×

INDI Library v1.9.4 Released (17 Jan 2022)

Bimonthly Stable INDI Library release introduces new drivers and fixes for existing ones. Some highlights:

10Micron LX200: different parking behavior when using Scheduler or INDI driver

  • Posts: 213
  • Thank you received: 53
I found that when using the scheduler shutdown procedure to park the mount, the mount switches to 'parked' status as soon as it gets moving.
I mean, it doesn't wait to be actually parked; that's quite dangerous as you have the roof start closing before the mount is in parking position.
Scheduler log:

2021-10-08T16:59:50 Manual shutdown procedure completed successfully.
2021-10-08T16:59:50 Shutdown complete.
2021-10-08T16:59:48 Dome already parked.
2021-10-08T16:59:47 Mount parked.
2021-10-08T16:59:46 Parking mount in progress...

At 16:59:46 the park command is issued and just 1 sec after (16:59:47) the mount is in 'parked' status (while it was still moving). Btw the 16:59:47 'Mount parked' message is not just an info message because the roof is allowed to move.

Further tests don't show the same behavior:
1- If keeping the same configuration but the 10Micron driver is switched to Telescope Simulator: the mount 'parked' status is correctly shown when the mount actually is parked.
2- If the 'park' command is issued by the INDI driver (10Micron_lx200) interface tab, 10Micron driver log:

2021-10-08T15:01:11: [INFO] Mount is parked.
2021-10-08T15:01:11: [INFO] Gstat changed from 2 to 5
2021-10-08T15:01:08: [INFO] Gstat changed from 7 to 2
2021-10-08T15:01:07: [INFO] Parking. 

Also here the 15:01:11 'Mount is parked' is shown only when the mount is physically parked (ok only 4 secs later, but was quite close to parking position in this test).

I'm not sure if it's a 10Micron INDI driver issue as suggested by test 1 because switching to a different driver solves. Or a Kstars Scheduler issue as test 2 would suggest because two different Kstars tabs show different behaviour using the same driver.

Tested on Kstars 3.5.4 / indi 1.9.2
Ferrante

PS: Originally posted on invent.kde.org/education/kstars/-/issues/136 but can be relevant for this forum too.
 
Last edit: 3 months 6 days ago by Ferrante Enriques.
3 months 6 days ago #76603

Please Log in or Create an account to join the conversation.

  • Posts: 916
  • Thank you received: 278
Hi Ferrante,
I need to take a closer look into the code, but I expect it being a driver issue.

Could you please check what happens when you press the park button on the mount tab? I think this uses the same command as the scheduler.

Wolfgang
TSA-120 + epsilon-160 + FSQ-85 + GSO 150/750 | Avalon Linear + M-zero | ASI 1600mm pro + 6200mm pro | KStars/INDI on Raspberry Pi 4/Intel NUC
The following user(s) said Thank You: Ferrante Enriques, Nguyễn Trọng Minh
3 months 4 days ago #76657

Please Log in or Create an account to join the conversation.

  • Posts: 213
  • Thank you received: 53
Hi Wolfgang,
issuing the Park command from the Mount Tab works as expected:
From Mount Tab:

2021-10-16T09:07:21: [INFO] Mount is parked. 
2021-10-16T09:07:21: [INFO] Gstat changed from 2 to 5 
2021-10-16T09:07:13: [INFO] Gstat changed from 0 to 2 
2021-10-16T09:07:12: [INFO] Parking. 

From parking to parked is 9 secs only because the mount was close to parking position.
With the same configuration I checked from the Scheduler:

2021-10-16T11:20:27 Dome parking failed. Restarting operation...
2021-10-16T11:19:54 Parking dome...
2021-10-16T11:19:53 Mount parked.
2021-10-16T11:19:52 Parking mount in progress...
2021-10-16T11:19:51 Cap parked.
2021-10-16T11:19:49 Parking Cap...
2021-10-16T11:19:48 Warming up CCD...
2021-10-16T11:19:48 No jobs left in the scheduler queue.
2021-10-16T11:19:47 Job 'NGC 4960' is complete.
2021-10-16T11:19:38 Job 'NGC 4960' capture is in progress...
2021-10-16T11:19:38 Job 'NGC 4960' slew is complete.
2021-10-16T11:19:37 Job 'NGC 4960' is slewing to target.
2021-10-16T11:19:36 Mount unparked.
2021-10-16T11:19:32 INDI devices connected.
2021-10-16T11:19:29 Scheduler started.
2021-10-16T11:19:29 Scheduler is awake.

