Crash Kernel Boot Runs Out of Memory and Fails With Large Logical Unit Number (LUN) Configurations
This document (7016542) is provided subject to the disclaimer at the end of this document.
SUSE Linux Enterprise Server 11 Service Pack 4 (SLES 11 SP4)
SUSE Linux Enterprise Server 12
On an Integrity Superdome X with BL920s Gen8 Server Blades or an HP ProLiant DL980 G7 server running SUSE LINUX Enterprise Server 11 SP3 and configured with a large number of logical unit numbers (which is typical with multipathing; for example, more than 250 LUN paths), the crash kernel boot may run out of memory, display output, and fail.
SUSE LINUX Enterprise Server 11 SP3 example output before the crash kernel boot unresponsiveness:
To avoid running out of memory booting the crash kernel on large LUN configurations, perform the following:
In /etc/sysconfig/kdump , add the kernel command line parameter:
udevd.children-max=2Rebuild the kdump initrd as follows:
For SUSE LINUX Enterprise Server 11, type the following command:
service boot.kdump restart
In addition, increase the crash kernel size to 768M. Larger sizes may be attempted up to the maximum of 896M, if required.
Each additional LUN requires approx. 50 KiB. If a system is connected to 100 logical devices through two FC switches, it will see 200 LUNs, increasing the required memory by 50*200 = 10,000 KiB.
Note: The exact memory consumption depends on which driver handles the LUNs, so it may be different for other HBAs
This Support Knowledgebase provides a valuable tool for 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:7016542
- Creation Date: 28-May-2015
- Modified Date:03-Mar-2020
- SUSE Linux Enterprise Server
For questions or concerns with the SUSE Knowledgebase please contact: tidfeedback[at]suse.com