Please check crash report for Limiters
-
Backtrace:
db:1:pfs> bt Tracing pid 0 tid 100011 td 0xfffff80001973740 kdb_enter() at kdb_enter+0x33/frame 0xfffffe00c59f7150 panic() at panic+0x43/frame 0xfffffe00c59f71b0 trap_fatal() at trap_fatal+0x40f/frame 0xfffffe00c59f7210 trap_pfault() at trap_pfault+0x4f/frame 0xfffffe00c59f7270 calltrap() at calltrap+0x8/frame 0xfffffe00c59f7270 --- trap 0xc, rip = 0xffffffff84446699, rsp = 0xfffffe00c59f7340, rbp = 0xfffffe00c59f7360 --- si_new() at si_new+0x109/frame 0xfffffe00c59f7360 ipdn_si_find() at ipdn_si_find+0x2c/frame 0xfffffe00c59f73b0 dummynet_io() at dummynet_io+0x18c/frame 0xfffffe00c59f7400 pf_dummynet_route() at pf_dummynet_route+0x3b9/frame 0xfffffe00c59f74f0 pf_route() at pf_route+0x96d/frame 0xfffffe00c59f75d0 pf_test() at pf_test+0xe28/frame 0xfffffe00c59f7780 pf_route() at pf_route+0x7ce/frame 0xfffffe00c59f7860 pf_test() at pf_test+0xe28/frame 0xfffffe00c59f7a10 pf_check_in() at pf_check_in+0x27/frame 0xfffffe00c59f7a30 pfil_mbuf_in() at pfil_mbuf_in+0x38/frame 0xfffffe00c59f7a60 ip_tryforward() at ip_tryforward+0x16d/frame 0xfffffe00c59f7b10 ip_input() at ip_input+0x349/frame 0xfffffe00c59f7b70 netisr_dispatch_src() at netisr_dispatch_src+0x22c/frame 0xfffffe00c59f7bc0 ether_demux() at ether_demux+0x149/frame 0xfffffe00c59f7bf0 ether_nh_input() at ether_nh_input+0x36d/frame 0xfffffe00c59f7c50 netisr_dispatch_src() at netisr_dispatch_src+0xaf/frame 0xfffffe00c59f7ca0 ether_input() at ether_input+0x69/frame 0xfffffe00c59f7d00 iflib_rxeof() at iflib_rxeof+0xc46/frame 0xfffffe00c59f7e00 _task_fn_rx() at _task_fn_rx+0x72/frame 0xfffffe00c59f7e40 gtaskqueue_run_locked() at gtaskqueue_run_locked+0x14e/frame 0xfffffe00c59f7ec0 gtaskqueue_thread_loop() at gtaskqueue_thread_loop+0xc2/frame 0xfffffe00c59f7ef0 fork_exit() at fork_exit+0x7f/frame 0xfffffe00c59f7f30 fork_trampoline() at fork_trampoline+0xe/frame 0xfffffe00c59f7f30 --- trap 0x4dad4dad, rip = 0x87d287d22e5f2e5f, rsp = 0x45614561ea95ba9d, rbp = 0x2d0e3d0e4a444244 ---
Panic:
config_aqm Unable to configure flowset, flowset busy! config_aqm Unable to configure flowset, flowset busy! config_aqm Unable to configure flowset, flowset busy! config_aqm Unable to configure flowset, flowset busy! Fatal trap 12: page fault while in kernel mode cpuid = 0; apic id = 00 fault virtual address = 0xffffffff00000096 fault code = supervisor read data, page not present instruction pointer = 0x20:0xffffffff84446699 stack pointer = 0x0:0xfffffe00c59f7340 frame pointer = 0x0:0xfffffe00c59f7360 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 (if_io_tqg_0) rdi: 3e3499dbe4058ac9 rsi: 3e3499db67f5a41a rdx: 0000000000017fa0 rcx: fffffe010603c000 r8: 0000000000000000 r9: 8000000000000000 rax: ffffffff0000003e rbx: fffff8002f7dc000 rbp: fffffe00c59f7360 r10: 0000000000000040 r11: 0000000000000002 r12: fffff8020bced900 r13: fffffe00c59f7488 r14: fffff802e6c56700 r15: 0000000000000000 trap number = 12 panic: page fault cpuid = 0 time = 1718626492 KDB: enter: panic
What exactly did you do to trigger that?
Steve
-
@stephenw10 when removing queue from Limiters
Also this bug is annoying
there were error(s) loading the rules: pfctl: opvnc1: driver does not support altq.
I have opvnc2, opvnc3 no any warning and this bug for opvnc1 -
You have added a ALTQ queue on that interface?
What the exact steps required to replicate this Limiters issue?
-
@stephenw10 No, after this crash I removed Limiters and going over traffic shaping wizard. Did set OpenVPN clients as separate interfaces WAN and got this altq bug now.
-
@stephenw10 No, after this crash I removed Limiters and going over traffic shaping wizard. Did set OpenVPN clients as separate interfaces WAN and got this altq bug now.
@stephenw10 said in Please check crash report for Limiters:
What the exact steps required to replicate this Limiters issue?
I told, with Limiters active (Floating rule as well active on WAN) I removed 2 queues from settings. Before get set 3queues on each Limiters and during removing 2 of them was crashed!
-
Were you routing traffic through those queues or through the parent limiter? I can't replicate that here.
-
@stephenw10 said in Please check crash report for Limiters:
through those queues
-
@stephenw10 said in Please check crash report for Limiters:
through the parent limiter
Sorry misclick
-
So rules were set to use the parent limiter not any of the queues you removed directly?
-
@stephenw10 If you mean floating rule, I did ll the same as described in pfsense docs with a setting against bufferbloat, only add more 2 queues during set up on the same parent Limiters and removed later 2 of them and remain one queue only
-
Ah, OK so now we're getting somewhere. You had the Limiters setup for codelq.
-
@stephenw10 Yes, during set up
Queue Management Algorithm
: CodelScheduler
: FQ CodelECN checked on parent and queues as well
-
Ok, let me see if I can replicate that.