372935 – After 1 hour of imaging kstars/ekos became unresponsive

Bug 372935 - After 1 hour of imaging kstars/ekos became unresponsive

Summary: After 1 hour of imaging kstars/ekos became unresponsive

Status:	RESOLVED WORKSFORME

Alias:	None

Product:	kstars
Classification:	Applications
Component:	general (show other bugs)
Version:	git
Platform:	Other Linux

Importance:	NOR normal
Target Milestone:	---
Assignee:	Jasem Mutlaq

URL:
Keywords:

Depends on:
Blocks:

Reported:	2016-11-25 22:53 UTC by Hans Lambermont
Modified:	2018-03-24 12:59 UTC (History)
CC List:	0 users

See Also:
Latest Commit:
Version Fixed In:

Attachments
No memory impact (13.65 KB, image/png) 2016-12-04 09:40 UTC, Hans Lambermont	Details
View All Add an attachment

Note You need to log in before you can comment on or make changes to this bug.

Description Hans Lambermont 2016-11-25 22:53:32 UTC

After imaging about 5*10 min frames kstars became very sluggish up to the point where it was not responding to anything anymore.
Gnome made the program dark as it's not responding to anything.
Click close button. Window is not responding, force quit ? Yes
Here's a backtrace in the hope that it's helpful


#0  0x00007fffec63e730 in xcb_setup_vendor_end () from /usr/lib/x86_64-linux-gnu/libxcb.so.1
#1  0x00007fffec63e759 in xcb_setup_pixmap_formats () from /usr/lib/x86_64-linux-gnu/libxcb.so.1
#2  0x00007fffe423f257 in ?? () from /usr/lib/x86_64-linux-gnu/libxcb-image.so.0
#3  0x00007fffe424037e in xcb_image_native () from /usr/lib/x86_64-linux-gnu/libxcb-image.so.0
#4  0x00007fffe424054f in xcb_image_shm_put () from /usr/lib/x86_64-linux-gnu/libxcb-image.so.0
#5  0x00007fffe4eda839 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5XcbQpa.so.5
#6  0x00007fffe4edac49 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5XcbQpa.so.5
#7  0x00007ffff46ad095 in QBackingStore::flush(QRegion const&, QWindow*, QPoint const&) () from /usr/lib/x86_64-linux-gnu/libQt5Gui.so.5
#8  0x00007ffff4ad9a7a in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#9  0x00007ffff4ada796 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#10 0x00007ffff4adc8bf in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#11 0x00007ffff4adca8c in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#12 0x00007ffff4afac1f in QWidgetPrivate::syncBackingStore() () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#13 0x00007ffff4b11d88 in QWidget::event(QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#14 0x00007ffff4c27d7b in QMainWindow::event(QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#15 0x00007ffff7033147 in KMainWindow::event(QEvent*) () from /usr/lib/x86_64-linux-gnu/libKF5XmlGui.so.5
#16 0x00007ffff706b4e5 in KXmlGuiWindow::event(QEvent*) () from /usr/lib/x86_64-linux-gnu/libKF5XmlGui.so.5
#17 0x00007ffff4acf05c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#18 0x00007ffff4ad4516 in QApplication::notify(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#19 0x00007ffff3cf338b in QCoreApplication::notifyInternal(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#20 0x00007ffff4add1bf in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#21 0x00007ffff4adde5d in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#22 0x00007ffff4afd612 in QWidget::repaint(QRect const&) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#23 0x00007ffff4c6e89e in QStatusBar::hideOrShow() () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#24 0x00007ffff4c6e94a in QStatusBar::showMessage(QString const&, int) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#25 0x00000000005e2b85 in INDI_D::updateMessageLog (this=0x3b7efe0, idv=<optimized out>, messageID=<optimized out>)
    at /build/kstars-bleeding-qSmJ00/kstars-bleeding-16.10+201611202010~ubuntu16.04.1/kstars/indi/indidevice.cpp:284
#26 0x00007ffff3d22c01 in QObject::event(QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#27 0x00007ffff4b11cdb in QWidget::event(QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#28 0x00007ffff4acf05c in QApplicationPrivate::notify_helper(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#29 0x00007ffff4ad4516 in QApplication::notify(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Widgets.so.5
#30 0x00007ffff3cf338b in QCoreApplication::notifyInternal(QObject*, QEvent*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#31 0x00007ffff3cf5786 in QCoreApplicationPrivate::sendPostedEvents(QObject*, int, QThreadData*) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#32 0x00007ffff3d493c3 in ?? () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#33 0x00007fffef0fc1a7 in g_main_context_dispatch () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#34 0x00007fffef0fc400 in ?? () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#35 0x00007fffef0fc4ac in g_main_context_iteration () from /lib/x86_64-linux-gnu/libglib-2.0.so.0
#36 0x00007ffff3d497cf in QEventDispatcherGlib::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#37 0x00007ffff3cf0b4a in QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#38 0x00007ffff3cf8bec in QCoreApplication::exec() () from /usr/lib/x86_64-linux-gnu/libQt5Core.so.5
#39 0x0000000000483c23 in main (argc=1, argv=<optimized out>)
    at /build/kstars-bleeding-qSmJ00/kstars-bleeding-16.10+201611202010~ubuntu16.04.1/kstars/main.cpp:277

