> On 28 Oct 2024, at 23:59, Miguel Landaeta <mig...@miguel.cc> wrote: > > On Wed, Oct 23, 2024 at 08:07:42AM +0000, Vitaliy Makkoveev wrote: >> On Mon, Oct 21, 2024 at 01:44:58PM +0200, Kirill A. Korinsky wrote: >>> On Mon, 21 Oct 2024 11:19:48 +0200, >>> "Anthony J. Bentley" <bent...@openbsd.org> wrote: >>>> >>>> Vitaliy Makkoveev writes: >>>>> I suspect vio(4). >>>> >>>> Maybe there are multiple bugs. But as we've discussed before, wg(4) is >>>> crashy on em(4) too. >>>> https://marc.info/?l=openbsd-bugs&m=171615647816882&w=2 >>>> >>> >>> I'd like to add one more issue with wg: >>> https://marc.info/?l=openbsd-bugs&m=172028863807274&w=2 >>> >>> -- >>> wbr, Kirill >> >> Does -tcplro on em(4) and vio(4) helps? Is it possible to test them >> witout ipv6 with -tcplro? > > I also got hit by this bug, I can provide a dmesg as data point > although my crash is almost the same as the one reported by > Kirill A. Korinsky. > > Thanks for pointing out -tcplro as possible mitigation. > > I got my VPS to stop crashing minutes after booting 7.6 by doing > any of the following: > > - Do not set up any wg(4) interface on the host. > > - Set up vio(4) interfaces with -tcplro and enable wg(4) interfaces as > usual (In my case, I'm routing IPv4 and IPv6 traffic through wg(4) > tunnel). >
So, problem lays in vio(4). > I can provide more info if required. > > After I disabled tcplro on my interfaces my box is stable as usual. > Fortunately, I didn't have to revert to 7.5. > > -Miguel >
