For an unknown reason PRTG server started to malfunction. So it was restarted and afterwards we discovered that the channels ids within the sensor changed. Sensor id has not changed, but all channel ids shuffled. E.g. from 15 to 17, from 11 to 7, etc. Is this the expected behavior? Is there no guarantee that the channel ids will stay the same during the lifetime of the sensor? Under what conditions do they change, what might be the event? A simple restart such as this one? Sensor is of type "REST Custom".
By default, the Channels IDs should not change during the lifetime of the Sensor. Therefore, this is quite strange. Please let me know whether the configuration of PRTG can be saved without an issue. You can check this via creating a backup. Please do this via "Setup > System Administration > Administrative Tools"
Yes, checked it, it worked, no problems. I understand what you are saying about the channel ids. After all, it was one assumption I created my app on. Imagine the damage it did when the channels ids shuffled. And yes, I am certain it happened this way. I have proof in the form of my app log files. There I save the channels config as it comes from PRTG. And when trying to update, at one moment the search using a channel id returned a record, in 5 mins following the restart it returned another record.
Thank you for the update. In the latest PRTG version we have no such errors. Therefore, please validate that you are using the latest PRTG version and how you query the channel IDs.
PRTG version is : "220.127.116.117+"
Thank you for the information.
I queried the channels of my Core Health Sensor "ID:1002" and could not reproduce the issue. However, I also use the latest PRTG version but this should also not occur in the version before. Can you reproduce the issue if you use the URL in your web browser?
No, unfortunately I cannot reproduce it no matter what I do. The whole thing is puzzling. I am absolutely certain this happened as we analyzed the issue within the team and we all agreed it did happen. However we do not know why PRTG behaved like that. It was not the first restart of the server and before there were never problems. In the meantime I have re-written the DB search algorithm for the PRTG objects and now I use another id, I am no longer considering the prtg id being unique. It should solve the future mysterious PRTG behavior, just in case. Thanks!
Thank you for the feedback here. I'm glad the issue no longer appears at the moment. It's really strange and this should not occur in PRTG.
I hope everything works now as expected. Have a nice day!