IBM Support

The LPM operation with a heavy workload on a large LPAR might result cause the LPAR to hang

News


Abstract

If you initiate a Live Partition Mobility (LPM) operation on a large logical partition (LPAR) that has more than 60 cores (480 CPUs) and memory of 8 TB or more, and when you run a heavy workload such as an Online Analytical Processing (OLAP) workload, the LPAR might hang or seem to hang, the network might not respond, and the partition console might get blocked. This behavior might be related to queued spin locking mechanism which is first enabled in SUSE Linux® Enterprise Server 15, Service Pack 3.

Content

Linux Releases Affected
SUSE Linux Enterprise Server 15, Service Pack 3

IBM Systems Affected
All IBM POWER9™ and Power10 systems that support SLES 15, SP3

Workaround

You can terminate the LPM operation. The LPAR might resume operation and applications that are running on the LPAR might cause the network connection requests to time out due to the network interruption. Reducing the workload before the LPM operation might help the LPM operation to complete successfully.

Fix Outlook

IBM is working with SUSE to release a fix for this issue. The fix for this issue should come as part of a future SLES release. Open a support ticket with SUSE if a test fix is needed before the next release.

[{"Business Unit":{"code":"BU054","label":"Systems w\/TPS"},"Product":{"code":"HW1W1","label":"Power -\u003EPowerLinux"},"ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Version(s)","Line of Business":{"code":"","label":""}}]

Product Synonym

SUSE Linux Enterprise Server 15, Service Pack 3

Document Information

Modified date:
22 September 2021

UID

ibm16218288