<hurricos>
OK, so no worries. The module is loaded.
<hurricos>
Your device is simply not connected.
<hurricos>
Dig through lsusb a lot further to find the reason
<FriendlyNGeeks>
lsusb shows | bus 001 device 005: id 0dba: 8151 Realteck usb 10/100 lan
<FriendlyNGeeks>
thats the thing its always connected since its part of the usb hub. is there an indepth lsusb command
<FriendlyNGeeks>
so i can trace
<hurricos>
if it's there and the module is too, then the question is why the module hasn't bound to the device
<hurricos>
I can't see it in e.g. `strings /lib/modules/5.10.135/r8152.ko`, but there should be a list of vendor/device names built into the kernel module which describe whether it will bind.
<FriendlyNGeeks>
right again my reasoning says either booting from usb even SD card reader is drawing two much power some how or its just loading so fast its not grabbing it
<FriendlyNGeeks>
it binds fine when i load the same image from SD card on the pi
<hurricos>
try `rmmod r8152; kmodloader`?
<FriendlyNGeeks>
when i burn image to usb or move SD card to card reader usb and change command line device to sda2 so it can find root... nope it doesnt load the lan port
<hurricos>
Any luck?
<hurricos>
RE: `rmmod r8152; kmodloader`?
<hurricos>
Let me know if that doesn't work. Thanks.
<FriendlyNGeeks>
so it unregistered and then loaded all modules usbcore: registered ***
<FriendlyNGeeks>
but ip link show still nothing
<FriendlyNGeeks>
when it loads from sd card i get about 4 lines showing its assigning eth0 during boot
<hurricos>
Right, I don't really know what SD card does, or care really. I'm trying to suss out what about the two is different. The same might be true of network booting.
goliath has quit [Quit: SIGSEGV]
<hurricos>
I'm not sure. I know a lot less about the boot implementation than I do about what is required for the module to load and bind to the device
<FriendlyNGeeks>
ahh fair enough
<hurricos>
I'm worried now that there are potentially more layers to peel back here.
<hurricos>
Is r8152-firmware installed?
<hurricos>
If it is, please try 'rmmod r8152; rmmod cdc_ncm; rmmod cdc_ether; rmmod usbnet; kmodloader;'
<hurricos>
RE: "more layers to peel back".
<FriendlyNGeeks>
firmware files say everything excetp r8152 ex
<hurricos>
I mean the OpenWrt package r8152-firmware.
<hurricos>
i.e. `opkg list-installed | grep -we r8152-firmware`.
<hurricos>
or `opkg info r8152-firmware`
<FriendlyNGeeks>
rtl8153a,c,56
<hurricos>
?
<hurricos>
Can you `opkg info r8152-firmware`?
<FriendlyNGeeks>
in the firmware folder
<FriendlyNGeeks>
sorry flipping inputs on monitor brb
<hurricos>
I know that some other files might be there, again just trying to suss out differences.
<FriendlyNGeeks>
yep its installed from that line
<FriendlyNGeeks>
but in folder im not seeing 8152.fw
<hurricos>
Shrug. Did you try the full `rmmod r8152; rmmod cdc_ncm; rmmod cdc_ether; rmmod usbnet; kmodloader;`?
<hurricos>
Where is the USB mounted at this point, is it your rootfs?
<FriendlyNGeeks>
so i have two copies one usb is mounted in my vm so i can look through files the sd card is in the pi so i can run cmd lines
<FriendlyNGeeks>
same image
<hurricos>
I mean to say, what's the output of `mount` on OpenWrt?
<FriendlyNGeeks>
i will try unregister all of that brb
<hurricos>
There's some difference. I'm sure it's the mount order, or at least if I had to guess. I don't know what your setup is, nor do I want to try and reproduce it, I just want to come up with some way to manually articulate the attachment of the network interface to the device, so we can then ask precisely how that normally happens to say with certainty why it isn't happening here
<hurricos>
If you can get that, and you should be able to, then you'll make everyone else's job feasible
<FriendlyNGeeks>
i unregistered each by hand then ran kmodloader still nothing on ip addr show
<hurricos>
:shrug: Not sure at this point, then. It must not be a mount order issue, then, but something much more subtle
<FriendlyNGeeks>
okay is it possible for me to boot off sd card and output dmesg to file somehow
<hurricos>
I'm guessing build issue which is why I'm bailing on trying to fix this
<FriendlyNGeeks>
hmm ive tried it with the offical image builder as well 22 or 23 i believe
<FriendlyNGeeks>
same thing happens
<hurricos>
it's not official if it's not mainline
<hurricos>
You said you were adding somesuch
<hurricos>
If you have a separate Linux machine running systemd / NetworkManager, you can always USB-network to it.
<hurricos>
Use USB -> power port on the rpi0
<hurricos>
The USB will become a gadget device for the Linux machine.
