Hello Janvier,
many thanks for the explanation, now I understand your use case. And more than 100 rows in the scheduler is not a solution, right!

The only thing I tried to clarify that your comment wasn’t a bug report, it is a feature request

The Scheduler module is „slightly“ unhandy in its current code base, and we are doing a lot of code refactoring in the background. We wanted getting this done before we touch the scheduler.

But I‘ll check whether we can provide a solution for your scenario without too massive effort.

HTH
Wolfgang

Hello Wolfgang,

Maybe a check button in Job Completion Conditions than reset all jobs automatically and restart again?

Thank You!!

Sounds interesting, Javier, thanks for the idea.

I'm afraid, if we want to have a simple solution, this will only work if the "Remember Job Progress" option is disabled. The downside will be that we will loose the restarting ability of the scheduler when using this feature. That means, if you did an incomplete session last night, it won't be completed first.

Cheers
Wolfgang

Hello Wolfgang,

I only work with the "Remember Job Progress" option disabled. Because is more simple add rows to the sheduler queue.

See this video to see like I work to make an idea.


Another hand, i dont understand this:

"The downside will be that we will loose the restarting ability of the scheduler when using this feature. That means, if you did an incomplete session last night, it won't be completed first."

Thanks,
Regards

OK, I think I understand. Could you please elaborate a little bit about your use case? What is the intention behind why you are shooting mosaics in combination with iterating over the mosaic positions? Or is the general idea visiting several positions in a row with a few captures per position, and repeating the entire row a certain amount of times or until the end of the session?

Hello Wolfgang,

I'll explain a little more.

To search exoplanets you need detect a variable intensity from light over star, like a sun eclipse, when the planet cross the star the light intensitivity go down a littel bit. For detect a true pattern of this, you need do a lot of captures with short exposure for example 1500 captures or 2000.

The first observation, each night normally are 300 captures by field. (some times its recomended double capture, because if the sky zone there are satellites, the satellite will broken the capture, to solve this doing a couple captures the after program detect satellite and erase it from final file ....)

When i finish to capture 300 (600) lights i go to another side in sky and start again with anoter field.

Finally with a couples of programs I process all information and it return a light curve from each star in field captured, i study each star light curve searching a potential star to be variable star or exoplanet transit. There are programs that searh it for you and return the potential stars in the field captured.

If the result its irrelevant (no stars potentials), then i search another field next night.
But if there are some potential star, I repeat the observation next night but only in this field all night, to capture 1500 or 2000 lights and verify with more data and details and be able to discart star variable or exoplanets transit false.

For now i do this work whitout mosaic, but this limite the night only a couple field.
The mosaic idea is make the field biggest for expand search possibilities.

That is the reason to do a mosaic 4x4 or 8x2 with 300 captures by field, and i detected that the actual scheduler, do not able to do automatically.

For now, the workarround solution is do less than 100 rows in scheduler queue and when finish, clear and restart with another 100 jobs, like video attached in post before. but this means be in front of scheduler to click on reset and start when finish.

The general Idea its go a full mosaic tour (in the example will be 16 rows in scheduler) and when finish , reset all and start again 300 more times if you have a sequency single capture or 150 more times if the sequency are 2 captures, instead of 4800rows for single captures or 2400 rows for double captures in scheduler.

Thanks again, if you need mor information, post here again.

NOTE: Sorry for my english
Regards,
Javier A.

Besides Wolfgang's questions, can you also please share the .esq and .esl files that cause the scheduler to go to 100% CPU for a while and crash? Please also share your approximate location (e.g. city) so I can try and re-create that in the simulator?

I doubt it would help, but also please try with the greedy scheduler algorithm instead of the classic algorithm.

Thanks,
Hy

I try it with greedy scheduler algorithm inform to you soon and i will attach the *.esq and *.esl files
Location city are: Las palmas de Gran Canaria (Spain)

Hello Hy,
I try with greedy scheduler algorithm but this do not solve the problem.
here are the .esq and .esl files.



to simulate this, Location: Las Palmas de Gran Canaria

Javier,
what about this simple "repeat" option? It would be available only if the "Remember Job Progress" is not active, since only in this case we should be able to reset the number of captured frames.


Do you need a certain amount of repeats, or should it simply repeat until all jobs' constraints are no longer met?

Cheers
Wolfgang

Hi Wolfgang,

I think that could be more pro with a certain amount of repeats, but if set to 0, will repeat for undefined time until all jobs' constraints are no longer met.

And this phrase "It would be available only if the "Remember Job Progress" is not active".... Of course!!

Many thanks for answer and give me a quickly workarround

Hi wolfgang,

I readed again your message, and i think:
- A simply solution its like you said. A simply repeat until all jobs' constraints are no longer met.
- A more advance solution, could be with a certain amount of repeats, but if set to 0, will repeat for undefined time until all jobs' constraints are no longer met.

Both solutions are valid and enough to solve the problem.

Many Thanks again!!!