Ticket and error report

Helpdesk + Releases + Feature requests
1

Hi
I hope my message is in the right section, you’ll tell me.
I wish I can report a strange behaviour in the use of the AstroColibri platform, anf more precisely in the use of the get light curve button.
But nothing critical.

First, I use:
image

How to reproduce:

  • Select one transient event from the upper bar
  • In the source of observations, check the desired ones
    image
    image319×165 5.94 KB
  • if one of them at least is empty (thus containing no observation), the get light curve button will run a long process ending with an error message

image
image521×116 8.59 KB

  • conversaly, if all the selected sources are empty, after the get light curve is pressed, the right error message is shown in the graph window

image
image1230×656 45.3 KB

(only RAPAS checked in this case)

Cheers
Fred Denjean

2

Thank you very much for reporting this issue! Yes, this is exactly the right place to do so.

We’ll investigate further, but at the moment it looks like the problem is caused by very low availability of the ATLAS server that provides the photometric data. The server responds, but the data transfer never actually starts. As a result, our server keeps waiting without receiving a proper error message. Eventually, our web interface assumes something went wrong and shows the red error banner you saw.

Unfortunately, since we rely on the official ATLAS software to make these requests, we don’t have much control over this behavior.

For now, the only workaround is to disable ATLAS data in your request.

Separately from the first issue: the retrieval of RAPAS data for AT 2025rkp fails simply because (as far as I know) no RAPAS member has reported observations for that source.

In addition, several other sources (which do have data from RAPAS) currently fail as well due to inconsistencies in the naming of the sheets in the RAPAS data collection. We’re working on implementing a more robust search method to handle these cases, which will be included in our next release.

3

Hi Fabian

Thank you for your fast feedback, it’s great.

I agree that relying on third-party servers can be tricky in case of non-responsiveness.
What I tried to point at is that it is not specific to one or another server.
Rather, it may be that a void return to a request results in high latency (timeout).

As you stated, a good workaround is to desable the servers with no observations.

Thanks for your work and for sharing it

Fred