<hurricos>
It'll have the right IDs to load usb_cdc / ncm / cdc_ether
<FriendlyNGeeks>
crazy thing is if i plug in a microsoft usb-nic it shows that one and maps eth0
<hurricos>
I'm sure it's a different device. This is why I'm suggesting your issue is more sinister
<FriendlyNGeeks>
the firmware selector i used to try as well was 21.02.3
<FriendlyNGeeks>
most def sinister.
<FriendlyNGeeks>
oh its powered from rpi0 right now ive tried poe and direct power same effect
<FriendlyNGeeks>
the only other thing i could try is load ubuntu on usb in the hub and see if it auto detects the eth0 in linux or install driver reboot and see if i can rule out power issue from booting from usb
<hurricos>
No, I mean that if you want to network with the device to export dmesg, you can use usb from a laptop to the USB "power port"
<FriendlyNGeeks>
put the pic of the poe hat chip is correct | im going to fresh image on sd card boot up make sure nic shows up then see if the firmware folder shows a file for 8152 brb
<FriendlyNGeeks>
okay so loaded fresh from sd card lib/firmware* still doenst show an exact file for r8152 but shows the others r8153----etc but the nic had loaded
<FriendlyNGeeks>
attempting to change cmd line text to sda2 and boot from usb on hub
SlimeyX has quit [Read error: Connection reset by peer]
Lechu has quit [Remote host closed the connection]
<FriendlyNGeeks>
@hurricos thanks for trying to troubleshoot earlier
Lechu has joined #openwrt-devel
SlimeyX has joined #openwrt-devel
danitool has quit [Ping timeout: 480 seconds]
<hurricos>
oh my god.
<hurricos>
PCIe paths have changed on AP3825i
<hurricos>
WTF
<hurricos>
from kernel version 5.10.135 -> 5.10.138
<hurricos>
hmm, maybe we should have a migration tool that attempts to backtrack netdev paths?
srslypascal has quit [Remote host closed the connection]
srslypascal has joined #openwrt-devel
goliath has joined #openwrt-devel
<hauke>
hurricos: how did the name change?
<Misanthropos>
fyi: I tried to setup mesh/SAE with rc6, tyring wpad-mesh-wolfssl and wpad-mesh-openssl - both failed with encryption: none. I went back to 21.2.3 and the same setup works.
robimarko has joined #openwrt-devel
<ynezz>
hauke: it's all in #10530
<ynezz>
tl;dr archaic platform problem
robimarko_ has joined #openwrt-devel
robimarko has quit [Ping timeout: 480 seconds]
<ynezz>
hauke: I'm wondering how could we QA kernel version upstep backports given the recent build failures in mt7622 and bcm27xx
Misanthropos has quit [Ping timeout: 480 seconds]
<ynezz>
hauke: I'm simply afraid, that if we're getting those build failures, I don't want to see the results of runtime testing failures
<f00b4r0>
the backport encap is only needed if mt76 isn't updated; nbd ack'd it.
<ynezz>
f00b4r0: IIRC I've already pushed those, so if you want to move forward with that remaining one I've referenced in that previous message, then please rebase it, thanks
Borromini has quit [Ping timeout: 480 seconds]
Misanthropos has joined #openwrt-devel
<hauke>
ynezz: you have to refersh the kernels also on 22.03
<hauke>
you could backport the github CI changes from master to 22.03 and create github pull request to build everything
<ynezz>
so the current idea is to copy&paste the CI stuff?
<hauke>
that would be my idea
Misanthropos has quit [Ping timeout: 480 seconds]
<hurricos>
ynezz: You can't do me like that
<hurricos>
:(
<hurricos>
I like my powerpc ....
<ynezz>
hurricos: sorry?
<ynezz>
what did I broke this time? :)
goliath has quit [Quit: SIGSEGV]
<hurricos>
Just the "archaic platform problem" bit.
<hurricos>
;^)
<hauke>
ynezz: are you currently fixing bcm27xx?
<hauke>
otehrwise I will make the patches apply again on 22.03
danitool_ has joined #openwrt-devel
Misanthropos has joined #openwrt-devel
danitool has quit [Ping timeout: 480 seconds]
<f00b4r0>
ynezz: unless I'm mistaken there's nothing to rebase. You already pushed everything :)
<f00b4r0>
nbd: thanks for the link, I'll take a look
Borromini has joined #openwrt-devel
<hauke>
ynezz: I refreshed the bcm27xx patches now too
Borromini has quit [Ping timeout: 480 seconds]
<ynezz>
hauke: yes, I'm on it.
