Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 17 Jun 1999 15:11:45 -0500 (EST)
From:      Alfred Perlstein <bright@rush.net>
To:        "Ronald F. Guilmette" <rfg@monkeys.com>
Cc:        questions@FreeBSD.ORG
Subject:   Re: What happens in this case?
Message-ID:  <Pine.BSF.3.96.990617150927.14320p-100000@cygnus.rush.net>
In-Reply-To: <24122.929649609@monkeys.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 17 Jun 1999, Ronald F. Guilmette wrote:

> 
> What will actually happen if I have a 140MB swap partition setup as
> my dump device, but I have 320MB of physical memory, and then a panic
> occurs?
> 
> Will the kernel see that it only has 140MB of disk to work with and
> then just write that much and then give up?  Or will it refuse to
> write anything?  Or (worse) will it just keep on writing stuff even
> past the end of the partition in question (thus overwriting the data
> in OTHER partitions)?

heh, no, the device should abort the dump routine.  you will get
an incomplete dump.

> And anyway, why the heck does the dumpdev have to big as big as
> physical memory?  I mean hay!  What if I have 320MB physical, but only
> 100MB of that is actually allocated or in use at the moment of the
> panic?  Then I should only need a 100MB swap partition to hold the
> panic dump, right?

the dump routines are "dumb" the reason they are dumb is so that they 
will work even when severe corruption of internal data structures occur.

this dumbness makes it difficult for them to only dump 
"the memory in use"

btw, freebsd documentation recommends 1-2x swap as ram.

-Alfred




To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.990617150927.14320p-100000>