kstars-bleeding      5:16.10+201611202010~ubuntu16.04.1
Ubuntu 16.04.1 LTS

Comment 1 Jasem Mutlaq 2016-11-26 00:07:53 UTC

Did you check memory consumption? Did you have INDI debug enabled to client? i.e. debug message getting printed in the INDI control panel?

Comment 2 Hans Lambermont 2016-11-26 21:35:44 UTC

(In reply to Jasem Mutlaq from comment #1)
> Did you check memory consumption?

Yes, kstars memory usage was normal, a few hundred MiB.

> Did you have INDI debug enabled to client?

Not anymore :-/ I had when I started the evening, but I did not re-enable debugging after an earlier kstars crash where I also restarted indiserver. Afaik INDI debug settings are not auto-restored.

I think I need to write a script that sets these extra wanted settings. Would help with other parts that do not self-restore like asi_ccd's raw-16-bit and USB Bandwidth.

> i.e. debug message getting printed in the INDI control panel?

I did not see any (otherwise I'd added them to this bug report).

Sorry. Not much to work with.

Comment 3 Hans Lambermont 2016-12-04 00:04:50 UTC

We can now narrow the issue down a bit further :

This happens when PHD2 sends a constant stream of guide commands to the mount via INDI. I have mount debugging enabled in INDI control panel (but not to the 'Scope Verbose level).
EKOS then quickly becomes completely unresponsive.
The INDI control panel also stopped showing updates of the mount commands and its window is unresponsive as well.
The Kstars status line at the bottom of the window does show the constant flow of guide commands.

The reason I do not always see this issue is that for me PHD2 only sends guide commands to the mount when the tip/tilt plate reaches a maximum tilt. PHD2 then instructs the mount to center the star again in the tip/tilt plate. I let PHD2 talk directly to the tip/tilt plate so its guide commands do not show up in INDI.

Here's just 3 seconds of log lines. It's not that many actually.

