gst pipeline closes (-5) without waiting for hardware trigger

[lc-spxl]

Describe the bug
gst pipeline fails with error

[14:43:29.379] 🅸 device> [GvDevice::new] Device endianness = big
[14:43:29.379] 🅸 device> [GvDevice::new] Packet resend     = yes
[14:43:29.379] 🅸 device> [GvDevice::new] Write memory      = yes
[14:43:29.379] 🅸 device> [GvDevice::new] Legacy endianness handling = yes
[14:43:29.604] 🅸 device> [GvDevice::create_stream] Number of stream channels = 1
[14:43:29.625] 🅸 device> [GvDevice::auto_packet_size] Current packet size check successfull (1400 bytes)
[14:43:29.633] 🅸 device> [GvStream::stream_new] Packet size = 1400 byte(s)
[14:43:29.637] 🅸 stream> [GvStream::stream_new] Destination stream port = 55890
[14:43:29.637] 🅸 stream> [GvStream::stream_new] Source stream port = 49153
[14:43:29.671] 🅸 stream> [GvStream::loop] Packet socket method
[14:43:30.379] 🅳 device> [GvDevice::Heartbeat] Ack value = 2
[14:43:31.381] 🅳 device> [GvDevice::Heartbeat] Ack value = 2

ERROR: from element /GstPipeline:pipeline0/GstAravis:aravis0: Internal data stream error.
Additional debug info:
../libs/gst/base/gstbasesrc.c(3127): gst_base_src_loop (): /GstPipeline:pipeline0/GstAravis:aravis0:
streaming stopped, reason error (-5)

[14:43:31.811] 🅸 stream>     bins    ;frame_retent; packet_time;inter_packet
                                    0;           0;           0;           0
                         -------------
                         >=      2000;           0;           0;           0
                         <          0;           0;           0;           0
                         min         :         n/a;         n/a;         n/a
                         max         :         n/a;         n/a;         n/a
                         last max at :           0;           0;           0
                         counter     :           0:           0:           0

To Reproduce

The pipeline is the following (I also tested it with a Hikrobot-MV-CA032-10GC)

aravissrc camera-name="FLIR-Blackfly S BFS-PGE-23S3C-20136026" features="TriggerSelector=FrameBurstStart TriggerSource=Line0 TriggerMode=On" ! video/x-raw, format=RGB,width=960,height=600,framerate=15/1 ! jpegenc idct-method=2 quality=100 ! multipartmux boundary="myboundary" ! tcpserversink sync=false host=0.0.0.0 port=8000

However, removing the features property it run properly

aravissrc camera-name="Hikrobot-MV-CA032-10GC-02K32049763" ! video/x-raw,framerate=15/1,fps=15/1,width=960,height=600,format=RGB ! jpegenc idct-method=2 quality=100 ! multipartmux boundary="myboundary" ! tcpserversink sync=false host=0.0.0.0 port=8000

Expected behavior

Pipeline should wait until the trigger is activated

Camera description:

Camera 1

• HIKROBOT
• MV-CA032-10GC
• eth

Camera 2

• HIKROBOT
• MV-CA032-10GC
• eth

Platform description:

• Aravis version: built from commit 97fcd70
• OS: Ubuntu 22.04
• Hardware x86_64

Additional context

I would point to #528 but I have no clarity on how the associated PR is solving the issue

[lc-spxl]

From the logs seems I am hitting the timeout of 2s as defined here

aravis/gst/gstaravis.c

Line 311 in 43b3ea4

GST_DEBUG_OBJECT (gst_aravis, "Buffer timeout = %" G_GUINT64_FORMAT " µs", gst_aravis->buffer_timeout_us);

0:00:00.411558950 11896 0x55ccb7514de0 DEBUG              aravissrc gstaravis.c:298:gst_aravis_set_caps:<aravis0> Buffer timeout = 2000000 µs

This then fails on timeout

aravis/gst/gstaravis.c

Lines 555 to 561 in 43b3ea4

	do {
	if (arv_buffer) arv_stream_push_buffer (gst_aravis->stream, arv_buffer);
	arv_buffer = arv_stream_timeout_pop_buffer (gst_aravis->stream, gst_aravis->buffer_timeout_us);
	} while (arv_buffer != NULL && arv_buffer_get_status (arv_buffer) != ARV_BUFFER_STATUS_SUCCESS);
	if (arv_buffer == NULL)
	goto error;

I would say #686 is not addressing it or maybe not solving it directly.

If the trigger is set I would skip the timeout completely, what do you think ?

For reference, this repository implements Flir camera plugin based on spinnaker SDK

https://github.com/thompd27/gst-plugin-spinnaker/blob/master/src/gstspinnaker.c#L1053

[EmmanuelP]

Closing as duplicate of #528

[EmmanuelP]

Thanks for the report.