Date: Mon, 11 Jun 2007 08:04:20 GMT From: KES<kes-kes@yandex.ru> To: freebsd-gnats-submit@FreeBSD.org Subject: sparc64/113556: panic: trap: memory address not aligned; Rebooting... Message-ID: <200706110804.l5B84KsQ094551@www.freebsd.org> Resent-Message-ID: <200706110810.l5B8A4wI003353@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 113556
>Category: sparc64
>Synopsis: panic: trap: memory address not aligned; Rebooting...
>Confidential: no
>Severity: critical
>Priority: high
>Responsible: freebsd-sparc64
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Mon Jun 11 08:10:03 GMT 2007
>Closed-Date:
>Last-Modified:
>Originator: KES
>Release: FreeBSD 6.2-STABLE #2: Mon May 14 17:40:17 EEST 2007
>Organization:
Aldec-Kharkov
>Environment:
[root@sparc /usr/home/aldec]# uname -a
FreeBSD sparc.aldec.com.pl 6.2-STABLE FreeBSD 6.2-STABLE #2: Mon May 14 17:40:17 EEST 2007 aldec@sparc.aldec.com.pl:/usr/obj/usr/src/sys/KES_KERN_SUN_v6 sparc64
>Description:
This two commands is a cause of server rebooting:
sysctl net.link.ether.bridge.enable=1
sysctl net.link.ether.bridge.config=rl0,rl1
------------------------------------
This also rebooting the server.
Here my mailing with mpd-users@
>How-To-Repeat:
1)
sysctl net.link.ether.bridge.enable=1
sysctl net.link.ether.bridge.config=rl0,rl1
2) Install mpd and UP vpn link
after connection server goes down
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
>KES wrote:
>> Hi.
>> My machine is:
>> FreeBSD 6.2-STABLE #1: Tue Apr 3 08:42:50 EEST 2007
>>
>> Timecounter "tick" frequency 360000000 Hz quality 1000
>> real memory = 201326592 (192 MB)
>> avail memory = 178716672 (170 MB)
>> cpu0: Sun Microsystems UltraSparc-IIi Processor (360.00 MHz CPU)
>
>MPD version? I've tested 4.1 on same arch and it works fine for me.
>
>> Connecting to mpd causing server to REBOOT
>> mpd.links
>> -------------------
>> pptp0:
>> set link type pptp
>>
>> mpd.conf
>> -------------------
>> default:
>> load pptp0
>> pptp0:
>> new -i ng00 pptp0 pptp0
>> load pptp_standart
>> pptp_standart:
>> set iface disable on-demand
>> set bundle disable multilink
>> set link no pap
>> set link enable chap
>> set link keep-alive 60 180
>> set ipcp yes vjcomp
>> #XXXX and YYYY -- my IP addresses
>> set ipcp dns 1.2.3.4
>> set ipcp dns 5.6.7.8
>> set iface disable proxy-arp
>> set pptp enable incoming
>> set pptp disable originate
>> set link yes acfcomp protocomp
>> log +auth
>
>Try looking at the console. The problem can be caused not directly by
>mpd. What exactly says kernel when failing?
#####################
NEXT MAIL
#####################
My mpd Version is:
[root@sparc /usr/home/aldec]# mpd4 -v
Version 4.1 (root@sparc.aldec.com.pl 14:10 29-Mar-2007)
I have ran mpd4 and try to connect from WinXP SP2.
[root@sparc /]# mpd4 > /vpntest
after server fail
[root@sparc /]# cat /vpntest
Multi-link PPP daemon for FreeBSD
process 1378 started, version 4.1 (root@sparc.aldec.com.pl 14:10
29-Mar-2007)
Label 'startup' not found
[pptp0] using interface ng0
[pptp1] using interface ng1
PPTP connection from 192.168.0.1 4765
pptp0: attached to connection with 192.168.0.1 4765
[pptp0] Accepting PPTP connection
[pptp0] opening link "pptp0"...
[pptp0] link: OPEN event
[pptp0] LCP: Open event
[pptp0] LCP: state change Initial --> Starting
[pptp0] LCP: LayerStart
[pptp0] attaching to peer's outgoing call
[pptp0] link: UP event
[pptp0] link: origination is remote
[pptp0] LCP: Up event
[pptp0] LCP: state change Starting --> Req-Sent
[pptp0] LCP: SendConfigReq #1
ACFCOMP
PROTOCOMP
MRU 1500
MAGICNUM c62da0c6
AUTHPROTO CHAP MSOFTv2
pptp0-0: ignoring SetLinkInfo
[pptp0] LCP: rec'd Configure Request #0 link 0 (Req-Sent)
MRU 1400
MAGICNUM 619e59e8
PROTOCOMP
ACFCOMP
CALLBACK
Not supported
[pptp0] LCP: SendConfigRej #0
CALLBACK
[pptp0] LCP: rec'd Configure Request #1 link 0 (Req-Sent)
MRU 1400
MAGICNUM 619e59e8
PROTOCOMP
ACFCOMP
[pptp0] LCP: SendConfigAck #1
MRU 1400
MAGICNUM 619e59e8
PROTOCOMP
ACFCOMP
[pptp0] LCP: state change Req-Sent --> Ack-Sent
[root@sparc /] cat /var/log/dmesg.today
panic: trap: memory address not aligned
Uptime: 1d5h46m18s
Cannot dump. No dump device defined.
Automatic reboot in 15 seconds - press a key on the console to abort
Rebooting...
Copyright (c) 1992-2007 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
The Regents of the University of California. All rights reserved.
FreeBSD is a registered trademark of The FreeBSD Foundation.
FreeBSD 6.2-STABLE #1: Tue Apr 3 08:42:50 EEST 2007
root@sparc.aldec.com.pl:/usr/obj/usr/src/sys/KES_KERN_SUN_v6
module_register: module ng_mppc already exists!
Module ng_mppc failed to register: 17
Timecounter "tick" frequency 360000000 Hz quality 1000
#########
NEXT MAIL
#########
>KES wrote:
>> My mpd Version is:
>> [root@sparc /usr/home/aldec]# mpd4 -v
>> Version 4.1 (root@sparc.aldec.com.pl 14:10 29-Mar-2007)
>
>> [root@sparc /] cat /var/log/dmesg.today
>> panic: trap: memory address not aligned
>
>This is your pain. Some netgraph modules should be rewrited because they
> do address memory the way sparc don't adore. At least I'm sure that
>ng_netflow is unusable on sparc.
>
>Try recompiling kernel without most NETGRAPH modules. You actually need
>only this 2 lines in your kernel config for mpd to work:
>
>options NETGRAPH
>options NETGRAPH_ETHER
>
>All other modules would be loaded automatically.
#######
NEXT MAIL
#######
I have recompiled kernel.
Server still rebooting. But mpd seems fail earlier
#cat /vpntest
Multi-link PPP daemon for FreeBSD
process 1044 started, version 4.1 (root@sparc.aldec.com.pl 14:10
29-Mar-2007)
Label 'startup' not found
[pptp0] using interface ng0
[pptp1] using interface ng1
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200706110804.l5B84KsQ094551>