RELENG_9 crash

[Mike Tancsa]

I upgraded to r256844, and still getting crashes under load. (about 500
connections).  RELENG_8 sans ipv6 is nice and stable under similar and
higher loads.  I will try without inet6 for now

Unread portion of the kernel message buffer:
current process         = 1158 (mpd5)
trap number             = 12
panic: page fault
cpuid = 2
KDB: stack backtrace:
#0 0xffffffff8092e5a6 at kdb_backtrace+0x66
#1 0xffffffff808f6e9e at panic+0x1ce
#2 0xffffffff80ca2cc0 at trap_fatal+0x290
#3 0xffffffff80ca2ff8 at trap_pfault+0x1e8
#4 0xffffffff80ca35fe at trap+0x3be
#5 0xffffffff80c8cc0f at calltrap+0x8
#6 0xffffffff80a75470 at sctp_add_addr_to_vrf+0x60
#7 0xffffffff80a3db63 at sctp_addr_change+0x113
#8 0xffffffff809c51cd at rt_newaddrmsg_fib+0x4d
#9 0xffffffff809c3ce7 at rtinit+0x1b7
#10 0xffffffff80a222d8 at in_ifinit+0x3a8
#11 0xffffffff80a239c2 at in_control+0x1072
#12 0xffffffff809b1f83 at ifioctl+0x803
#13 0xffffffff809404a6 at kern_ioctl+0x106
#14 0xffffffff809406ed at sys_ioctl+0xfd
#15 0xffffffff80ca25a0 at amd64_syscall+0x540
#16 0xffffffff80c8cef7 at Xfast_syscall+0xf7
Uptime: 8d9h15m51s
Dumping 1022 out of 7895
MB:..2%..11%..21%..32%..41%..51%..62%..71%..82%..91%

#0  doadump (textdump=<value optimized out>) at pcpu.h:234
234     pcpu.h: No such file or directory.
        in pcpu.h
(kgdb) #0  doadump (textdump=<value optimized out>) at pcpu.h:234
#1  0xffffffff808f6976 in kern_reboot (howto=260)
    at /usr/src/sys/kern/kern_shutdown.c:449
#2  0xffffffff808f6e77 in panic (fmt=0x1 <Address 0x1 out of bounds>)
    at /usr/src/sys/kern/kern_shutdown.c:637
#3  0xffffffff80ca2cc0 in trap_fatal (frame=0xc, eva=<value optimized out>)
    at /usr/src/sys/amd64/amd64/trap.c:879
#4  0xffffffff80ca2ff8 in trap_pfault (frame=0xffffff823bcaa320, usermode=0)
    at /usr/src/sys/amd64/amd64/trap.c:795
#5  0xffffffff80ca35fe in trap (frame=0xffffff823bcaa320)
    at /usr/src/sys/amd64/amd64/trap.c:463
#6  0xffffffff80c8cc0f in calltrap ()
    at /usr/src/sys/amd64/amd64/exception.S:232
#7  0xffffffff80a6a300 in sctp_find_ifn (ifn=0xfffffe0009406800,
    ifn_index=359) at /usr/src/sys/netinet/sctp_pcb.c:205
#8  0xffffffff80a75470 in sctp_add_addr_to_vrf (vrf_id=0,
    ifn=0xfffffe0009406800, ifn_index=359, ifn_type=53,
    if_name=0xfffffe0009406828 "ng347", ifa=0xfffffe0200cb1a00,
    addr=0xfffffe0200cb1b38, ifa_flags=0, dynamic_add=1)
    at /usr/src/sys/netinet/sctp_pcb.c:508
#9  0xffffffff80a3db63 in sctp_addr_change (ifa=0xfffffe0200cb1a00,
    cmd=<value optimized out>) at /usr/src/sys/netinet/sctp_bsd_addr.c:346
#10 0xffffffff809c51cd in rt_newaddrmsg_fib (cmd=1, ifa=0xfffffe0200cb1a00,
    error=0, rt=0xfffffe019c395000, fibnum=0)
    at /usr/src/sys/net/rtsock.c:1299
#11 0xffffffff809c3ce7 in rtinit (ifa=0xfffffe0200cb1a00, cmd=1, flags=5)
    at /usr/src/sys/net/route.c:1626
#12 0xffffffff80a222d8 in in_ifinit (ifp=0xfffffe0009406800,
    ia=0xfffffe0200cb1a00, sin=<value optimized out>, scrub=0)
    at /usr/src/sys/netinet/in.c:1068
#13 0xffffffff80a239c2 in in_control (so=<value optimized out>, cmd=0,
    data=0xfffffe01724ff340 "ng347", ifp=0xfffffe0009406800,
    td=<value optimized out>) at /usr/src/sys/netinet/in.c:550
#14 0xffffffff809b1f83 in ifioctl (so=0xfffffe01091c77f8, cmd=2151704858,
    data=0xfffffe01724ff340 "ng347", td=0xfffffe00096b3490)
    at /usr/src/sys/net/if.c:2569
#15 0xffffffff809404a6 in kern_ioctl (td=0xfffffe00096b3490, fd=149,
    com=2151704858, data=0xfffffe01724ff340 "ng347") at file.h:311
#16 0xffffffff809406ed in sys_ioctl (td=0xfffffe00096b3490,
    uap=0xffffff823bcaaa70) at /usr/src/sys/kern/sys_generic.c:692
#17 0xffffffff80ca25a0 in amd64_syscall (td=0xfffffe00096b3490, traced=0)
    at subr_syscall.c:135
#18 0xffffffff80c8cef7 in Xfast_syscall ()
    at /usr/src/sys/amd64/amd64/exception.S:391
#19 0x000000080227bbac in ?? ()
Previous frame inner to this frame (corrupt stack?)
(kgdb)


On 10/21/2013 9:38 AM, Mike Tancsa wrote:
> This was with a kernel from Aug 6th. (r253988). The box functions as an
> LNS. Similarly loaded RELENG_8 boxes with ipv6 disabled are quite
> stable.  Not sure if thats where the issue is as this is RELENG9
> 
> 	---Mike
> 
> 
> 
> 
> 
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at freebsd.org"


-- 
-------------------
Mike Tancsa, tel +1 519 651 3400
Sentex Communications, mike at sentex.net
Providing Internet services since 1994 www.sentex.net
Cambridge, Ontario Canada   http://www.tancsa.com/