Maybe it’s too late for an answer, but perhaps this could help others who bump into similar issue with kernel /proc
settings.
I guess a container has a virtual link to the /proc
path of the host it’s running on. It’s automatically mounted into the container.
In my case, an app warns that /proc/sys/net/core/rmem_max
and /proc/sys/net/core/wmem_max
must have values greater than the default. I edit /etc/sysctl.conf
on the host it’s running on to add the following:
# Allow a 25MB UDP receive buffer for JGroups
net.core.rmem_max = 26214400
# Allow a 1MB UDP send buffer for JGroups
net.core.wmem_max = 1048576
I run sysctl -p
, then I deploy the app. The app stops giving me the warnings.