#5 cameras

1 messages · Page 1 of 1 (latest)

dense bloom
#

What are the specs on the machine where you run HassOS?

grave badge
#

let me fetch that because I never remember and never can't see it within HA

#

Intel N5095
SSD 256GB

#

8GB RAM

dense bloom
#

i dont think it should be taking a long time then

grave badge
#

Beelink Mini S

dense bloom
#

could you try setting enabled: False under each camera and restarting?

grave badge
#

sure

dense bloom
#

wiat

#

wait

grave badge
#

to add some info, regarding the issue for the login page from yesterday

#

I am seeing 401 errors from 1 of the cameraas

#

just one

#

and that's the one having the issue

#

no idea if it might be related

dense bloom
#

also set

logger:
  frigate.record.util: debug
#

then restart

grave badge
#

thanks a lot for the help by the way

grave badge
dense bloom
#

you should get a End sync recordings. when its done

#

in the frigate yaml

#

at the top level

grave badge
#

and disable all cameras as well?

dense bloom
#

yea. just to ensure it focuses on this one task

grave badge
#
Config Error:

Traceback (most recent call last):
  File "/opt/frigate/frigate/api/app.py", line 244, in config_save
    FrigateConfig.parse_raw(new_config)
  File "/opt/frigate/frigate/config.py", line 1641, in parse_raw
    return cls.model_validate(config)
  File "/usr/local/lib/python3.9/dist-packages/pydantic/main.py", line 551, in model_validate
    return cls.__pydantic_validator__.validate_python(
pydantic_core._pydantic_core.ValidationError: 1 validation error for FrigateConfig
logger.`frigate.record.util`
  Extra inputs are not permitted [type=extra_forbidden, input_value='debug', input_type=str]
    For further information visit https://errors.pydantic.dev/2.7/v/extra_forbidden
```when trying to restart
#
logger:
  frigate.record.util: debug
#

is this correct? It's flagged as wrong

#

says property not allowed

#

fixed it using docs:

logger:
  logs:
    frigate.record.util: debug
dense bloom
#

sorry. i was literally looking at the docs and got it wrong

grave badge
#
2024-06-06 14:46:11.197600070  [2024-06-06 15:46:11] frigate.record.util            DEBUG   : Start sync recordings.
2024-06-06 14:46:11.749007393  [2024-06-06 15:46:11] frigate.record.util            INFO    : Deleting 18836 recording DB entries with missing files
2024-06-06 14:46:11.755484758  [2024-06-06 15:46:11] frigate.record.util            DEBUG   : Deleting 0.733775% of recordings DB entries, could be due to configuration error. Aborting...
2024-06-06 14:46:11.755685874  [2024-06-06 15:46:11] frigate.record.util            DEBUG   : End sync recordings.
#

there's the answer

#

something seems to be stopping it

dense bloom
#

i see, yea

#

thats a failsafe to prevent the database from emptying out if someone accidentally starts frigate with the storage improperly mounted

#

the best way to fix it up is to set the days to retain low enough to cleanup these older recordings the normal way

#

but if you are regularly deleting large numbers of recordings manually, this is going to keep happening

#

frigate already checks the disk usage every 5 minutes and deletes recordings if the disk could potentially fill within 1 hour of recordings

grave badge
#

I set up an alert myself from HA when it's going over 70% because once it got completely full, and then HA started malfunctioning. :S I guess HA needs some more room than that 1h maybe

#

I can use the retain days to do the manual cleanup if it's easier on Frigate tho

dense bloom
#

i mean, frigate will be just fine if the files dissappear

grave badge
#

I wish I get to a point where I don't need it, but tbh I find it very confusing to find out the causes of recording, so the best way I've found is clearing, changing setting, and then start again

dense bloom
#

the storage page just wont be accurate

grave badge
#

I get it

#

I like it to be accurate as I check it regularly, so I'll get by using the retain days

#

I'm only using motion mode tho, and 2 days. Does retain 0 work?

dense bloom
#

yea

#

that will only retain stuff that is during the time an object is being tracked

grave badge
#

cool. I have to check about events as I've only started using them. I'd like to set a retain limit for them too, as I fear they might end up fillling my storage themselves anyway

dense bloom
#

the default is 10 days

grave badge
#

and there's no "Select all" for them yet in the UI (feat request)

#

ah cool

dense bloom
#

and your storage should never fill

grave badge
#

that's enough I guess

dense bloom
#

frigate wont let it

#

unless you have something outside of frigate fill the disk

#

or if parts of frigate error out

grave badge
#

I understand, but that 1h room it protects maybe it's not enough for HA to perform tasks

dense bloom
#

i doubt it

#

for 5 cameras, thats going to be gigs

#

how big are the 10s segments for each camera?

grave badge
#

some backups maybe... they're sometimes quite large

dense bloom
#

it assumes 100% of them are retained

grave badge
dense bloom
#

so you can do (total of 10s segment size for all cameras)*6*60

grave badge
#

can't find the videos in the clips folder

dense bloom
#

recordings folder

grave badge
#

7.55MB each video. I guess that's what you mean.

#

the suggestionn from nick about making a partition for Frigate would be ideal for me, but I just saw it's not supported in HA. It'd need to be network storage. Don't have any of it for now.

#

anyway. Thanks a lot for the help. We can step out the thread now if you want.

#

If you can pick up the 401 error later when you can please

dense bloom
#

with 5 cameras having 7.55mb segments, Frigate checks every 5 minutes to see if the system has less than 13.5GB of free space on the drive. If it does, it deletes recordings until there is 27GB of free space.

#

so unless you have something creating ~13GB in under 5 minutes on the drive, it shouldnt be a problem

#

if thats not enough, you will need to use retain settings to ensure it stays within the limits that your system can handle

grave badge
#

cool. I just set retain to 0 and indeed it cleared most. However, now I set 0 for events to clear them out as well, but there's still 10GB used by Frigate somehow. I can see events in the UI. Is the 0 days keeping something?