Configure a XEN Virtual Machine (domu) for a kernel / core dump
This document (7001384) is provided subject to the disclaimer at the end of this document.
Environment
Novell Open Enterprise Server (NetWare 6.5)
SUSE Linux Enterprise Server 10 Service Pack 1
SUSE Linux Enterprise Server 10 Service Pack 2
SUSE Linux Enterprise Server 10 Service Pack 3
SUSE Linux Enterprise Server 11
SUSE Linux Enterprise Server 11 Service Pack 1
XEN
SUSE Linux Enterprise Server 10 Service Pack 1
SUSE Linux Enterprise Server 10 Service Pack 2
SUSE Linux Enterprise Server 10 Service Pack 3
SUSE Linux Enterprise Server 11
SUSE Linux Enterprise Server 11 Service Pack 1
XEN
Situation
XEN Virtual Machine (domu) is panicking / crashing /
abending.
Need to get kernel dump or core dump from a virtual machine (domu)
Need to get kernel dump or core dump from a virtual machine (domu)
Resolution
How to Configure XEN Virtual
Machines for kernel dumps
Configure a SLES 10 or SLES11 vm for a vmcore
On the XEN host (dom0)
1. Modify the following line in /etc/xen/xend-config.xsp
4. Test the configuration:
Configure NetWare for a core dump
System Requirements:
SLES 11 Host (unavailable for SLES 11 SP1 Host)
OES2 and Virtualized NetWare (NW65sp7 or later)
Install:
1. On Dom0 get, and install, the corresponding novell-nwcore rpm to facilitate coredumps on virtulized NetWare. Cliick below for current link:
Installation instructions:
System Requirements:
OES2 and Virtualized NetWare (NW65sp7 or NW65sp8)
Installation:
Install the correct rpm for SLES version and architecture as root user. Note that there will be two directories: One for SLES 10 SP1 and one for SLES10 SP2. Use the SLES 10 SP2 rpm's for SLES 10 SP2, SP3 and SLES 11 XEN hosts.
rpm -ivh novell-nwcore-1.0.4-1...
Managing the service:
Start ./etc/init.d/novell-nwcore start
Stop ./etc/init.d/novell-nwcore Stop
Status ./etc/init.d/novell-nwcore status
Default location for writing coredump:
/var/log/nwcore
If you wish to change the location of where the coredump is written, modify the /etc.init.d/novell-nwcore file and change the following:
"ARGS="/var/log/nwcore""
Test:
1- Enter 386debug on the NetWare VM server console.
Check the directory specified in example above to make sure the .img file (coredump) was written.
Known Problems and Limitations:
Notes: The keyboard may be slow to respond to key strokes while in the debugger. The system may become unresponsive if the ++ keys are pressed prior to coredump process being completed.
Configure a SLES 10 or SLES11 vm for a vmcore
On the XEN host (dom0)
1. Modify the following line in /etc/xen/xend-config.xsp
uncomment and change (enable-dump
no) to (enable-dump yes)
2. Modify the vm config file for desired behaviorEdit the following line(s) in the
/etc/xen/images/vm/<vm name> file:
on_crash=”coredump-restart” or "coredump-destroy"
3. As ROOT run rcxend
restart from a terminal4. Test the configuration:
- At a terminal prompt, shutdown corresponding vm if running and restart using
- xm create <vm name>
- Connect to the vm console by running xm console <vm name>& login as ROOT
- Run modprobe crasher call_panic=1
- Check /var/lib/xen/dump for the dump file
- As root user, from a terminal on XEN Host (dom0) run xm dump-core <vm name>
- Check /var/log/dump on the XEN Host (dom0)
Configure NetWare for a core dump
System Requirements:
SLES 11 Host (unavailable for SLES 11 SP1 Host)
OES2 and Virtualized NetWare (NW65sp7 or later)
Install:
1. On Dom0 get, and install, the corresponding novell-nwcore rpm to facilitate coredumps on virtulized NetWare. Cliick below for current link:
Installation instructions:
System Requirements:
OES2 and Virtualized NetWare (NW65sp7 or NW65sp8)
Installation:
Install the correct rpm for SLES version and architecture as root user. Note that there will be two directories: One for SLES 10 SP1 and one for SLES10 SP2. Use the SLES 10 SP2 rpm's for SLES 10 SP2, SP3 and SLES 11 XEN hosts.
rpm -ivh novell-nwcore-1.0.4-1...
Managing the service:
Start ./etc/init.d/novell-nwcore start
Stop ./etc/init.d/novell-nwcore Stop
Status ./etc/init.d/novell-nwcore status
Default location for writing coredump:
/var/log/nwcore
If you wish to change the location of where the coredump is written, modify the /etc.init.d/novell-nwcore file and change the following:
"ARGS="/var/log/nwcore""
Test:
1- Enter 386debug on the NetWare VM server console.
This will cause the server to enter
the debugger.Use the normal commands to navigate the
debugger. Note: ".c" will initiate the coredump.
2- Answer the prompts as follows for a typical core dump:Reboot the server after coredump
(y/n) - (Administrators choice)
Coredump type: 1) Full coredump 2)
Full without Cache - (2 is usually sufficient)
Compress coredump? 1) Yes 2)
No - (1 is usuall preferred)
Where should diagnostic coredump be
sent? 1) XENHOST 2) NETWORK 3) HDISK
Option"1" should be selected
to send the coredump to DOM(0).
Check the directory specified in example above to make sure the .img file (coredump) was written.
Known Problems and Limitations:
Notes: The keyboard may be slow to respond to key strokes while in the debugger. The system may become unresponsive if the ++ keys are pressed prior to coredump process being completed.
Additional Information
As stated above, use the rpm's created in the sles10-sp2 directory for SLES10 SP2, SLES10 SP3 & SLES11 XEN hosts.
Disclaimer
This Support Knowledgebase provides a valuable tool for NetIQ/Novell/SUSE customers and parties interested in our products and solutions to acquire information, ideas and learn from one another. Materials are provided for informational, personal or non-commercial use within your organization and are presented "AS IS" WITHOUT WARRANTY OF ANY KIND.
- Document ID:7001384
- Creation Date:17-SEP-08
- Modified Date:30-APR-12
- SUSESUSE Linux Enterprise Server
- NetIQPlateSpin Orchestrate
Did this document solve your problem? Provide Feedback