angry" as in "not a single interrupt is fired".
FPSUserename has joined #openwrt-devel
<FPSUserename>
Habbie, did you read the flash yet or will you (hopefully) do it tonight?
<Habbie>
FPSUserename, dunno yet, i'm working days -and- nights this week
<FPSUserename>
oof
<fda>
after every sysupgrad (keep settings), dropbear starts automatically even it was disabled before. as dropbear does not handle ipv6 properly im using xinetd for it.
<fda>
how could i prevent dropbear to start at boot? it starts before xinetd and gets the port first
<PaulFertser>
What's wrong about dropbear's IPv6 handling btw?
Borromini has joined #openwrt-devel
rejoicetreat has joined #openwrt-devel
<Habbie>
FPSUserename, yeah - US timezone conference, i made this poor choice myself ;)
<wb9688>
IETF111
<Habbie>
FPSUserename, how did you open it, btw?
<fda>
PaulFertser: if you set an "interface" dropbear should listen on, it listens only on ipv4. the ipv6(s) get assigned later and dropbear does not restart
<fda>
xinetd listens on ::, and you could set "allowed hosts" etc
<PaulFertser>
fda: I see
<fda>
in git, xinetd does not listen at all! i fixed it localy only
<fda>
.. at ipv6
<Habbie>
FPSUserename, i just read the manual for the experia wifi, and the letter that came with it - it does really claim that it 'takes over' wifi from the experiabox, and that wifi on the experiabox gets disabled
<Habbie>
FPSUserename, it also suggests that if you don't have long cables, allekabels.nl has a nice selection :D :D
<fda>
and dropbear needs a patch to get inetd sipport too
gladiac is now known as Guest2319
gladiac has joined #openwrt-devel
Guest2319 has quit [Remote host closed the connection]
<fda>
so, any idea to prevent dropbear to start automatically?
gladiac has quit [Remote host closed the connection]
<wb9688>
Habbie: Hahaha, wow, doesn't KPN also sell cables nowadays? Unfortunately I don't sell *long* cables
<wb9688>
But stupid to recommend allekabels.nl (unless they got paid for that) and now with how they handled that data breach I would never order there…
gladiac has joined #openwrt-devel
<Habbie>
FPSUserename, ok, opened it, only broke one clip
<Habbie>
FPSUserename, and, how did you remove the heatsink? i don't feel like doing my usual brute force approach there ;)
<Habbie>
lol, the stock firmware calls itself, among other things, MT7621.local
<Habbie>
FPSUserename, btw, the web interface claims i can download firmwares files from KPN, but I don't see them
netprince has quit [Remote host closed the connection]
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
aleasto has quit [Remote host closed the connection]
<FPSUserename>
Habbie, I still have a 20M ethernet cable, so I'm good with that
<FPSUserename>
What chip did you break
<FPSUserename>
allekabels is quite an expensive shop by the way, plus the data leak they had lmao, my data wasn't part of it gladly. I only bought a light for the oven from there once
<Habbie>
FPSUserename, i did not break a chip, just a black clip
<FPSUserename>
Habbie, the heatsink can be removed by unscrewing the screws and then try to twist it a bit while pulling.It really feels like the heat pads created a vacuum
<Habbie>
ah, the screws are on the bottom!
<FPSUserename>
wb9688 the extender works by copying the modem's configuration and exend the network. The issue on the older modems (before the experia box v10a), it would completely take over the network, meaning that you'd need to put one of the extenders near the modem. Not needed on the newer ones. I think I have a v12 modem
<FPSUserename>
And I think it's just a limitation on the older modems. I bet that the ISP is too lazy to give an update that fixes it. The ISP would rather see people ask for new equipment
<FPSUserename>
Habbie yeah, I saw that people already snooped around the v8 and v10 modems for openwrt. Lets hope that this v8 openwrt guide will be of any help to us for the extender
<FPSUserename>
Basically this extender is a mini router, so I don't really see a challenge with this one, just need to get through the password protection and hope that an openwrt image based on the mt7621 chip will work, if not, someone would have to make one. Not a programmer, so I can't help on that regard, besides testing
<wb9688>
FPSUserename: Could you get access to the shell (as in like bash)?
<FPSUserename>
Not without the password, any boot option other than the default (3, boot from flash) requires the password
<FPSUserename>
Unfortunately, the extender doesn't even have a web interface, and I doubt that you could SSH into it (I mean, if you know the username and possible password, plus if SSH is enabled)
<FPSUserename>
Oddly enough, the sticker on the bottom of the extender shows the SSID, the wireless key AND the web password
<FPSUserename>
Now what that web password may be, god may know
<FPSUserename>
It could be that the extender does have a web interface, but not accesible to me, because I use custom DNS servers, which KPN routers don't like (the modem's web interface isn't accessible either with custom DNS servers, even though navigating with the modem's IP address)
FPSUserename has quit [Remote host closed the connection]
FPSUserename has joined #openwrt-devel
<FPSUserename>
hmm, tried connecting to it through wifi and wired, no luck
<wb9688>
Does it not expose UART somehow?
