Nice!

hi, i'd like to buy driver support for RTL8125

... lol

for real though. its 2026, who tf is still using intel nics.

on purpose

i had loads of fun test driving omni in a virt-machine, just to find out i cant use it in my homelab now.

do you have time and brain to port the drivers from an OS that does support it

i was just investigating it. there are BSD drivers. (FreeBSD doesnt support it though, just tried). time is my worst enemy, thats why i said id like to buy lol

time is also limiting for anyone who you'd buy from

AnomicBSD: FWIW, I have always had better experiences with Intel NICs than Realtek NICs

even in 2026

hmm. Porting a driver sounds like a dope learning exercise.

alas, all my stuff is supported it seems.

SarahMalik at least he's offering money, which is nice.

fair.

if i'd the skills and the energy I'd consider taking them up.

jclulow: thanks for your update (omniosorg/illumos-omnios #1633), but after 28 days of uptime, that server just crashed again! :(

I do not have the clue if it's because of the same cause, but here is the dump: paste.omnios.org/?37ae78120437b76c#…W6NMyibRWtP2YRiZUKGnBEbA1M5rgtCJwaC

thats pretty amusing. stack is eaten away, but with mdb fffffe007aa94b50::dis should give some light. but fr_tcp_age only ever calls fr_movequeue and there are two cases where mutex_enter() is called, both get mutex address with & operator - therefore either oifq or nifq must be NULL.

My guess would be that we get NULL from ip_state.c:3928 from tqe->tqe_ifq see also how queues are handled in fr_setnatqueue()

nat_update() does not seem to check for NULL either...

warden: I agree it's almost certainly a separate cause at this point yeah

Presumably it is illumos.org/issues/15555 which may also be the same as illumos.org/issues/17584

→ BUG 15555: ipf: NULL pointer dereference in fr_tcp_age() / fr_movequeue() (New)

→ BUG 17584: Null pointer leading to a crash when using NAT66 (New)

jclulow: thanks, I’m absolutely clueless on this matter, but it might be someway useful to know that I’ve the vague feeling that it happened after I configured IPv6 rules (both filter and NAT). I’m not 100% sure because IPv6 was enabled just a few weeks after that server was put in production.

win 32