Panic with RACCT on a jailed process

Jeremie Le Hen jlh at
Tue Jun 18 22:08:31 UTC 2013


I've been bit by a panic three times over the last month.

I'm currently running:
    FreeBSD obiwan.piupiu.local 10.0-CURRENT FreeBSD 10.0-CURRENT #7 r251519: Sun Jun  9 22:37:09 CEST 2013 root at obiwan.piupiu.local:/usr/obj/usr/src/sys/OBIWAN  amd64

The panic message is:
    panic: destroying non-empty racct: 1007616 allocated for resource 4

I quicky tried to track this down, but in half an hour I didn't get very
far.  Resource 4 is RACCT_RSS.  The faulty process seems to be 66555
which looks like a 32-bits cron(8) running in a jail.

This is the call trace:

    kdb_enter() at kdb_enter+0x3e/frame 0xffffff80e63b3610
    vpanic() at vpanic+0x146/frame 0xffffff80e63b3650
    kassert_panic() at kassert_panic+0x136/frame 0xffffff80e63b36c0
    racct_destroy() at racct_destroy+0x8e/frame 0xffffff80e63b36f0
    uifree() at uifree+0x5e/frame 0xffffff80e63b3710
    crfree() at crfree+0x48/frame 0xffffff80e63b3730
    thread_wait() at thread_wait+0xac/frame 0xffffff80e63b3750
    proc_reap() at proc_reap+0x43f/frame 0xffffff80e63b37a0
    proc_to_reap() at proc_to_reap+0x340/frame 0xffffff80e63b37f0
    kern_wait6() at kern_wait6+0x1fa/frame 0xffffff80e63b3890
    kern_wait() at kern_wait+0x5a/frame 0xffffff80e63b39e0
    freebsd32_wait4() at freebsd32_wait4+0x2e/frame 0xffffff80e63b3ae0
    ia32_syscall() at ia32_syscall+0x26c/frame 0xffffff80e63b3bf0

The panic happened each three times because of cron(8), but both with
32-bits and 64-bits versions.

crashinfo + ddb textdump available here:

Note that the dmesg included in the core contains another panic I got
with ZFS.  I usually wait to see it twice before reporting, but if you
want I still have the core and the crashinfo around.

Jeremie Le Hen

Scientists say the world is made up of Protons, Neutrons and Electrons.
They forgot to mention Morons.

More information about the freebsd-current mailing list