- Documentation
Elastic Compute Service
Elastic Compute Service
Troubleshooting
Operating system issue
Linux operating system issue
How do I resolve the abnormal kworker load issue caused by rhashtable problems in Debian 9?
Linux operating system issue
How do I resolve the abnormal kworker load issue caused by rhashtable problems in Debian 9?
How do I resolve the abnormal kworker load issue caused by rhashtable problems in Debian 9?
Problem description
In the Debian 9 (EOL) operating system, there is a probability that the kernel kworker process occupies nearly 100% of the CPU even under low-load conditions.
Problem description:
When I use the top command to view the CPU usage of the instance, the kernel kworker process occupies 100% of the CPU.
Causes
This issue occurs because when the server runs workloads with a large number of bursty fragmented packets, it may get stuck in a state where the rht_shrink() function returns -EEXIST. This triggers continuous rescheduling of the delayed work of the hash table, thereby occupying the CPU core.
Solutions
- Solution 1: Replace the operating system
- Solution 2: Install a newer version of the kernel
- In the Backports repository of Debian 9, this issue has been fixed in the kernel version 4.19+105+deb10u4~bpo9+1. You can upgrade the kernel version based on your needs. Learn more.
Last updated: 2026.06.18 17:49:50