How to reset a OAK4-D PRO?

isk

How to reset a OAK4-D PRO?
The camera is new. I try to set it up as a OAK4-D. It didnt work. Now the led is blue (device is ready). How can i do a factory reset?

jakaskerl

isk
If you can access the device via ADB, you can set manually set the IP of the device if that is what failed.
More here.

Thanks,
Jaka

isk

Thank you for the answer. I was able to login via adb and made a factory-reset. So it was possible to configure the device manually (with agentconfd, configuration via browser was not possible). Login via network is possible, but the examples from an extern host is not able to find the device. What can be done?
Is wlan on this device (at least "ip a" returns it)?

jakaskerl

isk

isk Login via network is possible, but the examples from an extern host is not able to find the device. What can be done?

DEPTHAI_DEVICE_NAME_LIST=<ip> env variable should do it.

isk Is wlan on this device (at least "ip a" returns it)?

Will be possible in the future.

Thanks,
Jaka

isk

Thank you for the answer. But it didnt work.
Setting
export DEPTHAI_DEVICE_NAME_LIST=10.10.XX.XX
and running the following script:

#!/usr/local/bin/python3
import depthai as dai
from typing import List
print('Searching for all available devices...\n')
infos: List[dai.DeviceInfo] = dai.DeviceBootloader.getAllAvailableDevices()
if len(infos) == 0:
    print("Couldn't find any available devices.")
    exit(-1)
for info in infos:
    state = str(info.state).split('X_LINK_')[1]
    print(f"Found device '{info.name}', MxId: '{info.mxid}', State: '{state}'")
print(f"\nBooting the first available camera ({infos[0].name})...")
with dai.Device(dai.Pipeline(), infos[0], usb2Mode=False) as device:
    print("Available camera sensors: ", device.getCameraSensorNames())
    calib = device.readCalibration()
    eeprom = calib.getEepromData()
    print(f"Product name: {eeprom.productName}, board name {eeprom.boardName}")

gives the following output:

root@fdc56e272bbd:/depthai-core/examples/python# ./get_cameras.py 
Searching for all available devices...

Found device '10.10.XX.XX', MxId: '2140128077', State: 'GATE_BOOTED'

Booting the first available camera (10.10.XX.XX)...
Traceback (most recent call last):
  File "/depthai-core/examples/python/./get_cameras.py", line 13, in <module>
    with dai.Device(dai.Pipeline(), infos[0], usb2Mode=False) as device:
RuntimeError: No available devices (1 connected, but in use)
root@fdc56e272bbd:/depthai-core/examples/python#

PS: Was trying the oakctl tutorial. got this problem:

root@f2b2a854af50:/# bash -c "$(curl -fsSL https://oakctl-releases.luxonis.com/oakctl-installer.sh)"
  % Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100 19.4M  100 19.4M    0     0  31.6M      0 --:--:-- --:--:-- --:--:-- 31.5M
Downloading data
  [00:00:00] [████████████████████████████████████████] 6.49 MiB/6.49 MiB (0.0s)Os { code: 2, kind: NotFound, message: "No such file or directory" }

jakaskerl

isk
I know it sounds stupid, but have you tried rebooting the device 🙂?
I think the depthai gate service might be down.. Rebooting should restart it.

Thanks,
Jaka

isk

On the camera:

~ # systemctl | grep depthai
  depthai_gate.service                                                                                     loaded    active     running         DepthAI Gate

On the host:

root@81615ac68cfd:/depthai-core/examples/python# python3 Camera/camera_output.py 
Traceback (most recent call last):
  File "/depthai-core/examples/python/Camera/camera_output.py", line 7, in <module>
    with dai.Pipeline() as pipeline:
RuntimeError: No available devices
root@81615ac68cfd:/depthai-core/examples/python# export DEPTHAI_DEVICE_NAME_LIST=10.10.XX.XX
root@81615ac68cfd:/depthai-core/examples/python# python3 Camera/camera_output.py 
Traceback (most recent call last):
  File "/depthai-core/examples/python/Camera/camera_output.py", line 7, in <module>
    with dai.Pipeline() as pipeline:
RuntimeError: No available devices (1 connected, but in use)
root@81615ac68cfd:/depthai-core/examples/python# python3 Camera/camera_output.py 
Traceback (most recent call last):
  File "/depthai-core/examples/python/Camera/camera_output.py", line 7, in <module>
    with dai.Pipeline() as pipeline:
RuntimeError: No available devices (1 connected, but in use)
root@81615ac68cfd:/depthai-core/examples/python#