The parking distance was the same as the previous test. But here the Parked status is immediately after the Parking command.
(the 2hrs time offset between the 2 tests is due to the mount working in UTC and Ekos using local time)

Thanks
Ferrante
Last edit: 3 months 4 days ago by Ferrante Enriques.
3 months 4 days ago #76664

Please Log in or Create an account to join the conversation.

  • Posts: 916
  • Thank you received: 278
Hi Ferrante,
sorry for the late response, but this confirms what I thought. It looks like the being a driver issue.

Fixing this without a 10micron is very difficult. Are you more or less familiar with building from sources and debugging? I could assist remotely. Just send me a PN.

HTH
Wolfgang
TSA-120 + epsilon-160 + FSQ-85 + GSO 150/750 | Avalon Linear + M-zero | ASI 1600mm pro + 6200mm pro | KStars/INDI on Raspberry Pi 4/Intel NUC
The following user(s) said Thank You: Ferrante Enriques
1 month 2 weeks ago #78277

Please Log in or Create an account to join the conversation.

  • Posts: 213
  • Thank you received: 53
hi Wolfgang,
the behavior is exactly the same as reported in this thread: 
indilib.org/forum/ekos/10807-scheduler-w...ck-policy.html#78185
In this case it's a different mount (CEM70) so it's not likely a 10micron driver issue.
But I can build and debug the 10 micron driver if needed.
​​​​​​ bestFerrante
1 month 2 weeks ago #78283

Please Log in or Create an account to join the conversation.

  • Posts: 213
  • Thank you received: 53
hi Wolfgang,

sorry to bother you with this issue again, but I still struggle to understand the LX200 driver/Scheduler interaction when the shutdown procedure runs.
I see a different behavior when switching from LX200 mount driver to Simulators (Simulators has the expected behavior so the issue is likely on LX200 side), see logs below.

I'm not familiar with the scheduler code, what triggers the 'mount parked' status? It would help me to understand if the LX200 driver is issuing the wrong message.

LX200 logs:
[2021-12-26T21:00:40.086 CET INFO ][           org.kde.kstars.indi] - "FITS" file saved to "/home/ferrante/NGC2081/Light/Luminance/IC_4632_Light_Luminance_1.500_secs_2021-12-26T21-00-40_002.fits"
[2021-12-26T21:00:40.143 CET INFO ][   org.kde.kstars.ekos.capture] - "Download Time: 0.19 s, New Download Time Estimate: 0.20 s."
[2021-12-26T21:00:40.287 CET INFO ][   org.kde.kstars.ekos.capture] - "Received image 2 out of 2."
[2021-12-26T21:00:40.535 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'NGC2081' is complete."
[2021-12-26T21:00:40.700 CET INFO ][ org.kde.kstars.ekos.scheduler] - "No jobs left in the scheduler queue."
[2021-12-26T21:00:40.700 CET INFO ][ org.kde.kstars.ekos.scheduler] - Starting shutdown process...
[2021-12-26T21:00:41.690 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Parking mount in progress..."
[2021-12-26T21:00:41.703 CET INFO ][           org.kde.kstars.indi] - LX200 10micron :  "[INFO] Parking. "
[2021-12-26T21:00:41.729 CET INFO ][           org.kde.kstars.indi] - Hakos Roll Off Roof :  "[INFO] Telescope status changed. Lock is set to: unlocked "
[2021-12-26T21:00:41.962 CET INFO ][           org.kde.kstars.indi] - LX200 10micron :  "[INFO] Gstat changed from 0 to 2 "
[2021-12-26T21:00:42.699 CET INFO ][     org.kde.kstars.ekos.guide] - "Mount is moving. Resetting calibration..."
[2021-12-26T21:00:42.700 CET INFO ][     org.kde.kstars.ekos.guide] - "Calibration is cleared."
[2021-12-26T21:00:42.708 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Mount parked."
[2021-12-26T21:00:43.694 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Parking dome..."
[2021-12-26T21:00:43.696 CET INFO ][org.kde.kstars.ekos.observatory] - "Rolloff roof is closing..."
[2021-12-26T21:00:43.726 CET INFO ][           org.kde.kstars.indi] - Hakos Roll Off Roof :  "[INFO] Roof is parking... "
[2021-12-26T21:00:44.724 CET INFO ][           org.kde.kstars.indi] - Hakos Roll Off Roof :  "[INFO] Dome state PARKING. "
[2021-12-26T21:00:44.725 CET INFO ][org.kde.kstars.ekos.observatory] - "Rolloff roof stopped."
[2021-12-26T21:00:49.732 CET INFO ][           org.kde.kstars.indi] - Hakos Roll Off Roof :  "[INFO] Dome state IDLE. "
[2021-12-26T21:01:04.720 CET INFO ][           org.kde.kstars.indi] - Hakos Roll Off Roof :  "[INFO] Dome state IDLE. "
[2021-12-26T21:01:07.119 CET INFO ][           org.kde.kstars.indi] - LX200 10micron :  "[INFO] Gstat changed from 2 to 5 "
[2021-12-26T21:01:07.129 CET INFO ][           org.kde.kstars.indi] - LX200 10micron :  "[INFO] Mount is parked. "
[2021-12-26T21:01:10.118 CET INFO ][           org.kde.kstars.indi] - Hakos Roll Off Roof :  "[INFO] Resuming Park command. "
[2021-12-26T21:01:10.119 CET INFO ][org.kde.kstars.ekos.observatory] - "Rolloff roof is closing..."
[2021-12-26T21:01:10.128 CET INFO ][           org.kde.kstars.indi] - Hakos Roll Off Roof :  "[INFO] Roof is parking... "
...
It seems that the scheduler sets the mount as parked at 21:00:42.708 when the LX200 driver is still parking ('Gstat changed from 0 to 2'*).
But the LX200 driver issues the mount parked message ~25 sec later 21:01:07.119 ('Gstat changed from 2 to 5'*).

