- We have a tested, repeatable problem where our database client/server connection fails with a connection timed out / connection reset that most likely is caused by a tcp keepalive timeout.
- After upgrading from kernel 3.10.0-514.el7 our application reports connectivity loss for 30 or more seconds
- Red Hat Enterprise Linux (RHEL) 7.5 and later
- Physical, as well as virtual (eg. also applies to VMware vSphere, Microsoft Azure Cloud)
- 10 Gbps network interface, drivers such as
mlx5_core, and others
- Virtual 10 Gbps network interfaces, drivers such as
hv_netsvc, and others
- TCP network connection
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.