2016-12-04T00:21:57.036 - DEBG - Celestron GPS :  "Moving toward West. "
2016-12-04T00:21:57.132 - DEBG - Celestron GPS :  " END-OF-TIMER "
2016-12-04T00:21:57.231 - DEBG - Celestron GPS :  "   USE_PULSE_CMD = 0 "
2016-12-04T00:21:57.328 - DEBG - Celestron GPS :  "   GUIDE_DIRECTION = 2 "
2016-12-04T00:21:57.454 - DEBG - Celestron GPS :  "   CALL_DIRECTION = 2 "
2016-12-04T00:21:57.551 - DEBG - Celestron GPS :  "CMD (50 02 10 24 00 00 00 00) "
2016-12-04T00:21:57.657 - DEBG - Celestron GPS :  "RES (#) "
2016-12-04T00:21:57.753 - DEBG - Celestron GPS :  "Movement toward West halted. "
2016-12-04T00:21:57.848 - DEBG - Celestron GPS :  " CALL SendPulseStatusCmd "
2016-12-04T00:21:57.944 - DEBG - Celestron GPS :  " PULSE STATUS REQUEST: (FD:03, DIR:02) "
2016-12-04T00:21:58.040 - DEBG - Celestron GPS :  "   COMMAND (50 03 10 27 00 00 00 01) "
2016-12-04T00:21:58.138 - DEBG - Celestron GPS :  "   ISSUING COMMAND "
2016-12-04T00:21:58.234 - DEBG - Celestron GPS :  "   WAITING FOR REPLY "
2016-12-04T00:21:58.330 - DEBG - Celestron GPS :  "    NBYTES = 2, RESPONSE = 01 23 "
2016-12-04T00:21:58.427 - DEBG - Celestron GPS :  " PULSE STILL IN PROGRESS, POSSIBLE MOUNT JAM. "
2016-12-04T00:21:58.522 - DEBG - Celestron GPS :  "GUIDE CMD COMPLETED "
2016-12-04T00:21:58.625 - DEBG - Celestron GPS :  "GUIDE CMD: W 84 ms "
2016-12-04T00:21:58.721 - DEBG - Celestron GPS :  " PULSEGUIDE IS OFF "
2016-12-04T00:21:58.817 - DEBG - Celestron GPS :  " PULSEGUIDE OFF, STARTING MOTION "
2016-12-04T00:21:58.913 - DEBG - Celestron GPS :  "CMD (50 02 10 24 01 00 00 00) "
2016-12-04T00:21:59.009 - DEBG - Celestron GPS :  "RES (#) "
2016-12-04T00:21:59.106 - DEBG - Celestron GPS :  "Moving toward West. "
2016-12-04T00:21:59.205 - DEBG - Celestron GPS :  "GUIDE CMD: N 132 ms "
2016-12-04T00:21:59.300 - DEBG - Celestron GPS :  " PULSEGUIDE IS OFF "
2016-12-04T00:21:59.396 - DEBG - Celestron GPS :  " PULSEGUIDE OFF, STARTING MOTION "
2016-12-04T00:21:59.493 - DEBG - Celestron GPS :  "CMD (50 02 11 24 01 00 00 00) "
2016-12-04T00:21:59.589 - DEBG - Celestron GPS :  "RES (#) "
2016-12-04T00:21:59.686 - DEBG - Celestron GPS :  "Moving toward North. "
2016-12-04T00:21:59.783 - DEBG - Celestron GPS :  " END-OF-TIMER "
2016-12-04T00:21:59.918 - DEBG - Celestron GPS :  "   USE_PULSE_CMD = 0 "

I hope this helps.

Comment 4 Jasem Mutlaq 2016-12-04 07:00:23 UTC

So same question again about memory consumption?

Comment 5 Hans Lambermont 2016-12-04 09:40:30 UTC

Created attachment 102611 [details]
No memory impact

Comment 6 Jasem Mutlaq 2016-12-04 09:54:44 UTC

Try to turn off "INDI Messages in Status Bar" and see if that makes any difference. It is found in the INDI Settings in KStars Settings.

Comment 7 Hans Lambermont 2016-12-04 23:01:09 UTC

I disabled "INDI Messages in Status Bar" and it has an effect, but very little.
Sorry, not much to work with.

Comment 8 Jasem Mutlaq 2017-09-19 13:11:42 UTC

Can you reproduce this in 2.8.3?

Comment 9 Jasem Mutlaq 2018-03-24 12:59:45 UTC

If you still experience this in 2.9.3, please reopen.