minimal has joined #openwrt-devel
<fda> well, wenn i get 5 ipv6 GUA prefixes on thernet plungged, i know it gets to much RAs :)
<dwfreed> But you don't know if Windows is just ignoring the tags or if they're not there :P
<fda> windows can ignores vlan tags :-\
<dwfreed> Windows is weird; wireshark at least lets you know that the driver reports there's a vlan tag
<fda> dwfreed: i've put the notebook on the switch and added the 802.1q column, its just empty
<fda> im wondering if its possible to see other vlans, maybe it depends on the device driver
<dwfreed> might be worth booting a linux rescue distro
<fda> burn a cdrom? im not sure if the cd writer is still dust free :)
<slh> USB sticks should work for most options as well
<fda> that does not work with this old device
<slh> in many cases you could even do a regular linux installation (exclusively-) to a USB stick (>=8 GB, the details of how to contain that might differ from distro to distro though)
<slh> anything since ~2005 should be able to boot from a USB stick
<fda> its a dual core
<fda> dont know from when
<fda> btw, if it is a vlan problem, why are ipv4-dhcp working?
<fda> im getting the correct ipv4 subnet depending on vlan
dangole has joined #openwrt-devel
Tapper has quit [Ping timeout: 480 seconds]
<fda> dwfreed: with ipv4 the correct dhcp answer: https://ibb.co/zm9FtPC + https://ibb.co/JpGtKKx
<mangix> slh: C7 uses ar8327. 37 is used by v4 and v5. What's funny is the DTS files claim they all use 27. Anyway, Ansuel's working on a 27 unit.
<mangix> It seems he's close to getting it working
<slh> yeah, I have a tl-wdr3600 and tl-wdr4300 myself (with ar8327), so I'm definately more than interested to follow ;)
goliath has quit [Quit: SIGSEGV]
Tapper has joined #openwrt-devel
rua has quit [Quit: Leaving.]
valku has quit [Quit: valku]
minimal has quit []
Tapper has quit [Ping timeout: 480 seconds]
danitool has quit [Ping timeout: 480 seconds]
victhor has quit [Ping timeout: 480 seconds]
dangole has quit [Ping timeout: 480 seconds]
Grommish__ has quit [Read error: Connection reset by peer]
Grommish has joined #openwrt-devel
<fda> seems multicase leaks of ipv6 in vlans is known for many router operating systems
<fda> multicast
<fda> so from 1 vlan you could send fakre RAs to another vlan
<fda> maybe it helps to disable "Enable multicast support" in openwrt (by default on!) for all interfaces etc
<aparcar[m]> Tusker: ping
kenny2 has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
f5 has quit [Ping timeout: 480 seconds]
f5 has joined #openwrt-devel
pmelange has joined #openwrt-devel
pmelange has quit [Quit: Leaving.]
rmilecki has joined #openwrt-devel
Borromini has joined #openwrt-devel
<mangix> hmmm
<mangix> ed25519 is shown as
<mangix> ECDH, Curve25519 in LuCI. This is totally wrong
<mangix> it uses neither ECDH nor Curve25519
<mangix> Curve25519 is a Montgomery curve. ed25519 uses an Edwards curve
rua has quit [Quit: Leaving.]
<PaulFertser> mangix: isn't curve25519-sha256 KEX used?
<PaulFertser> And rsa-sha2-256 as host key algo
<PaulFertser> debug1: expecting SSH2_MSG_KEX_ECDH_REPLY
rua has joined #openwrt-devel
<PaulFertser> So curve25519-sha256 is a variant of ECDH key exchange algorithm.
<PaulFertser> And it's used for key exchange. Not the actual key.
rua has quit []
rua has joined #openwrt-devel
rua has quit []
rua has joined #openwrt-devel
rua has quit []
rua has joined #openwrt-devel
rua has quit []
rua has joined #openwrt-devel
<mangix> PaulFertser: sure but it has nothing to do with ed25519 keys
<mangix> the curve used in ed25519 is birationally equivalent to curve25519 but not the same
<mangix> I'm specifically talking about the LuCI ssh keys menu
<PaulFertser> mangix: if LuCI is talking about authorized keys with that wording then it sounds really wrong, I agree.
<PaulFertser> I've never seen it :)
<mangix> I sent a PR for it
robin_ has quit [Ping timeout: 480 seconds]
wvdakker has joined #openwrt-devel
<mangix> rsalvaterra: did you ever send your zstd jffs2 patches
<rsalvaterra> mangix: Wow, I totally forgot about that!
<rsalvaterra> Yeah, I did…
<rsalvaterra> … but nobody came.
<rsalvaterra> (Couldn't resist the Undertale reference. :P)
<rsalvaterra> No comments or replies.
