RHEL7: ntpdate can loop when the network is under stress and hw timestamping is enabled
Issue
- This box is an lsyncd mirror server. We suspect it was a number of concurrent rsync processes since we were syncing about 200G of data to ~60 servers globally when the issue first appeared.
- The ntpdate command started looping indefinitely.
Environment
- Red Hat Enterprise Linux 7
- kernel-3.10.0-123.9.3.el7
- Dell PowerEdge R730xd w/ version 1.0.4, 08/28/2014
- 2 socket, 16 cores/threads total
- Intel(R) Xeon(R) CPU E5-2667 v3 @ 3.20GHz
- 64 GB RAM in 8 DIMMs
- 2 socket, 16 cores/threads total
- Mellanox Ethernet
- HW time stamping
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
