raaaimund / internet-speed-monitor Goto Github PK

When i try to start the stack with portainer via repository i get the following error:

Container internet-speed-monitor-speedtester-1 Started 
Error response from daemon: OCI runtime create failed: container_linux.go:367: starting container process caused: process_linux.go:495:
container init caused: rootfs_linux.go:60: mounting "/data/compose/21/telegraf/telegraf.conf" to rootfs at "/var/lib/docker/overlay2/17ac96b8523ba2911532177a13c052635cb5b03b825d1a02c07aade1db8b59eb/merged/etc/telegraf/telegraf.conf" 
caused: not a directory: unknown: Are you trying to mount a directory onto a file (or vice-versa)? Check if the specified host path exists and is the expected type

Is there is difference in using portainer or running it by hand? I am not very familiar with compose. I know basic docker.

speedtest.net has changed cli connection, and failing, is there a solution to correct this error?

#Error for speedtest.net
crond: USER root pid 12 cmd /usr/local/bin/speedtest --accept-license -s 5025 -f json > $(mktemp -u -p /var/log/speedtest XXXXXX).json
speedtester_1 | [2021-04-20 13:15:06.201] [error] Configuration - SSL connect error (UnknownException)
speedtester_1 | [2021-04-20 13:15:06.201] [error] Configuration - Cannot retrieve configuration document (0)
speedtester_1 | [2021-04-20 13:15:06.202] [error] ConfigurationError - Could not retrieve or read configuration (Configuration)
speedtester_1 | [2021-04-20 13:15:06.202] [error] ConfigurationError - Could not retrieve or read configuration (Configuration)
speedtester_1 | {"type":"log","timestamp":"2021-04-20T13:15:06Z","message":"Configuration - Could not retrieve or read configuration (ConfigurationError)","level":"error"}

Nice project!

Maybe adding a tag that we have fixed values for cpu architecture might help others to start faster

e.g. armhf in the following lines needs to be changed to amd64 on x86 host systems
https://github.com/raaaimund/internet-speed-monitor/blob/master/telegraf/Dockerfile#L4-L6

speedtest/Dockerfile:

Consider updating this line from:
RUN echo "${SPEEDTEST_CRON_SCHEDULE} root /usr/bin/speedtest --accept-license --accept-gdpr -s ${SPEEDTEST_SERVER} -f json > \$(mktemp -u -p /var/log/speedtest XXXXXX).json" > /etc/cron.d/speedtest
...to:
RUN echo "${SPEEDTEST_CRON_SCHEDULE} root /usr/bin/speedtest --accept-license --accept-gdpr -s ${SPEEDTEST_SERVER} -f json > /var/log/speedtest/results.json" > /etc/cron.d/speedtest

There is really no need to dump the results into unique json file every time. From long-term perspective and in conjuction with the speedtest being run every 1 minute, you will end up having 1440 files daily, 525k files in one year !!!!! That's crazy overkill and waste of resources.

Especially since telegraf agent is set to read all the files every 10s hence it reads THE SAME DATA of all thousands and thousand of files 6 times every one minute !!!!
Surely you can update telegraf agent interval to "1m" but that won't fix the inefficient usage of filesystem due to gazillion of files created by the speedtest and then being read by telegraf.

In the end, here below is the best of the best approaches.....
By doing this, telegraf agent executes the speedtest binary which dumps the json into stdout and telegraf reads and processed stdout and writes into influxdb.
Another advantage is also that you end up having just single container with both telegraf and speedtest in one nice package :)

[[inputs.exec]]
commands = ["/usr/bin/speedtest -f json-pretty"]
name_override = "Speedtest"
timeout = "1m"
interval = "5m"
data_format = "json"
json_string_fields = [ "interface_externalIp",
                       "server_name",
                       "server_location",
                       "server_host",
                       "server_ip",
                       "result_url" ]