While with simulators, the driver and scheduler act as expected.
Simulators log:
[2021-12-28T12:31:26.689 CET INFO ][           org.kde.kstars.indi] - "FITS" file saved to "/home/ferrante/NGC_7242/Light/Luminance/IC_4632_Light_Luminance_1.500_secs_2021-12-28T12-31-26_004.fits"
[2021-12-28T12:31:26.753 CET INFO ][   org.kde.kstars.ekos.capture] - "Download Time: 0.20 s, New Download Time Estimate: 0.21 s."
[2021-12-28T12:31:26.890 CET INFO ][   org.kde.kstars.ekos.capture] - "Received image 2 out of 2."
[2021-12-28T12:31:27.026 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Job 'NGC 7242' is complete."
[2021-12-28T12:31:27.360 CET INFO ][ org.kde.kstars.ekos.scheduler] - "No jobs left in the scheduler queue."
[2021-12-28T12:31:27.361 CET INFO ][ org.kde.kstars.ekos.scheduler] - Starting shutdown process...
[2021-12-28T12:31:28.328 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Parking mount in progress..."
[2021-12-28T12:31:28.330 CET INFO ][           org.kde.kstars.indi] - Telescope Simulator :  "[INFO] Parking to RA:  1:05:33 - DEC:  0:00:00 "
[2021-12-28T12:31:29.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - Parking mount in progress...
[2021-12-28T12:31:29.336 CET INFO ][     org.kde.kstars.ekos.guide] - "Mount is moving. Resetting calibration..."
[2021-12-28T12:31:29.337 CET INFO ][     org.kde.kstars.ekos.guide] - "Calibration is cleared."
[2021-12-28T12:31:30.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - Parking mount in progress...
[2021-12-28T12:31:31.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - Parking mount in progress...
[2021-12-28T12:31:32.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - Parking mount in progress...
[2021-12-28T12:31:33.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - Parking mount in progress...
[2021-12-28T12:31:34.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - Parking mount in progress...
[2021-12-28T12:31:35.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - Parking mount in progress...
[2021-12-28T12:31:35.402 CET INFO ][           org.kde.kstars.indi] - Telescope Simulator :  "[INFO] Mount is parked. "
[2021-12-28T12:31:35.403 CET INFO ][           org.kde.kstars.indi] - Dome Scripting Gateway :  "[INFO] Telescope status changed. Lock is set to: unlocked "
[2021-12-28T12:31:35.404 CET INFO ][           org.kde.kstars.indi] - Telescope Simulator :  "[INFO] Telescope slew is complete. Parked "
[2021-12-28T12:31:36.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Mount parked."
[2021-12-28T12:31:37.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Dome already parked."
[2021-12-28T12:31:39.327 CET INFO ][ org.kde.kstars.ekos.scheduler] - "Shutdown complete."
[2021-12-28T12:31:39.329 CET INFO ][ org.kde.kstars.ekos.scheduler] - Scheduler is stopping...

