Fatal trap 12: page fault while in kernel mode cpuid = 2; apic id = 02 fault virtual address = 0x58aa5c6 fault code = supervisor read data, page not present instruction pointer = 0x20:0xffffffff80d57a78 stack pointer = 0x28:0xfffffe00002e72c0 frame pointer = 0x28:0xfffffe00002e72c0 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 (hvevent2) trap number = 12 panic: page fault cpuid = 2 KDB: enter: panic db:0:kdb.enter.default> show pcpu cpuid = 2 dynamic pcpu = 0xfffffe026cb25580 curthread = 0xfffff80005dd2000: pid 0 "hvevent2" curpcb = 0xfffffe00002e7cc0 fpcurthread = none idlethread = 0xfffff8000525a000: tid 100005 "idle: cpu2" curpmap = 0xffffffff834f1c40 tssp = 0xffffffff835a33a0 commontssp = 0xffffffff835a33a0 rsp0 = 0xfffffe00002e7cc0 gs32p = 0xffffffff835a9ff8 ldt = 0xffffffff835aa038 tss = 0xffffffff835aa028 tlb gen = 5304139 db:0:kdb.enter.default> bt Tracing pid 0 tid 100120 td 0xfffff80005dd2000 kdb_enter() at kdb_enter+0x3b/frame 0xfffffe00002e6f70 vpanic() at vpanic+0x19b/frame 0xfffffe00002e6fd0 panic() at panic+0x43/frame 0xfffffe00002e7030 trap_pfault() at trap_pfault/frame 0xfffffe00002e7080 trap_pfault() at trap_pfault+0x49/frame 0xfffffe00002e70e0 trap() at trap+0x29d/frame 0xfffffe00002e71f0 calltrap() at calltrap+0x8/frame 0xfffffe00002e71f0 --- trap 0xc, rip = 0xffffffff80d57a78, rsp = 0xfffffe00002e72c0, rbp = 0xfffffe00002e72c0 --- m_tag_locate() at m_tag_locate+0x28/frame 0xfffffe00002e72c0 bpf_mtap() at bpf_mtap+0x19d/frame 0xfffffe00002e7330 vlan_start() at vlan_start+0x1b6/frame 0xfffffe00002e73b0 if_transmit() at if_transmit+0x16e/frame 0xfffffe00002e73f0 ether_output_frame() at ether_output_frame+0x98/frame 0xfffffe00002e7420 ether_output() at ether_output+0x6d7/frame 0xfffffe00002e74b0 ip_output() at ip_output+0x138d/frame 0xfffffe00002e75e0 ip_forward() at ip_forward+0x2c3/frame 0xfffffe00002e7680 ip_input() at ip_input+0x724/frame 0xfffffe00002e7710 netisr_dispatch_src() at netisr_dispatch_src+0xa2/frame 0xfffffe00002e7760 ether_demux() at ether_demux+0x15b/frame 0xfffffe00002e7790 ether_nh_input() at ether_nh_input+0x32c/frame 0xfffffe00002e77f0 netisr_dispatch_src() at netisr_dispatch_src+0xa2/frame 0xfffffe00002e7840 ether_input() at ether_input+0x26/frame 0xfffffe00002e7860 vlan_input() at vlan_input+0x215/frame 0xfffffe00002e7910 ether_demux() at ether_demux+0x144/frame 0xfffffe00002e7940 ether_nh_input() at ether_nh_input+0x32c/frame 0xfffffe00002e79a0 netisr_dispatch_src() at netisr_dispatch_src+0xa2/frame 0xfffffe00002e79f0 ether_input() at ether_input+0x26/frame 0xfffffe00002e7a10 hn_chan_callback() at hn_chan_callback+0xf29/frame 0xfffffe00002e7af0 vmbus_chan_task() at vmbus_chan_task+0x22/frame 0xfffffe00002e7b20 taskqueue_run_locked() at taskqueue_run_locked+0x185/frame 0xfffffe00002e7b80 taskqueue_thread_loop() at taskqueue_thread_loop+0xb8/frame 0xfffffe00002e7bb0 fork_exit() at fork_exit+0x83/frame 0xfffffe00002e7bf0 fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe00002e7bf0 --- trap 0, rip = 0, rsp = 0, rbp = 0 --- That trace doesn't look familiar to me, but it seems to have crashed when trying to find an mbuf tag, so maybe it's an issue with mbuf exhaustion. (Though usually that would crash when trying to allocate one...) What features do you have enabled? (Things like captive portal, limiters, CARP/HA, etc) What does netstat -m show?

