×

INDI Library v2.0.7 is Released (01 Apr 2024)

Bi-monthly release with minor bug fixes and improvements

Read More...

Issue indi_v4l2_ccd: frame received while not exposing, force-aborting capture

  • Posts: 26
  • Thank you received: 1

Issue indi_v4l2_ccd: frame received while not exposing, force-aborting capture was created by Alexandre Chapellon

Whenever I try to do long exposure frames with my spc900 and the v4l2_ccd driver, the driver goes unresponsive and spits the error message bellow repeatedly:

<code>2021-01-31T20:20:23: Driver indi_v4l2_ccd: newFrame: frame received while not exposing, force-aborting capture</code>

I have a webcam Phillips SPC900NC which I modified using the long exposure mod controled by LED.
I have compiled the pwc-lxled driver for my led, which loads successfully and both normal capture (bellow 0.025 seconds or stacked) and streaming works ok in the indi client UI.
On the other hand, long exposure pictures gives me the error above, though I can see the LED turns on or the duration of the exposure and turns off after exposure time. Only then the error starts to appear in the logs.
I have checked the verbose logs but couldn't find anything obvious (see attached file

File Attachment:

File Name: indi_log.gz
File Size:50 KB
).
I'm using a fairly recent version of libindi:

<code>INDI Library: 1.8.7
Code v1.8.6-11-g92c1e617. Protocol 1.7.</code>

The error message seem to suggest the driver didn't really notice it is actually exposing but that doesn't help me much.

Any help is much appreciated as I have spent a fair amount of time moding this cam and would really like to get this working at some point :)
Last edit: 3 years 2 months ago by Alexandre Chapellon.
3 years 2 months ago #66751
Attachments:

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

  • Posts: 26
  • Thank you received: 1

Replied by Alexandre Chapellon on topic Issue indi_v4l2_ccd: frame received while not exposing, force-aborting capture

Taking a look at the code I saw the error occurs in an "else" part of an "if" statement:
if ( PrimaryCCD.isExposing() )
    {
        non_capture_frames = 0;
...
}
    else
    {
        non_capture_frames++;
 
        if( non_capture_frames > 10 ) {
            /* If we arrive here, PrimaryCCD is not exposing anymore, we can't forward the frame and we can't be aborted neither, thus abort the exposure right now.
            * That issue can be reproduced when clicking the "Set" button on the "Main Control" tab while an exposure is running.
            * Note that the patch in StartExposure returning busy instead of error prevents the flow from coming here, so now it's only a safeguard. */
            IDLog("%s: frame received while not exposing, force-aborting capture\n", __FUNCTION__);
            AbortExposure();
...

So it confirms the CCD should be exposing but indi reports it is not, although I can see the LED is on for the time of exposure and the error message says it's receiving frames in the error message...

I'm lost here, if anyone can shed a light on how frames should be grabbed and what can lead to this error? Is it the capture which doesn't happen at the right time? could it be indi v4l2 driver which do not talk to the device appropriately? Or could it be my mod which is just not right? (BTW, here is the mod I used: spc900 LED mod LXhttp://www.astrosurf.com/djibb/new/modif.htm
3 years 2 months ago #66822

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

  • Posts: 26
  • Thank you received: 1

Replied by Alexandre Chapellon on topic Issue indi_v4l2_ccd: frame received while not exposing, force-aborting capture

File Attachment:

File Name: v4l2_drive...atch.txt
File Size:1 KB
So this is a reply just in case someone find it useful one day, given this topic didn't exactly get much traction...
So eventually managed to get indi_v4l2_ccd working again with the SPC900NC modified for long exposure controlled by LED.

In addition to the modification documented here , I had to also patch the indi_v4l2_ccd in order to avoid reaching a part of the code - which according to the comment should anyway not be reached - where the capture was failing.
The patch is attached bellow, but BEWARE: I'M NO INDI DEVELOPER AND I DONT COMMIT TO ANYTHING, DON'T BLAME IT ON ME if it breaks your device, insult your mother or reelect Georges Bush (<- a geek reference is hidden here!)

Thanks to @geehalel for his help and for updating his driver for recent kernels (5.4)
Last edit: 1 year 7 months ago by Alexandre Chapellon.
3 years 2 months ago #67053
Attachments:

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

Time to create page: 0.448 seconds