-
Notifications
You must be signed in to change notification settings - Fork 1.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Edit: vcos_abort error and raspistill hangs #1574
Comments
Duplicate of https://www.raspberrypi.org/forums/viewtopic.php?f=43&t=310668 (now locked) Raspistill command line used is apparently |
Sorry for duplicating, I did not know it's the same. I did additional steps. First of all, I've rolled the firmware back to the 5.10.25 version, as it works on a different hardware. ` ` Unfortunately, the issue stays the same. Anything else to check ? thanks.
`913359.390: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 1913421.775: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2093336.683: assert( temps->last_tone_mapping_params_valid ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_agc.c::isp_tuner_brcm_agc_switch_mode line 2745 rev f66a788 2093385.633: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2093447.848: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2273368.909: assert( temps->last_tone_mapping_params_valid ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_agc.c::isp_tuner_brcm_agc_switch_mode line 2745 rev f66a788 2273417.337: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2273481.815: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2453464.997: assert( temps->last_tone_mapping_params_valid ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_agc.c::isp_tuner_brcm_agc_switch_mode line 2745 rev f66a788 2453514.938: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2453577.272: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2453633.656: assert( q->guardword ==0xa55a5aa5 ) failed; ../../../../../vcfw/rtos/common/rtos_common_malloc.c::rtos_pool_aligned_free line 194 rev f66a788 |
Please find below additional log info, collected after reboot and no raspistil in crontab. login as: pi The programs included with the Debian GNU/Linux system are free software; Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent [dozen of messages like this:] No exceptions found pi@AllSkyCamPi:~ $ pi@AllSkyCamPi:~ $ raspistill -v "raspistill" Camera App (commit 4a0a19b88b43 Tainted) Camera Name imx477 GPS output Disabled Quality 85, Raw no Preview Yes, Full screen Yes |
Hey Frank, I am having the exact same problem. It does seem to be related to the latest updates and upgrades as my raspistill code was stable for a year prior to last month. A couple things that I have followed that have helped (but not cured... it still hangs but not as often) are: Increasing GPU memory: https://www.raspberrypi.org/documentation/configuration/config-txt/memory.md Like I said, this has improved things but not completely resolved the issue. |
thanks for replying. I already have gpu_mem = 256 |
Here again. please, any advice would be appreciated. pi@AllSkyCamPi:~/sky $ sudo vcdbg log assert 1648829.099: assert( temps->last_tone_mapping_params_valid ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_agc.c::isp_tuner_brcm_agc_switch_mode line 2745 rev f66a788 1648880.463: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 1648941.105: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 1828910.726: assert( temps->last_tone_mapping_params_valid ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_agc.c::isp_tuner_brcm_agc_switch_mode line 2745 rev f66a788 1828958.006: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 1829022.623: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2007940.183: assert( temps->last_tone_mapping_params_valid ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_agc.c::isp_tuner_brcm_agc_switch_mode line 2745 rev f66a788 2007987.136: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2008049.588: assert( pass == ISP_TUNER_PASS_NONE ) failed; ../../../../../middleware/ISP/tuner/isp_tuner_brcm_focus.c::master_focus_post_isp_update line 1912 rev f66a788 2008149.450: assert( *p==0xa55a5aa5 ) failed; ../../../../../vcfw/rtos/common/rtos_common_malloc.c::rtos_pool_aligned_free line 205 rev f66a788 2010143.383: assert( Timed out waiting to destroy ril.camera ) failed; ../../../../../interface/mmal/vc/ril/mmal_ril.c::fail_destroy line 1471 rev f66a788 pi@AllSkyCamPi: pi@AllSkyCamPi:~/sky $ sudo vcdbg log ex |
gpu_mem = 1024 is much much too big,. Maximum you will ever need on a running (F)KMS Pi4 is 256, probably less. Not related to the issue I suspect, but a waste of memory, and large values of gpu_mem are known to cause other issues. |
I agree...
I had to reboot the Pi and now it is no longer reachable... As it is in a
remote location, I need to figure out how to go there and restart it.
To me is still a mystery what vcos_abort means and how to fix it.
Hope someone here may help..
Il giorno gio 29 apr 2021 alle ore 11:18 James Hughes <
***@***.***> ha scritto:
… gpu_mem = 1024 is much much too big,. Maximum you will ever need on a
running (F)KMS Pi4 is 256, probably less. Not related to the issue I
suspect, but a waste of memory, and large values of gpu_mem are known to
cause other issues.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#1574 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AKWXGV4HZPZSLOPZQMT5PQDTLEP65ANCNFSM43XEI2UQ>
.
|
You've already had that answered on the forum thread (and this why it is frowned upon to start the same conversation in two places.
Asserts from I've tested the latest rpi-update release on an 8GB Pi and HQ camera with your command line. No issues observed so far. Now running on a tight loop.... |
6by9, thanks for the answer, but the info you provided are not helping me to overcome the issue.
why ? A new Raspberry Pi4 with 4GB and Raspbian Lite is just supposed to collect an image with a raspistill every 3 minutes. The task itself should take a few seconds to get done. It worsk for some hours and, all of the sudden... vcos_abort.
By who ? each 3 minutes raspistill is supposed to generate an image with a different name (with timestamp). Raspistill hangs, but after 3 minutes a new instance of raspistill is launched, as it is in crontab job. As I am not an expert, if you were me, what would you do ? Anything you would check, try or change ? Should I return the hardware (in this case, camera or Pi board) ? |
My test rig has now been running for an hour with no issues (no asserts logged other than the early You're using raspistill with the preview disabled. Is the camera connection actually reliable? Check the preview image is stable. You are NOT running the latest firmware.
