Getting series of errors when scrubbing the timeline extensively in Safari, ultimately causing failure to convert /tmp/cache files to /recordings #14243
Unanswered
NathanVanCleave
asked this question in
General Support
Replies: 1 comment 3 replies
-
this seems like some type of disk issue where the disk activity causes issues, we can specifically see the error |
Beta Was this translation helpful? Give feedback.
3 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Checklist
Describe the problem you are having
This appears to only happen in Safari on desktop and mobile. I have not been able to reproduce it in Chrome on desktop.
If I scrub through the timeline quickly, either moving forward and backwards or past the timeline extents which causes it to move faster, I get several this error
sqlite3.OperationalError: unable to open database file
followed immediately by this errorpeewee.OperationalError: unable to open database file
thenUnable to convert /tmp/cache
... for most cameras, which results in further errors for each camera occurrence fromfrigate.record.maintainer
. About 1 second later this series of errors repeats and continues quickly for a while.Sometimes Frigate seems to recover and the errors go away after a few minutes, other times it continues creating thousands of logs a minute and stops recording until I restart Frigate. The first time it happened, I didn't notice it until about an hour later, during which time no recordings were made (the reason I think this is worth creating an issue over).
Watching CPU usage by the docker container via Glances I can see it bounce up and down (well above and back to normal usage) with every update (every 2s). Typically, this happens for a few minutes before settling back down to normal, and after another few minutes the main errors disappear. However, I continue getting
Unable to keep up with recording segments in cache for <camera>. Keeping the 6 most recent segments out of 7 and discarding the rest...
and frigate shows theNo frames have been received, check error logs
for each camera in the UI, though everything else including viewing events and previous recordings is still functional. I did not include the logs from this below because there are tens of thousands of lines in between the initial errors and when it gets to this state. Let me know and I can provide another snippet or the full logs.Steps to reproduce
Version
0.14.1-f4f3cfa
In which browser(s) are you experiencing the issue with?
Safari Version 18.0 (20619.1.26.31.6), iOS 18.1 Public Beta 2
Frigate config file
docker-compose file or Docker CLI command
Relevant Frigate log output
Relevant go2rtc log output
# Frigate is too unresponsive to load go2rtc logs.
Operating system
Debian
Install method
Docker Compose
Network connection
Wired
Camera make and model
Mostly Amcrests (IP5M-T1277EW-AI and AD-410), as well as a few Eufy Solos (indoor and outdoor) and a few Wyze V3s
Screenshots of the Frigate UI's System metrics pages
Any other information that may be helpful
I have Home Assistant automations to alert me of a camera being offline by watching for the overall detection FPS sensor as well as each camera's FPS sensor in HA to drop. However, when this happens the FPS sensors remain constant causing the automation to not run and I don't notice the issue right away.
I have noticed the HA Intel-Qsv gpu load sensor stops updating while the initial errors are happening and once I start seeing the
Unable to keep up with recording segments
the gpu sensor goes back to normal.Beta Was this translation helpful? Give feedback.
All reactions