I'm not sure if this is related, but I had problem with 3.4.31 and
solrais 9 sparc and ipnat as well.
as soon as I created nat config, system started to crash,
[114] [EMAIL PROTECTED]:crash/muhv> mdb unix.0 vmcore.0
Loading modules: [ unix krtld genunix ufs_log ip usba s1394 logindmux
ptm cpc random ipc nca lofs nfs ]
> ::stack
fr_timeoutstate+0x2c(78065c18, 1, a, 7808bd20, 7808ac00, 300000529c7)
ipfr_slowtimer+0x40(0, 2a10006bd40, 20, 1, 2a10006bd40, 0)
callout_execute+0x90(30000247038, 10, bffffffffff89f78, 1, 0, 0)
taskq_thread+0x188(30000251e60, 0, 1437888, 1437888, 30000251e16, 0)
thread_start+4(30000251e08, 0, 0, 0, 0, 0)
0x30001529d5f: IP Filter: attach to [hme0,0] - IPv4
0x30001529adf: IP Filter: attach to [hme1,1] - IPv4
0x3000152985f: IP Filter: attach to [pcpsm0,0] - IPv4
0x300015295df: IP Filter: attach to [hme0,0] - IPv6
0x3000152935f: IP Filter: v3.4.31, attaching complete.
0x300028e2d60:
panic[cpu0]/thread=2a10006bd40:
0x300028e2fe0: BAD TRAP: type=34 rp=2a10006b7b0 addr=30000052a47 mmu_fsr=0
0x300015294a0:
0x30001529720: sched:
0x300028e33a0: alignment error:
0x3000103fb60: addr=0x30000052a47
0x3000103f660: pid=0, pc=0x7807818c, sp=0x2a10006b051,
tstate=0x80001606, context=0x0
0x3000103f160: g1-g7: 144e000, 97c0, 8, 1, 60, 0, 2a10006bd40
0x3000103ec60:
0x3000103e763: 000002a10006b4e0 unix:die+a4 (34, 2a10006b7b0,
30000052a47, 0, 100b96c, 0)
0x30000e39763: %l0-3: 0000000000000000 00000300000587e8
000003000000eb00 000003000000eab0
%l4-7: 0000000000000034 000000000142d638 0000030000039ac0 0000030000e302e0
0x30000e39c63: 000002a10006b5c0 unix:trap+59c (2a10006b7b0, 0, 10000,
10200, 300, 1)
0x30000bba1e3: %l0-3: 0000030000014248 000000000080000b
0000000001437888 0000000000000034
%l4-7: 0000000000000000 0000000078065de8 0000000000000000 0000000000000000
0x30000bba6e3: 000002a10006b700 unix:ktl0+48 (30001a41988, ffff, 22, 1,
2a10006bd44, 30000bcdc28)
0x30000bbabe3: %l0-3: 0000000000000007 0000000000001400
0000000080001606 000000000102aaf8
%l4-7: 0000000001456c00 0000030000274ee0 0000000000000000 000002a10006b7b0
0x30000bbb0e3: 000002a10006b850 ipf:fr_timeoutstate+9c (78065c18, 1, a,
7808bd20, 7808ac00, 300000529c7)
0x30000bbb363: %l0-3: 0000000000000087 0000030000052a47
0000000000000000 000000000000d802
%l4-7: 0000000001400090 000000000142d5b8 0000000000000000 000000007808ea08
0x30000bbb5e3: 000002a10006b910 ipf:ipfr_slowtimer+40 (0, 2a10006bd40,
20, 1, 2a10006bd40, 0)
0x30000bbb863: %l0-3: 0000000000000001 0000000000000003
0000000000000000 0000000078065c18
%l4-7: 0000000000000004 00000000001f0000 0000000000000002 000003000026f3b0
0x30000bbbc23: 000002a10006b9d0 genunix:callout_execute+90
(30000247038, 10, bffffffffff89f78, 1, 0, 0)
0x30000bbbea3: %l0-3: 0000000078079f68 0000030001a1cd78
000000000000d802 000000000000d802
%l4-7: 0000030000247000 8000000000000000 0000000000000000 0000030000248038
0x30000b80263: 000002a10006ba80 genunix:taskq_thread+188 (30000251e60,
0, 1437888, 1437888, 30000251e16, 0)
0x30000b804e3: %l0-3: 0000000001079164 0000030000251e28
000003000024ff88 0000000000010000
%l4-7: 0000030000251e08 0000030000251e3a 0000030000251e38 0000030000251e30
0x30000b80760:
0x30000b809e3: syncing file systems...
Tang Li Yong wrote:
> It seems that there exists a very boring bug in IPNAT. Does anyone know how
> to solve it?
>
> Here's the description of this problem:
> We must use ipnat to map our internal hosts to an address pool. First I
> established a testing environment as below:
> |
> fxp0|192.168.101.10
> +--+--+
> | GW |
> +--+--+
> fxp1|10.0.0.1
> |
> +----------+---------- +
> | | |
> A (.11) B (.12) C (.13)
>
> GW is a FreeBSD 4.7 box with ipfilter 3.4.29. A, B and C are 3 internal
> hosts. The ipnat script in GW is:
> map fxp0 from any to any -> 192.168.101.10-192.168.101.11 portmap
> TCP/UDP auto
> map fxp0 from any to any -> 192.168.101.10-192.168.101.11
> I tried to "ping 192.168.101.1" on A, everything is OK. And then I tried to
> "ping 192.168.101.1" on B without let A stop, everything is OK also. But
> when C took part in this game, the GW crashed immediately. :-(
> If I increased the number of the addresses in the pool, I can let more hosts
> ping the same host simultaneously. But when the number of hosts goes beyond
> the number of addresses, the same problem happen again.
>
> I will be very glad if anybody can tell me how to do. Thanks ahead.
>
> With best regards,
>
> --Leroy Tang
>
>
>
--
toomas
--
I have already given two cousins to the war and I stand ready to
sacrifice my wife's brother.
-- Artemus Ward
