Date: Fri, 1 Jan 1999 15:35:18 -0800 (PST) From: Matthew Dillon <dillon@apollo.backplane.com> To: Garrett Wollman <wollman@khavrinen.lcs.mit.edu>, current@FreeBSD.ORG Subject: Re: object inconsistant state: RPC: 1, RC: 0 Message-ID: <199901012335.PAA18223@apollo.backplane.com> References: <199901012329.PAA18171@apollo.backplane.com>
next in thread | previous in thread | raw e-mail | index | archive | help
::yesterday's latest, and I'm suddenly getting a lot of these messages:
::
::Jan 1 15:50:40 xyz /kernel: object inconsistant state: RPC: 3, RC: 0
::Jan 1 15:50:40 xyz /kernel: object inconsistant state: RPC: 14, RC: 0
::Jan 1 15:50:40 xyz /kernel: object inconsistant state: RPC: 3, RC: 0
:
: Try fsck'ing all your filesystems manually from single-user.
:
: It also looks like this code segment runs at splbio(), which
: is not high enough since vm_page_free() can mess with the vnode_tobefree
: list without using the vnode_free_list_slock. It needs to run at splvm()
: at the very least. Try changing:
:
: s = splbio();
:
: to:
:
: s = splvm();
I take it back. No network interrupt should be calling vm_page_remove()
(directly or indirectly) - unless there is a bug in one of the interrupt
service routines that run at splnet... for example, a routine calling a
normal kernel malloc() with the wrong arguments causing it to recycle a
cache page. If changing splbio() to splvm() fixes the problem, there's
a bug in a network driver somewhere. If it doesn't, then that isn't the
problem and the code should remain at splbio().
-Matt
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199901012335.PAA18223>