DietPi-Config | No-IP: Doubled process appears

[johnnyt83]

Hi Micha,

I just installed No-IP on a new pi zero with DietPi (6.26.3) and noticed it did not go "online" after successfully entering login credentials. I gave "5" as the new refresh interval and noticed this:

https://abload.de/img/thumbnail_img_2019102lpkk9.jpg

https://abload.de/img/thumbnail_img_2019102dwkqz.jpg

CPU usage did not even drop after 10 minutes.
Is the plugin buggy?
BTW: does the plugin prevent me from having to login into the webUI every 30 days to make my domain not expire?
best wishes.

[MichaIng]

@johnnyt83
Many thanks for your report. Strange is that there is already an instance of noip2 running (with close to 0% usage) while the second one then has high usage. Should be only one instance.

I just checked the code:

• https://github.com/MichaIng/DietPi/blob/master/dietpi/dietpi-config#L4517-L4523
  • Service is stopped, when using the option
  • Config file creation (input prompt) is called
  • Service is restarted
• https://github.com/MichaIng/DietPi/blob/master/dietpi/dietpi-software#L8340
  • Leads to process being killed gracefully and should wait/delay until it is really stopped.

Did it probably fail to stop the service? However this should produce an error prompt then. But to check:

systemctl status noip2

---

Found the reason for the wrong online state: 9b70bd5
However this does not explain the doubled binary call yet. Finally systemctl stop noip2 must success to kill/stop all of them via killall -w noip2.

Fix for wrong online state is there, but investigation of doubled process still required.

[MichaIng]

@johnnyt83
I have no No-IP account, hence cannot test 100%, but noip2 -C which causes the 100% CPU usage is the config creation, hence the interactive call where you enter your mail and password. After you did so, does it hang or does Press any key to continue... appear which lets dietpi-config menu reappear?

Once idea I have is that noip2 -C, if credentials are correct, as well goes into background to run as regular daemon. If we then start the service, there would be two instances, probably causing the issue. A reboot should solve it. If this is really true, we need to manually kill the spawned noip2 process before starting the service, or add ExecStartPre=/usr/bin/killall -w noip2 to the service so that is always kills all probably running processes before starting itself. However I doubt that actually, since the noip2 -C process in your case is still running as root while it spawns the actual daemon executed with user nobody.

[MichaIng]

Marking as closed due to missing reply. Feel free to reopen if the issue is still present.