hi Hrvoje,

Dec 26, 2022 08:46:43 Hrvoje Popovski <[email protected]>:

> On 26.12.2022. 5:06, Csillag Tamas wrote:
>> hi,
>>
>> the crash repeated again
>>
>> uvm_fault(0xffffffff823ed470, 0x0, 0, 1) -> e
>> fatal page fault in supervisor mode
>> trap type 6 code 0 rip ffffffff81e4c208 cs 8 rflags 10246 cr2 0 cpl 0 rsp 
>> ffff8000225a4060
>> gsbase 0xffff80001e119ff0  kgsbase 0x0
>> panic: trap type 6, code=0, pc=ffffffff81e4c208
>> Starting stack trace...
>> panic(ffffffff81f27c9e) at panic+0x12c
>> kerntrap(ffff8000225a3fb0) at kerntrap+0x114
>> alltraps_kern_meltdown() at alltraps_kern_meltdown+0x7b
>> pf_state_export(fffffd8002e8a1c0,fffffd952aa907e0) at pf_state_export+0x38
>> pfsync_sendout() at pfsync_sendout+0x5e4
>> pfsync_update_state(fffffd9526f4f3f0) at pfsync_update_state+0x15b
>> pf_test(2,1,ffff800001eb5800,ffff8000225a4438) at pf_test+0x117a
>> ip_input_if(ffff8000225a4438,ffff8000225a4444,4,0,ffff800001eb5800) at 
>> ip_input_if+0xcd
>> ipv4_input(ffff800001eb5800,fffffd805f671000) at ipv4_input+0x39
>> ether_input(ffff800001eb5800,fffffd805f671000) at ether_input+0x3b1
>> carp_input(ffff800001ecd800,fffffd805f671000,5e000101) at carp_input+0x196
>> ether_input(ffff800001ecd800,fffffd805f671000) at ether_input+0x1d9
>> if_input_process(ffff8000014e3050,ffff8000225a4618) at if_input_process+0x6f
>> ifiq_process(ffff8000014e5700) at ifiq_process+0x69
>> taskq_thread(ffff800000037200) at taskq_thread+0x100
>> end trace frame: 0x0, count: 242
>> End of stack trace.
>>
>> Regards,
>> Tamas
>
> Hi,
>
> can you upgrade to latest snapshot with sysupgrade?
> If that won't solve your panic can you try this diff
> https://www.mail-archive.com/[email protected]/msg72582.html
>
> this was my panic
> https://www.mail-archive.com/[email protected]/msg18583.html
> and that diff solved it ...

Thanks for the suggestion. As I understand (from the fedi, commit messages and 
some emails) there are multiple recent changes to pf, so this strategy might 
not work.

Currently, we disabled the pfsync interface in a hope this would bypass the 
code causing the issue.

We might downgrade to 7.1 because while we were following upgrades every few 
months for the last 3 years we have not seen such an issue.

Regards,
Tamás on behalf of the HU hostmasters

Reply via email to