<wb9688>
On the board
<Habbie>
FPSUserename, the web interface is on 192.168.2.150 - it does not serve DHCP
<Habbie>
i can see the webinterface from here on my other desk ;)
<FPSUserename>
Well, the odd thing is that the UART doesn't post anything after SDK 5.0.S.0
<Habbie>
5.0.S.0?
<FPSUserename>
while I've seen similar boot logs from the MT7621 chip on the openwrt page that start linux after it
<FPSUserename>
Just look at the line where it says "SDK 5.0.S.0". My log ends there, nothing else is outputted on the uart
<Habbie>
any change the speed changes there?
<FPSUserename>
While that mi router spits out linux related stuff
<PaulFertser>
Usually with a wrong baud you'd see at least something
<Habbie>
PaulFertser, also my experience, to be fair yes
<tmn505>
FPSUserename: have You tried shorting clock and data out pin of NOR chip just before kernel is read from flash?
<FPSUserename>
You mean the baudrate? I don't think so. Usually when the baudrate is different, you'll get incorrect ascii chars on your display, but it's just empty, not even newlines
<FPSUserename>
I mean, I already had to play a bit with the baudrate to see the correct stuff
<FPSUserename>
How does the web interface of the extender look like? Did let your modem configure it or just using it standalone, like it came out of the box?
<Habbie>
the extender has never seen my modem
<PaulFertser>
tmn505: that might work
<Habbie>
i took it out of the box and did not connect it to the internet
<Habbie>
it has css files with _voda_ in the name
<Habbie>
i'll make a picture
<PaulFertser>
Before or during to make CRC calculation fail.
<PaulFertser>
But probably it'll just reboot
<tmn505>
possible, but You never know.
<PaulFertser>
Who knows, probably worth checking even though it's a bit risky hardware-wise.
<FPSUserename>
so I might have a faulty unit perhaps, software wise. I mean, my boot log does show "old glbcfg partition is broken, magic=[0xffffffff][0xffffffff]!"
<Habbie>
what would be the point of that shorting?
<tmn505>
PaulFertser: usually that's the first thing I do before soldering.
<PaulFertser>
Habbie: forcing CRC error hoping that the bootloader will give a prompt failing to load the image
<Habbie>
ah
<Habbie>
and that prompt might not be password protected?
<PaulFertser>
If they're lazy and you're lucky
<Habbie>
right
<Habbie>
nice
<FPSUserename>
Hmm, booted my other extender, but it also just doesn't let me see a web interface
<Habbie>
FPSUserename, did you configure an IP in 192.168.2.0/24 on your computer?
<FPSUserename>
Very unfortunate that I can't make use of the wifi to ethernet sharing, because windows puts it on 192.168.137.x instead of keeping it to 192.168.2.x. Can't access the interface that way
bookworm has quit []
bookworm has joined #openwrt-devel
FPSUserename has quit [Remote host closed the connection]
FPSUserename has joined #openwrt-devel
<FPSUserename>
Tried bridging, but I just lost wifi for a sec
FPSUserename_ has joined #openwrt-devel
FPSUserename has quit [Remote host closed the connection]
FPSUserename_ has quit []
FPSUserename has joined #openwrt-devel
<FPSUserename>
Habbie, I bet someone with experience in networking could snoop out the file that the extender pulls when doing an update, if you use a machine in between the router and extender
<Habbie>
unless it uses https
<Habbie>
and you only get one shot
<Habbie>
and the update might lock the device down even more
<dwfreed>
yeah, but only if it validates the cert
<Habbie>
sure
<dwfreed>
mitmproxy for fun and profit
<Habbie>
and if the config backup button is any indication, the firmware might be encrypted with a key we don't have
<dwfreed>
if the firmware is encrypted, you have the key, you just may have to put a lot of work into extracting it
<Habbie>
universe-ending amounts of work, if you're unlucky
<dwfreed>
I mean, the device has to be able to decrypt the firmware, so the device has the key
<Habbie>
sure
Borromini has quit [Quit: leaving]
Weasel____ has joined #openwrt-devel
Weasel___ has quit [Ping timeout: 480 seconds]
<FPSUserename>
Habbie, any luck so far?
<Habbie>
heatsink is still on
<Habbie>
little time to tinker with hardware today
<FPSUserename>
okay. If only we'd have an insider that could leak a stock firmware build
Tapper has quit [Ping timeout: 480 seconds]
Tapper has joined #openwrt-devel
FPSUserename has quit [Quit: Page closed]
Acinonyx has joined #openwrt-devel
Acinonyx_ has quit [Ping timeout: 480 seconds]
danitool has joined #openwrt-devel
aleksander0m has joined #openwrt-devel
aleksander has quit [Read error: Connection reset by peer]