* rsalvaterra goes to double check the email
<rsalvaterra> mangix: Oh, I remember now: https://marc.info/?l=linux-kernel&m=161942697517216&w=4
<rsalvaterra> I got stuck here, basically.
<rsalvaterra> To be honest, I don't know exactly if the userspace changes are strictly needed. dwmw2_gone?
pmelange has joined #openwrt-devel
wvdakker has quit []
rmilecki has quit [Ping timeout: 480 seconds]
danitool has joined #openwrt-devel
wvdakker has joined #openwrt-devel
pmelange has quit [Quit: Leaving.]
goliath has joined #openwrt-devel
Acinonyx_ has joined #openwrt-devel
Tapper has joined #openwrt-devel
Acinonyx has quit [Ping timeout: 480 seconds]
rua has quit [Remote host closed the connection]
rua has joined #openwrt-devel
fda- has joined #openwrt-devel
victhor has joined #openwrt-devel
fda has quit [Ping timeout: 480 seconds]
fda- has quit [Quit: ZNC - https://znc.in]
fda has joined #openwrt-devel
Borromini has quit [Ping timeout: 480 seconds]
<mangix> rsalvaterra: so not upstreamed. Bummer.
jhepp has joined #openwrt-devel
Borromini has joined #openwrt-devel
<jhepp> Hello! When a new openwrt Release is flashed via sysupdate then all packages are deleted. I created a script which reinstalls the packages after such an update. I find this very helpful for devices connected only over vpn which can't be access without wireguard app installed. How can I give this script back to openwrt? Regards Jochen
<olmari> jhepp: Usually such advanced users compiles own openwrt with wanted packages already in, or uses imagebuilder to achieve same.. That being said, I dno't see why any tool that can ease life is bad :)
<jhepp> Hello olmari! I have multiple devices with different packages on each device. So I take the official openwrt sysupgrade image and flash all with the same. They afterward install each the packages they need and are updated.
fda- has joined #openwrt-devel
fda| has joined #openwrt-devel
fda has quit [Ping timeout: 480 seconds]
<Borromini> jhepp: you can script the imagebuilder to do that before you, as olmari suggested
<rsalvaterra> mangix: Yeah. Not yet, at least.
fda- has quit [Ping timeout: 480 seconds]
<jhepp> Borromini: Isn't it easier to install only the official openwrt images and have a e.g. /etc/package text file which contains the name of the packes in uci-syntax and use a bash script to install all missing packages vi opkg instead of compiling an image for each device?
rua has quit [Quit: Leaving.]
<Borromini> i'd go for the imagebuilder option, but of course you already have a framework in place. so there's a small learning curve
<jhepp> Borromini: I thought the script would be helpful for others which flash the original images of openwrt and then install packages and customize there device. Via sysupgrade all customized files are still there but the packages are missing.
<Borromini> jhepp: of course it would, and you can share it on the forum e.g.
<Borromini> all we're saying is this usecase is covered by the imagebuilder, so standard OpenWrt tools, if you'd wish
jhepp has quit [Quit: Page closed]
jhepp has joined #openwrt-devel
<jhepp> Borromini: sorry firefox restared because of update
<Borromini> what did you get?
<Borromini> from messages i mean, last timestamp
<jhepp> Borromini: thank you for the advice with the forum I didn't know and try to use it - and I look who to customize imagebuilder but I think this is too cumbersome for me
<jhepp> Borromini: the last conversation is cleared before my new login but I remebered it in my head
Tapper has quit [Ping timeout: 480 seconds]
<Borromini> ttps://paste.debian.net/1211345/
<Borromini> sorry
<jhepp> Borromini: the imagebuilder with the use of PACKAGES= would have the same result as my script does with the cost of compiling an image for each device with different packages or one with all packages wanted
<jhepp> Borromini: thank you for the pasted conversation!
<Borromini> yw. it's a tradeoff between cycles and bandwith on your computer or on your network devices
rua has joined #openwrt-devel
rua has quit []
Borromini has quit [Quit: leaving]
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
rmilecki has joined #openwrt-devel
<jhepp> Borromini: I searched i the forum and found a relevant thread where I added it
<jhepp> Borromini: thank you very much with your advice
fda has joined #openwrt-devel
fda| has quit [Ping timeout: 480 seconds]
fda- has joined #openwrt-devel
fda has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
fda has joined #openwrt-devel
fda- has quit [Ping timeout: 480 seconds]
rua has quit [Quit: Leaving.]
rua has joined #openwrt-devel
rua has quit []
wvdakker has quit []
wvdakker has joined #openwrt-devel
rua has joined #openwrt-devel
Tapper has joined #openwrt-devel
jhepp has quit [Remote host closed the connection]
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
kenny2 has joined #openwrt-devel
dedeckeh has joined #openwrt-devel
rua has quit [Remote host closed the connection]
rua has joined #openwrt-devel
<kenny2> I'm looking at my tplink C2V5. It's similar to the C1V1 but they've switched the switch chip to a qca8337. Looking at dts files I see some collection of magic `qca,ar8327-initvals` values. Any suggestions on how to find these values?
