misc/183417: panic: fault on nofault entry in softdep_setup_freeblocks

Rick Reed rr at whatsapp.com
Mon Oct 28 21:20:02 UTC 2013

>Number:         183417
>Category:       misc
>Synopsis:       panic: fault on nofault entry in softdep_setup_freeblocks
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-bugs
>State:          open
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Mon Oct 28 21:20:01 UTC 2013
>Originator:     Rick Reed
>Release:        9.1
FreeBSD mms983.whatsapp.net 9.1-RELEASE-p7 FreeBSD 9.1-RELEASE-p7 #0 r255864M: Tue Sep 24 18:12:17 PDT 2013     root at mms983.whatsapp.net:/usr/obj/usr/src/sys/bsdnine64  amd64
We've seen this panic multiple times on hosts which are getting uncorrectable I/O errors from a spinning SATA disk:

root at mms983$ cat panic.txt
vm_fault: fault on nofault entry, addr: ffffffa010748000

b:0:kdb.enter.panic>  show pcpu
cpuid        = 12
dynamic pcpu = 0xffffff807faf8200
curthread    = 0xfffffe09675938e0: pid 57943 "beam.smp"
curpcb       = 0xffffffa31038ebc0
fpcurthread  = none
idlethread   = 0xfffffe0034935470: tid 100014 "idle: cpu12"
curpmap      = 0xfffffe01230ccbe8
tssp         = 0xffffffff80957340
commontssp   = 0xffffffff80957340
rsp0         = 0xffffffa31038ebc0
gs32p        = 0xffffffff80955498
ldt          = 0xffffffff809554d8
tss          = 0xffffffff809554c8
db:0:kdb.enter.panic>  bt
Tracing pid 57943 tid 43087119 td 0xfffffe09675938e0
kdb_enter() at kdb_enter+0x3b
panic() at panic+0x1c7
vm_fault_hold() at vm_fault_hold+0x1eea
vm_fault() at vm_fault+0x73
trap_pfault() at trap_pfault+0x41d
trap() at trap+0x365
calltrap() at calltrap+0x8
--- trap 0xc, rip = 0xffffffff8054c43c, rsp = 0xffffffa31038e490, rbp = 0xffffffa31038e520 ---
softdep_setup_freeblocks() at softdep_setup_freeblocks+0x3ec
ffs_truncate() at ffs_truncate+0x530
ufs_inactive() at ufs_inactive+0x1e5
vinactive() at vinactive+0x8d
vputx() at vputx+0x2d8
kern_unlinkat() at kern_unlinkat+0x1df
amd64_syscall() at amd64_syscall+0x5d6
Xfast_syscall() at Xfast_syscall+0xf7
--- syscall (10, FreeBSD ELF64, sys_unlink), rip = 0x80125e32c, rsp = 0x7ffffd1b3ea8, rbp = 0x825b5af40 ---

After reboot, we get a second panic:
root at mms983$ cat panic.txt
ufs_dirbad: /d6: bad dir ino 9708991 at offset 1536: mangled entry

After reboot and fsck, filesystem is working again.




More information about the freebsd-bugs mailing list