Thanks,
Ferrante

* As to the LX200 driver, the status codes are:
0, GSTAT_TRACKING: SCOPE_TRACKING;
1, GSTAT_STOPPED: SCOPE_IDLE;
2, GSTAT_PARKING: SCOPE_PARKING;
3, GSTAT_UNPARKING: SCOPE_TRACKING;
4, GSTAT_SLEWING_TO_HOME: SCOPE_SLEWING;
5, GSTAT_PARKED: SCOPE_PARKED;
 
Last edit: 3 weeks 2 days ago by Ferrante Enriques.
3 weeks 2 days ago #78888

Please Log in or Create an account to join the conversation.

  • Posts: 916
  • Thank you received: 278
Hi Ferrante,
that's really weird. It indeed looks like the 10micron driver has a problem reporting the parking process. Could you please check what happens when you manually press the "PARK" button in the INDI dialog? The sequence should be the following:
 
 
 

As long as the mount slews to its parking position, the LED should turn yellow and switch to green as soon as the parking position has been reached.
Cheers
Wolfgang
 
TSA-120 + epsilon-160 + FSQ-85 + GSO 150/750 | Avalon Linear + M-zero | ASI 1600mm pro + 6200mm pro | KStars/INDI on Raspberry Pi 4/Intel NUC
The following user(s) said Thank You: Ferrante Enriques
3 weeks 1 day ago #78897
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 213
  • Thank you received: 53
hi Wolfgang,
you're absolutely right: The parking light is always green, also during slewing and parking. But the messages in the log are reporting parking and parked correctly (see attachments).
So the scheduler rely on the status light been green/yellow to proceed to the next shutdown step? isn't there a 'parking status' variable in Ekos? 

Thanks,
Ferrante


 
 
  
3 weeks 1 day ago #78902
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 916
  • Thank you received: 278
Sure, there is a parking variable (and maybe even more than one :-) ) But this variable is derived from information the INDI driver sends, which is displayed in the INDI driver tabs. As you could see from your screenshots, there is no difference in the Parking variable between PARKING and PARKED.

That's why the scheduler assumes the mount being parked although the slew is in progress. The PARKING state is signaled by "Park(ed)" selected and the status LED is yellow. As soon as it turns to green, the state changes to PARKED.

It looks like we need to fix the INDI driver for 10micron.

Wolfgang
TSA-120 + epsilon-160 + FSQ-85 + GSO 150/750 | Avalon Linear + M-zero | ASI 1600mm pro + 6200mm pro | KStars/INDI on Raspberry Pi 4/Intel NUC
The following user(s) said Thank You: Ferrante Enriques
3 weeks 1 day ago #78905

Please Log in or Create an account to join the conversation.

  • Posts: 213
  • Thank you received: 53
Looking at the code seems that the park() method is not managing lights. 
I will open an issue on github providing these informations. Probably Jasem should look into it because the bug could also be in the parent lx200 and I immediately got lost in the class hierachy.

Thanks a lot for your help.
Ferrante
3 weeks 1 day ago #78921

Please Log in or Create an account to join the conversation.

  • Posts: 916
  • Thank you received: 278
I did a first check in the code. For me surprisingly LX200_10MICRON is derived from LX200Generic and not from LX200Telescope, in contrast to the telescope simulator. Maybe that change would help, but I do not have a mount for testing.

@Ferrante: if you want to test it, I could make these changes on my INDI fork.

Wolfgang
TSA-120 + epsilon-160 + FSQ-85 + GSO 150/750 | Avalon Linear + M-zero | ASI 1600mm pro + 6200mm pro | KStars/INDI on Raspberry Pi 4/Intel NUC
The following user(s) said Thank You: Ferrante Enriques
3 weeks 21 hours ago #78928

Please Log in or Create an account to join the conversation.

  • Posts: 213
  • Thank you received: 53
 it seems that Jasem already fixed this issue as part of another one.
github.com/indilib/indi/issues/1582

I will test and report here if it works fine with the scheduler as well. thanks again for your support.

ferrante
3 weeks 12 hours ago #78935

Please Log in or Create an account to join the conversation.

Time to create page: 0.806 seconds