panic: Journal overflow

[Adam Vande More]

On Wed, Sep 23, 2009 at 6:35 PM, Doug Poland <doug at polands.org> wrote:

> Hello,
>
> I've got a 7.2-RELEASE (i386) box that has recently experienced
> several panics when writing to a gjournaled partition.  On this
> particular partition, the journal device is "mixed" in with the data
> provider.  In both cases, I was trying to copy a 19GB file from an
> external USB device (da1p16.journal) to the filesystem on a 3Ware
> RAID-10 (da7p0.journal).
>
> This afternoon I saw...
>
> panic: Journal overflow (joffset=xxxxxxx active=yyyyyyyy
> inactive=zzzzzz)
> cpuid = 1
> Uptime:
> Physical memory: 1011 MB
> Dumping 351MB
>
> Fatal trap 12: page fault while in kernel mode
> cpuid = 0, apic id = 00
> fault virtual address    = 0xea60
> fault code               = supervisor write, page not present
> instructor pointer       = 0x20: 0xc04715b6
> stack pointer            = 0x28: 0xe6ab0320
> frame pointer            = 0x28: 0xe6ab03bc
> code segment             = base 0x0, limit 0xfffff, type 0x1b
>                           DPL 0, pres 1, def32 1, gran 1
> processor eflags         = interrupt enabled, resume, 10PL = 0
> current process          = 48 (gjournal da0p7)
> trap number              = 12
>
> Does this mean my journal is not big enough for the amount of data I'm
> trying to push through?


No, it's kinda like the journal is a is a buffer and you overflowed it.  You
can address this by two methods:

1)  slow the rate of transfer
2)  make a bigger journal



> Any suggestions on how to resolve this?
>

http://www.freebsd.org/doc/en_US.ISO8859-1/articles/gjournal-desktop/understanding-journaling.html

>
> I can provide "a lot" more data if necessary.  Thanks for your help!
>
> --
> Regards,
> Doug
>


-- 
Adam Vande More