Today I had a Crash Dump during startup (may be intresting)

[louis2]

Today I had a crash dump at start-up using latest development build. I do not have a clue why. For those interested file attached.

Dump header from device: /dev/gptid/3eb1c198-a592-11ea-b190-6cb3113bbb90
Architecture: amd64
Architecture Version: 4
Dump Length: 84992
Blocksize: 512
Compression: none
Dumptime: Wed Jun 10 11:14:52 2020
Hostname: pfSense.lan
Magic: FreeBSD Text Dump
Version String: FreeBSD 12.1-STABLE f1de4082be8(devel-12) pfSense
Panic String: page fault
Dump Parity: 2509497656
Bounds: 0
Dump Status: good

Dump file attached. System is running normal now, as far as I can see.

Louis

textdump.tar.0

[jimp]

Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address	= 0x130002ff
fault code		= supervisor read data, page not present
instruction pointer	= 0x20:0xffffffff80eb020d
stack pointer	        = 0x0:0xfffffe00004477e0
frame pointer	        = 0x0:0xfffffe00004478b0
code segment		= base 0x0, limit 0xfffff, type 0x1b
			= DPL 0, pres 1, long 1, def32 0, gran 1
processor eflags	= interrupt enabled, resume, IOPL = 0
current process		= 0 (config_0)
trap number		= 12
panic: page fault
cpuid = 0
time = 1591780492
KDB: enter: panic

db:0:kdb.enter.default>  bt
Tracing pid 0 tid 100013 td 0xfffff80004344000
kdb_enter() at kdb_enter+0x37/frame 0xfffffe00004474a0
vpanic() at vpanic+0x197/frame 0xfffffe00004474f0
panic() at panic+0x43/frame 0xfffffe0000447550
trap_fatal() at trap_fatal+0x391/frame 0xfffffe00004475b0
trap_pfault() at trap_pfault+0x4f/frame 0xfffffe0000447600
trap() at trap+0x286/frame 0xfffffe0000447710
calltrap() at calltrap+0x8/frame 0xfffffe0000447710
--- trap 0xc, rip = 0xffffffff80eb020d, rsp = 0xfffffe00004477e0, rbp = 0xfffffe00004478b0 ---
rtsock_msg_mbuf() at rtsock_msg_mbuf+0xed/frame 0xfffffe00004478b0
rt_newmaddrmsg() at rt_newmaddrmsg+0x108/frame 0xfffffe0000447960
if_delmulti_locked() at if_delmulti_locked+0x2a/frame 0xfffffe0000447990
if_delmulti_ifma_flags() at if_delmulti_ifma_flags+0x54/frame 0xfffffe00004479c0
in6m_release_task() at in6m_release_task+0xcd/frame 0xfffffe0000447a20
gtaskqueue_run_locked() at gtaskqueue_run_locked+0x121/frame 0xfffffe0000447a80
gtaskqueue_thread_loop() at gtaskqueue_thread_loop+0xb6/frame 0xfffffe0000447ab0
fork_exit() at fork_exit+0x7e/frame 0xfffffe0000447af0
fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe0000447af0
--- trap 0, rip = 0, rsp = 0, rbp = 0 ---

That backtrace doesn't look familiar to me, but given that it's happening in routing (or perhaps multicast + routing) it could be from https://redmine.pfsense.org/issues/9544 or perhaps https://redmine.pfsense.org/issues/10558. 9544 is more likely.

[louis2]

Jimp,

I now and then have IPV6 related FW-rule-error messages, and also notice just like others "pfSense radvd[84278]: can't join ipv6-allrouters on ix1.116"

The FW-rule-load errors I sometimes see are related to throwing back traffic from the PCLAN towards the WAN-gateway, by explicit defining the WAN-gateway in the rule. The reason to do that is that accessing my services from my own network should pass the same filterrules as if they were comming from the internet

There were error(s) loading the rules: /tmp/rules.debug:530: no routing address with matching address family found. - The line in question reads [530]: pass in quick on $PCLAN $GWWAN_VLAN6_DHCP6 inet6 proto { tcp udp } from <myipv6-range-for-pc-subnet> to $MyPublicServersIPV6 tracker 1559247992 keep state label "USER_RULE: Route this internal traffic via the front door"
@ 2020-06-11 11:04:23

So that could be related as well.

Louis