Replies: 3 comments 4 replies
-
You should always be receiving MQTT messages about new Observations. Even if there are many, and regardless of how the entity was created (Directly, via deep-insert, via batch-requests or via DataArrays). When there flow is really large some messages may be lost, but FROST should complain about queue overruns when that happens, and indicate which settings control the relevant queue. Usually this only happens when using DataArrays to create Observations. |
Beta Was this translation helpful? Give feedback.
-
Do you mean Observations sent as an array to frost? Like: As far as i can see it from our script code, we do send multiple observations as an array to FROST. But i do not see any errors in the docker frost logs. Also i don't get any result when greping the container internal logs for "overrung":
|
Beta Was this translation helpful? Give feedback.
-
See the DataArray extension in the specification: https://docs.opengeospatial.org/is/18-088/18-088.html#create-observation-dataarray The message you would be looking for is: Failed to add message to send-queue But even if there are buffer overruns, you should at least get some MQTT messages. If you are getting no messages at all, they you are either using the wrong subscription, or there is another issue somewhere. |
Beta Was this translation helpful? Give feedback.
-
Hi,
i am not receiving any updates via MQTT, when large amount of Observations are imported into FROST via HTTP/REST. Is this intended behaviour?
On the other hand, when i create an Observation with my python script (using the Python FROST client), from my dev machine i do get a published message in my MQTT-Explorer.
In both cases i communicate with the same server.
Here's the python script:
Do understand the concept correctly: I should receive newly published entities via MQTT independently of how the entity has been created in FROST?
Beta Was this translation helpful? Give feedback.
All reactions