<ynezz>
hauke: it doesn't compile, so need to take another attempt
<ynezz>
but if yours is ready, feel free to push it
<rmilecki>
f00b4r0: nbd: oh, do we have "unetd" now? another undocumented project? :/
<rmilecki>
it'll make a nice company to usteer, ucwmp, ucentral
<mrnuke>
hurricos: I'm silently waiting for the poe/podl patchset to be merged in a -next branch before I start messing with it
SamantazFox has joined #openwrt-devel
mcbridematt has quit [Ping timeout: 480 seconds]
<hauke>
ynezz: jow: Is there still something blocking a OpenWrt 22.03 final release?
<hauke>
Should we better do an RC7 or a final? I think it is ok to do final
<hauke>
we should wait till ~ Wednesday till becasue of all the recent changes
<rotanid_>
regarding this i wrote on 2022-07-07 over in openwrt-adm :
rotanid_ is now known as rotanid
<rotanid>
"i think it would be good if the number of changes between a release and the last RC isn't as high as between e.g. rc4 and rc5, because otherwise it's not really an "RC" anymore and the risk of introducing new problems between RC and release is high..."
jamzz has joined #openwrt-devel
mcbridematt has joined #openwrt-devel
csharper2005 has joined #openwrt-devel
<csharper2005>
Hi guys! Is it enough to add luci to the master branch: echo 'CONFIG_PACKAGE_luci-ssl=y' >> .config ?
<PaulFertser>
And "make defconfig" after that.
<PaulFertser>
Should work if the feed was installed.
<csharper2005>
PaulFertser: sure (defconfig). I'm experimenting with gh action...
<csharper2005>
22.03 build for mt7621 takes ~5 hours.
<csharper2005>
llvm compilation eats almost half the time. lol
<robimarko_>
Are you maybe using the Buildbot config?
<robimarko_>
That would explain LLVM getting built
<csharper2005>
robimarko_: yeah. official config.buildinfo
<robimarko_>
Yeah, that is why you are getting LLVM built
guidosarducci_ has quit [Remote host closed the connection]
<csharper2005>
robimarko_: What will be lost if I remove llvm: CONFIG_BPF_TOOLCHAIN_BUILD_LLVM=y, CONFIG_SDK_LLVM_BPF=y, CONFIG_USE_LLVM_BUILD=y?
<robimarko_>
BPF stuff wont work
cbeznea has joined #openwrt-devel
jamzz has quit [Remote host closed the connection]
minimal has joined #openwrt-devel
<hurricos>
hauke: RE: final release, are there any plans to bump kernel past 5.10.135?
<hurricos>
I'm hoping not, so as to avoid mpc85xx borkups
Misanthropos has quit [Ping timeout: 480 seconds]
<hauke>
hurricos: it is already done
<mrnuke>
hurricos: What is your opinion on tagging realtek-poe-1.0 ?
Misanthropos has joined #openwrt-devel
jamzz has joined #openwrt-devel
<hurricos>
mrnuke: Probably, "tag it now while it's still hot".
<hurricos>
I will go do it and submit the packages pr for it ...
<hurricos>
if you can tell me how to configure /etc/config/network and /etc/config/firewall for a NAS device which needs a public IPv6 address :^)
<hurricos>
I'm guessing, a wan6 interface on the same eth0 device as lan, and manually configure /etc/config/firewall for services I want?
<hurricos>
hauke: Bump already applied to both, huh? Rats, OK.
<hurricos>
I don't want that address rotating, I have to test a quick fix using linux,pci-domain in the device tree
<hurricos>
alright, time to YOLO. If I lose connectivity with my NAS I'll never forgive myself
<hurricos>
it came back.
<hurricos>
even trippier, I no longer am resolving the ipv6 address. Nice! I guess the network restart caused a withdrawl of the ipv6 records. Can't complain
<f00b4r0>
reminds me I wanted to check what SoC is on the Zyxel GS1200-8HP: the onboard sw is shitty as hell, I'd sure like to run openwrt. Too many things to do, too little time ;)
<jamzz>
csharper2005: Thanks for the tip. The problem is I've added it to the DEVICE_PACKAGES option in the build file, but it still doesn't want to build with the image
<jamzz>
I'm trying to add support for other Arcadyan routers, would like to include it
dangole has quit [Ping timeout: 480 seconds]
Larhzu has quit [Quit: leaving]
<csharper2005>
jamzz: delete .config and make "make menuconfig" after DEVICE_PACKAGES update
cbeznea has quit [Ping timeout: 480 seconds]
<mangix>
hauke: I vote for release instead of another RC
<csharper2005>
jamzz: Arcadyan WG460223? Do you know what wireless chips are installed? mt7915 I guess
<jamzz>
WG630223, almost identical as MTS WG430223. MT7915 is correct
<csharper2005>
Mangix: me too. Updated ath10 driver was rejected :( 22.03 is stable enough
<mangix>
updated ath10k driver?
<jamzz>
I've submitted a PR on github, which had DEVICE_PACKAGES set to include openssl and uencrypt. Will have to fix that later...