@bmeeks Does tracert (windows) use ports? Anyway, I did two test with pfSense and two with Windows to an AWS IP... I can't see nothing. Spoiler [image: 1592837803123-capture.png]

@Gertjan thanks

@Gertjan So an update, after a week of issues I've narrowed this down to pfBlockerNG-devel/DNS issues. I've removed pfBlockerNG-devel and reverted to AdGuard Home and I've had a stable WAN connection and box for 3 days. I did notice this and wasn't sure if that was a contributing factor: https://www.reddit.com/r/PFSENSE/comments/88wg6g/issue_with_pfblockerng_dnsbl_and_cloudflares_1111/ @Rico - I made these changes and it made no difference when DNS was down or been blocked by pfBlockerNG-devel So thanks all for the help and pointers, all really useful and I defiantly improved my knowledge of PFSense and DNS over the last week!

@awebster i've seen you help https://forum.netgate.com/topic/145578/ldaps-ad-bind/19. Can you please point me in direction where i need to troubleshoot this issue?

@Derelict We actually have 2 phone systems from a merger. A on premise phone system (that works great) and AT&T hosted voice. Calls on the AT&T drop randomly. AT&T refuses to be of any help because it's a network problem and to be fair if i put the phones on a router with the udp timeout changed it does work. The reason I am trying to move it off the current router is QOS. Which has been a problem before because we were running 3 routers for different network segments off the same internet connections and 1 router could use all the bandwidth and starve the other routers.

Ok, just found out pfSense is indeed truncating: https://forum.netgate.com/topic/152220/suricata-eve-json-cutting-off-in-remote-logging/9

@cajunzman Okay.

@elduder You might want this: https://github.com/homebridge/homebridge May find valuable info here too: https://forum.netgate.com/topic/154163/can-i-use-ips-to-trigger-a-custom-action

You can not put the wan and lan on the same network.. Sure looks like your going to connect the wan and lan to the same network.

UFS keeps extra space that is not available to regular users so when the filesystem is completely filled for user and non-user data, it shows 108%. : tunefs -p / tunefs: POSIX.1e ACLs: (-a) disabled tunefs: NFSv4 ACLs: (-N) disabled tunefs: MAC multilabel: (-l) disabled tunefs: soft updates: (-n) enabled tunefs: soft update journaling: (-j) enabled tunefs: gjournal: (-J) disabled tunefs: trim: (-t) disabled tunefs: maximum blocks per file in a cylinder group: (-e) 4096 tunefs: average file size: (-f) 16384 tunefs: average number of files in a directory: (-s) 64 tunefs: minimum percentage of free space: (-m) 8% tunefs: space to hold for metadata blocks: (-k) 6408 tunefs: optimization preference: (-o) time tunefs: volume label: (-L) Note the value for -m

Using something like IPMI? With that I can attach an ISO image on my PC and boot the data center server off of it. Takes a while to boot of course as it has to upload but it can be done. pfSense isn't involved in that scenario.

Update: Looks like the NIC is buggered. Threw PFsense onto another PC and a different second NIC and getting full 1Gb transfers as expected. Put the quad-card into the PC and it refused to boot with it in. RMA time!!

@Gertjan Very educating... thank you!

ck42, it looks like it is single-queued

Update/reboot when there's a remote vuln that can affect my box. All ports blocked by default, so pretty much has to be a FreeBSD kernel issue. Had 400+ days at some point the past year. Been hanging out on 2.4.4-p3 for a bit now.