Both those asserts were removed from the firmware source tree on 25th March, and have been in all rpi-update releases since 26th March (Hexxeh/rpi-firmware@c0cf93a) |
Thank you 6by9. Initially, I had the problem once per day. It was the release 5.17.10 pi@AllSkyCamPi:~ $ vcgencmd version pi@AllSkyCamPi:~ $ raspistill -v "raspistill" Camera App (commit 4a0a19b88b43 Tainted) Camera Name imx477 GPS output Disabled Quality 85, Raw no Preview Yes, Full screen Yes |
Hi, I did a new, fresh installation on a completely different Pi4 board with same 4GB RAM and Raspbian Lite and a different Pi HQ Camera. I got the Pi4 frozen after 1 day. Looking at sudo vcdbg log assert, I have found the same message:
Please, note that there was no vcos_abort, but just the above message. So I decided to update the firmware to the last version. Now, the firmware is the following:
looking at the logs, I can see this messages:
Not sure what this means. Any insight ? I will let the Pi4 working and let's see if it hangs again. |
I had this problem also with intervals that became shorter and shorter. At last it stopped working within just a few minutes. |
I just run sudo apt update and sudo apd update. I got the version
According to raspberry.org, this is last stable version. Wondering if I should take the risk to install something that is not considered stable... |
and then the Pi was no longer reachble.
The discussion is here |
Was there a definitive solution to this problem? I have slightly different hardware (https://electreeks.de/shop/raspberry-pi-kamera-nachtsicht-mit-infrarot-sperrfilter-175-grad-weitwinkel-objektiv/) and I'm using uv4l to stream motion jpg or the raw h264 stream. After some time, I'm also suffering from the I'm already on the latest firmware:
|
Hi,
I have a Pi4+ with 4GB RAM and Raspbian Lite. This morning I have updated the firmware to the last version.
Now I have:
uname -a Linux AllSkyCamPi 5.10.32-v7l+ #1413 SMP Tue Apr 27 14:43:13 BST 2021 armv7l GNU/Linux
all of a sudden, my (every-3-minutes job in crontab) raspistill hangs. It works for around 30 minutes and then.. it hangs.
EDIT: it is no longer every 30 minutes. Let's say "frequently".
In order to investigate, I have added start_debug=1 into config.txt.
checking with sudo vcdbg log assert, I get:
04876.873: assert( timeout != 0 ) failed; ../../../../../vcfw/drivers/chip/vciv/2708/clock_2711.c::clock_wait_busy line 2014 rev a48d332 vcdbg_ctx_get_dump_stack: dump_stack failed
checking with sudo vcdbg log msg, I get:
425.381: camsubs: Camera found OK 089350.841: imx477_open: read id 1143 succes 0! 116519.256: imx477_open: read id 1143 succes 0! 226123.930: imx477_open: read id 1143 succes 0! 296520.729: imx477_open: read id 1143 succes 0! 476564.772: imx477_open: read id 1143 succes 0! 656604.678: imx477_open: read id 1143 succes 0! 836599.585: imx477_open: read id 1143 succes 0! 1016624.792: imx477_open: read id 1143 succes 0! 1196640.645: imx477_open: read id 1143 succes 0! 1376645.807: imx477_open: read id 1143 succes 0! 1382107.500: vcos_abort: Halting 1384096.620: mmal: fail_destroy: Timed out waiting to destroy ril.camera
after this error, if I launch raspistill from command line, it does not respond at all. Ctrl-C is not working.
I only have to close the SSH session, open a new SSH session and reboot.
After a while, it hangs again.
of course, in my ps -ef I get a lot of raspistill (as they are in crontab every 3 minutes)
After reboot, when it hangs, I get additional messages in log:
`004876.873: assert( timeout != 0 ) failed; ../../../../../vcfw/drivers/chip/vciv/2708/clock_2711.c::clock_wait_busy line 2014 rev a48d332
vcdbg_ctx_get_dump_stack: dump_stack failed
1382107.453: assert( *p==0xa55a5aa5 ) failed; ../../../../../vcfw/rtos/common/rtos_common_malloc.c::rtos_pool_aligned_free line 205 rev a48d332
vcdbg_ctx_get_dump_stack: dump_stack failed
1384096.690: assert( Timed out waiting to destroy ril.camera ) failed; ../../../../../interface/mmal/vc/ril/mmal_ril.c::fail_destroy line 1471 rev a48d332
vcdbg_ctx_get_dump_stack: dump_stack failed
`
last reboot I had this one:
`debug_sym: vc_mem_copy: Unable to open '/dev/fb0': No such file or directory(2)
004977.204: assert( timeout != 0 ) failed; ../../../../../vcfw/drivers/chip/vciv/2708/clock_2711.c::clock_wait_busy line 2014 rev a48d332
vcdbg_ctx_get_dump_stack: dump_stack failed
2115614.742: assert( object->magic == OBJECT_MAGIC ) failed; ../../../../../helpers/vc_pool/vc_pool.c::vc_pool_offsetof line 461 rev a48d332
vcdbg_ctx_get_dump_stack: dump_stack failed`
If I run
sudo vcgencmd get_config int
it completely hangs. Ctrl-C does not work.
p.s. I have a different hardware (Pi4 and HQ Camera) but with 5.10.25-v7l+ #1408 SMP and NO issues at all
An idea could be going back to the 5.10.25 and check if it still happens..
Please help me..
thanks
frank
The text was updated successfully, but these errors were encountered: