a docker network latency plotter coded which is a part of the bundle docker swarm management and this subpart will check and plot all the docker network latency and also prune them if they are sleeping or are dead. this is the subpart of the docker swarm and management bundle which i already finished coding and pushing in a short while. A snippet of which is here if you want to use it separately. if will show you visual graphs for each of the container that is on the network including the bridge and the local ones. A bigger bundle you can find at docker_management which is in release.
I finished almost implementing this sub module into a graphical interface and it looks like and you can search the profile with the regular expression and it will explicit look into the network id. Have a look here and i am integrating this with the streamer library from here for seamless integration into python pyobject and GTK for the linux kernel gtk
when you will press ok, all the options will appear, so a quick launch.
if [[ $ip_address ]]
then
ping "$ip_address" > "${ip_address}".log && cat "${ip_address}".log | \
cut -f 7 -d " " | cut -f 2 -d "=" | uplot bar -d, -t "time latency"
else
for i in $(docker network ls | cut -f 1 -d " " | sed "/NETWORK/d") \
do echo ping "$i" > "${i}".log && cat "${i}".log | cut -f 7 -d " " | \
cut -f 2 -d "=" | uplot bar -d, -t "timelatency"
done
fi
fi
if [[ $network == "yes" ]] &&
[[ $option == "yes" ]]
then
for i in $(docker network ls | cut -f 1 -d " " | sed "/NETWORK/d")
do
echo ping "$i" > "${i}".log && cat "${i}".log
done
for file in $(pwd)/*.log
do
if [[ -f $file ]] &&
[[ -z $file ]]
then
docker network prune -y
# entire code there this is just a part
Gaurav
Academic Staff Member
Bioinformatics
Institute for Biochemistry and Biology
University of Potsdam
Potsdam,Germany