[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: problem with kvm on virtual server

From: Martin Pala
Subject: Re: problem with kvm on virtual server
Date: Thu, 14 Apr 2005 01:47:58 +0200
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.6) Gecko/20050324 Debian/1.7.6-1

This is related to the virtual server environment permissions - freebsd's kvm_open call failed.

I will look on it (monit should not stop given this condition) ... possible workaround could be to not use process checks in restricted virtual server environment (--without-resource configure option).


Mike Jackson wrote:
I'm the sys admin for a small web hosting company. We have several virtual servers from Verio. Here's the uname -a output from the one where I'm testing the 4.5.1 install:

FreeBSD 4.7-RELEASE-p28 FreeBSD 4.7-RELEASE-p28 #40: Fri Mar 11 14:16:49 MST 2005 address@hidden:/usr/src/sys/compile/VKERN i386

We've been using monit 4.1.1. I know there were some issues getting it working in this environment, but that was before my time here. I just compiled 4.5.1; I had to turn off SSL support to get it to compile (gmake failed at some point, but I didn't save the output). However, upon launching it I get this error:

cypress1# ./monit -I -c /usr/local/etc/monitrc
Starting monit daemon with http interface at [*:2812]
Starting monit HTTP server at [*:2812]
monit HTTP server started
system statistic error -- cannot initialize kvm interface
AssertException: at process.c:390
Abort (core dumped)

I don't know jack about reading core dumps, but here's a relevant-looking snippet from 'strings monit.core':

$FreeBSD: src/lib/libkvm/kvm_i386.c,v 2001/09/21 04:01:51 peter Exp $
cannot allocate vm
bad namelist
cannot read IdlePTD
cannot read PTD
vatop called in live kernel!
_kvm_vatop: lseek
_kvm_vatop: read
invalid address (%x)
$FreeBSD: src/lib/libkvm/kvm.c,v 2002/09/13 14:53:43 nectar Exp $
: %s
: %s
can't allocate %u bytes: %s
exec file name too long
bad flags arg
%s: not physical memory device
%s: %s
invalid address (%x)
short read
short write
kvm_write not implemented for dead kernels
@(#)kvm_proc.c  8.3 (Berkeley) 9/23/93

Any ideas how to work around/fix this issue in this particular environment? I'd stick with the working 4.1.1, but I'd really like to use the nifty new mysql protocol tests on these servers. Thanks for any help you can provide.

To unsubscribe:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]