kdump takes a long time to collect vmcore
Issue
- A crash dump stopped responding in the kdump kernel context due to an excessive interrupt load.
[ 6.986829] hpet1: lost 119 rtc interrupts
[ 8.729969] hpet1: lost 109 rtc interrupts
- Kdump slowly collect vmcore. It took around two hours to complete collecting vmcore. While capturing vmcore, the kdump console continuously shows: hpet1: lost XXXX rtc interrupts
Environment
- Red Hat Enterprise Linux 8.1
- kernel-4.18.0-147.64.1.el8_1
- Lenovo ThinkSystem SR650 V2
- HPE Apollo 2000 Gen10 Plus
- HPE ProLiant MicroServer Gen10 Plus
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.