Help with core debugging

Hi all, hoping you can help me and I'm fairly new to Solaris. We have a production email server running Solaris 10. The server itself is a Sun Netra T1 200. The other day, it rebooted on it's own. After checking the syslog message I found out that the reboot was caused by a kernel panic:

reboot after panic: BAD TRAP: type=31 rp=2a1000471e0 addr=0 mmu_fsr=0 occurred in module "unix" due to a NULL pointer dereference

I ran mdb on the core files, and was just wondering I could have some help deciphering them. It appears that ipfilters is what caused the NULL pointer deference which caused the kernel panic, and was wondering if it could be a DoS attack of some sort:

Loading modules: [ unix krtld genunix specfs ufs sd md ip sctp usba fctl nfs random ipc crypto ptm ]

> ::status

debugging crash dump vmcore.0 (64-bit) from mail

operating system: 5.10 Generic (sun4u)

panic message:

BAD TRAP: type=31 rp=2a1000471e0 addr=0 mmu_fsr=0 occurred in module "unix" due

to a NULL pointer dereference

dump content: kernel pages only

> $c

mutex_enter+4(30016a82618, 0, 70140e98, 30016a826ec, 30016a82540, 30016a82540)

fr_updatestate+0x50(2a100047560, 30016a82540, 70140e98, 0, 2, 0)

fr_checkstate+0x238(2a100047560, 2a100047554, 0, 2a100047568, 30004334f40,

40004502)

fr_check+0x358(1, 14, 0, 30007455000, 7013c0d0, 2a1000478a8)

pfil_precheck+0x4f4(300024275e8, 2a1000478a8, 0, 1, 30001b3ace0, 0)

pfilmodrput+0x23c(30001b3ace0, 3000196fed8, 2a100042000, 8400, 0, 300056b8080)

putnext+0x218(30001b3aed0, 30001b3ace0, 30007455000, 100, 30001b3b490, 0)

eri_intr+0x43c(30001c0e000, 10280, 80000000, 10238, 1c000, 20)

pci_intr_wrapper+0x9c(300000b63f8, 300000b9048, 0, 0, 0, 30001bb1b00)

intr_thread+0x144(701408c0, 0, 120, 300049c79c0, 30003c1f4c0, 70b)

fr_addstate+0x900(2a1016c0bf0, 200000, 40004502, 30016a82540, 0, 0)

fr_firewall+0x380(2a1016c0bf0, 2a1016c0be4, 8, 2, 0, 1)

fr_check+0x384(1, 14, 0, 30001f18980, 7013c0d0, 2a1016c0f38)

pfil_precheck+0x4f4(300018cf308, 2a1016c0f38, 0, 2, 30001b3add8, 0)

pfilmodwput+0xa8(30001b3add8, 30001b9d600, 3000196fed8, 8000d0a, 0, 3001ec72400)

putnext+0x218(30001b3aed0, 30001b3add8, 30001f18980, 100, 30001b3ab48, 0)

ip_wput_ire+0x1514(3000497ea48, 10000, 0, 0, 30001b9d7c8, 30001b3ab48)

ip_output+0x67c(c7acc002, 0, 0, 0, 0, 3000404b340)

putnext+0x218(30004f2f778, 30004f2f680, 30001f18980, 100, 30001b3b2f8, 0)

udp_wput+0x61c(30001b3b2f8, 1d74ee60, 30001f18980, 3001ec72490, ff, 14)

putnext+0x218(30001b3b3f0, 30001b3b2f8, 30001e476c0, 100, 300079d9390, 0)

strput+0x1d8(30004b0c4c8, 30001e476c0, 25c7, 30001e476c0, 30003d458e0, 0)

kstrputmsg+0x328(2a1016c1ab0, 3001c033380, 0, 0, 3, 4)

sosend_dgram+0x248(3001c033380, 3000e2d3480, 10, 2a1016c1ab0, 18a2000,

30005409b30)

sotpi_sendmsg+0x44c(30005409b30, 2a1016c1a80, 2a1016c1ab0, 0, 0, 0)

sendit+0x138(5, 2a1016c1a80, 2a1016c1ab0, 3000e2d3480, 30005409b30, fcbae100)

sendto+0x64(5, fd26fa1c, 22, 0, fd26f910, 10)

syscall_trap32+0xcc(5, fd26fa1c, 22, 0, fd26f910, 10)

Any help with this is much appreciated.

--

Chris

[3319 byte] By [cpedro] at [2008-2-23]
# 1

It looks like you're running into

6283314 frequent panics in ipf:fr_movequeue: NULL pointer dereference

which deals with race conditions due to improper locking.

This bug has been fixed in S10 01/06 and in the latest

kernel patch 118822-30

(<a href="http://sunsolve.sun.com/search/document.do?assetkey=urn:cds:docid:1-21-11 8822-30-1" target="_blank"> http://sunsolve.sun.com/search/document.do?assetkey=urn:cds: docid:1-21-118822-30-1</a>)

Regards,

Joep

Joep at 2007-7-5 > top of java,General,Sun Alert and Security Discussion...