Yesterday I noticed some Sensor Factory sensors telling me they have no data for all channels (this is in the Overview tab). However, the graphs show me data, and there are values in the table.
Did some more digging: The Sensor Factory sensors displaying this issue have one channel paused. If I add one of the still working channels a 2nd time, all channels suddenly display data again. Removing the duplicate channel again cause all channels to display 'No data'.
I don't want to re-create the Sensor Factory sensors, because they're used in several maps, added to Business Process sensors etc. Also, the now paused channels will come back in a while.
We just updated PRTG to the newest version this afternoon (17.3.33.2753+), but this didn't solve the problem.
- Why are all channels stating 'No data', where there's only one channel paused (and the other channels DO display data on tab 2-5) - Why if I duplicate a channel with data, all of a sudden all other channels with data display their data again - Why if I remove the duplicate channel, all channels go to 'No data' again
Kind regards,
Corné van den Bosch
Article Comments
Hi Stephan,
The Channel Definition:
#1:CdrFileSummary not updated Channel(6272,3) #2:Unknown Msisdn Errors Channel(6274,3) #3:Count declined CDRs Channel(6276,3) #4:CDR Mediation Error Channel(6277,3)
This looks like a pretty standard channel definition to me.
Sensor 6277 here is the paused channel. If I remove it, all other 3 channels show values again. If I re-add the channel, all other 3 channels show 'No data' again in both Channel overview (channel list in the center of the page) and in the gauges (left side of the page). But the graphs actually DO show data, both on the right side of the screen, as well as in their respective tabs.
I have a couple Sensor Factory sensors displaying this problem...
Sep, 2017 - Permalink
I fixed the formatting - what type are the source sensors?
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink
Hi Stephan,
All 4 channels are Microsoft SQL v2 sensors. The 3 active channels as well as the paused channel.
I checked one of the other sensors with this behavior as well: This other sensor has 2x EXE/Script sensors.
Sep, 2017 - Permalink
Hi Corné,
Thanks for clarifying. The issue seems to apply to all sensor types. I'll create an issue ticket regarding the matter and let you know the outcome of it.
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink
Hi Stephan,
Appreciate it :-)
Odd that no one noticed this so far. Am I the only one keeping an eye on my monitoring environment? :-p Or is this maybe only an issue here (can't really believe that)?
Sep, 2017 - Permalink
Hehe :D There already was an issue ticket raised with this. Could you please increase the sensor interval of the factory sensor to be larger than the largest source sensor interval (e.g. if all have 60s and one 90s, set its interval to 100s). This way, I could reproduce the issue and fix it.
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink
Hi Stephan,
Sorry for the delay. Kinda busy here.
The 3 active channels are checking every 5 minutes (one of them scheduled to only check during the day), the paused channel was checking every 60 seconds. The Sensor Factory sensor was set to check every 5 minutes. I've changed it to check every 10 minutes. No direct result though.
Sep, 2017 - Permalink
Hi Cornè,
No worries, same here :) Is your PRTG available from outside of your network so I could check the sensor myself?
You can share the credentials here, I won't publish the posting.
Kind regards,
Stephan Linke, Tech Support
Sep, 2017 - Permalink
Should've been more specific - I don't have access to the object with that ID :)
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink
Oops,
We work with group access rights on our sensors, so the departments only see the sensors they're responsible for.
Changed group membership for your account. You should have full access now...
Sep, 2017 - Permalink
Weird. That account is member of the PRTG group PRTG Administrators. From what I understood, this group has access everywhere, and you can't remove this access.
However, the account was in multiple usergroups. I've changed it, and it's now only member of PRTG Administrators. Maybe that'll help...?
Sep, 2017 - Permalink
| If a Sensor Has No Data | Calculate the factory channels and use zero as source value |
|---|
....et voila :)
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink
Good it's fixed. Still weird it had values in the table and graphs, but nothing in the gauges and saying it has no data on the other channels (whilst, as said, there is data in the channels and graphs).
Thank you for your time! I'll change the other Sensor Factory sensors accordingly.
Sep, 2017 - Permalink
You're most welcome :) The channels don't get calculated then - the graphs seemingly do, however.
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink
Might be related: I noticed a couple of SSH Disk Space sensors claiming they had no data, yet table and graphs actually showed data.
What I did in the end to get rid of the 'no data' message: Change the primary channel to a channel which gave another value than '0 free space' Default channel was initially set to a channel which had 0MB free space.
After changing the default channel, the sensors changed their status from Unknown to OK, and the message from the clusternodes to showing data instead of claiming 'no data'...
Sep, 2017 - Permalink
Hi Corné,
Sorry for the delay in response. The behavior is intended. If the primary channel has no data, it will override the state of the sensor, despite other channels receiving actual data.
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink
Hi Corné,
That might be a bug in the sensor factory sensor. What actual source channels are you using?
Kind regards,
Stephan Linke, Tech Support Team
Sep, 2017 - Permalink