<hauke> xdarklight: there is a locking problem in the lantiq PCI driver: https://pastebin.com/N39R9XJV
<hauke> this could cause a daedlock which could make the device hang
<hauke> xdarklight: I think you already looked into cleaning up the driver
Borromini has joined #openwrt-devel
Borromini has quit [Quit: Lost terminal]
wvdakker has quit []
wvdakker has joined #openwrt-devel
wvdakker has quit []
wvdakker has joined #openwrt-devel
wvdakker has quit []
wvdakker has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
<wvdakker> exit
<wvdakker> quit
wvdakker has quit []
Tapper has quit [Ping timeout: 480 seconds]
ecloud has quit [Ping timeout: 480 seconds]
ecloud has joined #openwrt-devel
danitool has joined #openwrt-devel
wvdakker has joined #openwrt-devel
minimal has joined #openwrt-devel
ecloud has quit [Ping timeout: 480 seconds]
<mkresin> hauke: xdarklight: I'm playing with xrx200 DSA on kernel 5.4.143 and I'm not sure if I'm doing something wrong or hitting an corner case which isn't implemented
ecloud has joined #openwrt-devel
<mkresin> hauke: xdarklight: HH5a, lan1-lan4 are bridge member. wan should be used standalone with a vlan tag. But 'ip link add link wan name wan.2 type vlan id 2' ends with RTNETLINK answers: Not supported
<mkresin> hauke: xdarklight: Am I doing something wrong or is it supposed to work this way?
<fda> as im using ULAs with NAT6 (no PD) i disabled RAs. slaac-only android gets then no ipv6
ecloud has quit [Ping timeout: 480 seconds]
Tapper has joined #openwrt-devel
<dwfreed> fda: nice
<fda> not really ...
<hauke> mkresin: DSA does not really support the classical VLAN interfacs, but you can do VLAN filtering on a brigde with the bridge command
ecloud has joined #openwrt-devel
<mkresin> hauke: okay, that is unexpected. till now my understanding was that DSA adds the dsa header to distinguish from/to what switch port an ethernet packet should travel
<mkresin> hauke: there might be done more done/offloaded in case a port is a member of a software bridge
kenny2 has quit [Ping timeout: 480 seconds]
<hauke> mkresin: yes the prots are seperated based on the cpu tag added to the packet
<hauke> *ports
<hauke> but the vlan handling is vlan filtering
<mkresin> hauke: and vlan filtering is possible/configured via vlan aware software bridge?
<hauke> yes
<mkresin> means, add br-lan [lan1, lan2, lan3, lan4] and br-wan [wan]. the vlan config is done one br-wan.
<hauke> yes
<hauke> but I think "ip link add link wan name wan.2 type vlan id 2" should work if it is not in a bridge
<hauke> then it is done in software
<mkresin> hauke: yes, that was my expectation as well. but beside the interface not being in a bridge, it doesn't work
<hauke> hmm strange
<mkresin> one would had to messure what is better performance wise. adding the vlan tag in hardware with the downside that the paket need to travel the bridge code vs. vlan tag added in software but no bridge code which the paket needs to travel along
<hauke> mkresin: when you add a new interface with a vlan like wan.2 on top of wan you can send packets on both interfaces
<hauke> the vlan tag is not stored in the switch specail tag as far as I remember
<hauke> the hardware would not be able to differentiate them
<hauke> some hardware supports vlan offlaoding, but I think the xrx200 does not spport it
Borromini has joined #openwrt-devel
Borromini has quit [Quit: Lost terminal]
kenny2 has joined #openwrt-devel
Tapper has quit [Ping timeout: 480 seconds]
Grommish has quit [Read error: Connection reset by peer]
Tapper has joined #openwrt-devel
minimal has quit []
dedeckeh has quit [Remote host closed the connection]
PaulFertser has quit [Ping timeout: 480 seconds]
PaulFertser has joined #openwrt-devel
rmilecki has quit [Ping timeout: 480 seconds]
wvdakker1 has joined #openwrt-devel
wvdakker1 has quit [Remote host closed the connection]
wvdakker1 has joined #openwrt-devel
wvdakker1 has quit [Remote host closed the connection]
wvdakker has quit []
wvdakker has joined #openwrt-devel
wvdakker has quit []
wvdakker has joined #openwrt-devel
goliath has joined #openwrt-devel
cp- has quit [Read error: Connection reset by peer]