Nmap returns:

root@81615ac68cfd:/depthai-core/examples/python# nmap -sT -p- 10.10.XX.XX
Starting Nmap 7.80 ( https://nmap.org ) at 2025-02-04 11:25 UTC
Nmap scan report for 10.10.XX.XX
Host is up (0.0029s latency).
Not shown: 65528 closed ports
PORT      STATE SERVICE
22/tcp    open  ssh
3030/tcp  open  arepa-cas
5000/tcp  open  upnp
8080/tcp  open  http-proxy
8765/tcp  open  ultraseek-http
9998/tcp  open  distinct32
11492/tcp open  unknown

Nmap done: 1 IP address (1 host up) scanned in 0.78 seconds
root@81615ac68cfd:/depthai-core/examples/python#

systemctl status depthai_gate returns:

[[0;1;32m●[[0m depthai_gate.service - DepthAI Gate

Loaded: loaded ([]8;;file://kalama/lib/systemd/system/depthai_gate.serviceG/lib/systemd/system/depthai_gate.service[]8;;G; e

nabled; vendor preset: enabled)

Active: [[0;1;32mactive (running)[[0m since Wed 2025-02-05 11:20:20 CET; 1min 11s ago

Main PID: 60277 (python3)

Memory: 19.6M

CGroup: /system.slice/depthai_gate.service

└─[[0;38;5;245m 60277 /usr/bin/python3 /usr/bin/depthai_gate[[0m

Feb 05 11:20:20 kalama depthai_gate[60277]: ERROR:root:[(<class 'depthai_gate.exceptions.PublicError'>, <function main.<locals>.h

andle_exception at 0x7fbac2be20>)]

Feb 05 11:20:20 kalama depthai_gate[60277]: * Serving Flask app 'depthai_gate' (lazy loading)

Feb 05 11:20:20 kalama depthai_gate[60277]: * Environment: production

Feb 05 11:20:20 kalama depthai_gate[60277]: WARNING: This is a development server. Do not use it in a production deployment.

Feb 05 11:20:20 kalama depthai_gate[60277]: Use a production WSGI server instead.

Feb 05 11:20:20 kalama depthai_gate[60277]: * Debug mode: off

Feb 05 11:20:20 kalama depthai_gate[60277]: INFO:werkzeug: * Running on all addresses (0.0.0.0)

Feb 05 11:20:20 kalama depthai_gate[60277]: WARNING: This is a development server. Do not use it in a production deployment.

Feb 05 11:20:20 kalama depthai_gate[60277]: * Running on http://127.0.0.1:11492

Feb 05 11:20:20 kalama depthai_gate[60277]: * Running on http://10.10.33.55:11492 (Press CTRL+C to quit)

jakaskerl

isk
Maybe the netmask and/or gateway were incorrectly set during setup?
check ifconfig and https://rvc4.docs.luxonis.com/hardware/platform/deploy/oak4-deployment-guide/#OAK4%20Getting%20Started-Advanced-Set%20static%20IP

systemctl restart depthai_gate should reset the "one in use" issue.

Thanks,
Jaka

isk

The network mask is correct. Restarting the depthai_gate service gives the error from above.

jakaskerl

Hi @isk
Apparently there is an demo app on there that starts automatically.. Will add to docs..
List apps
oakctl app list

Stop, then disable app
oakctl app stop <id>
oakctl app disable <id>

Thanks,
Jaka

isk

Hi,
thanks. This is the result:
~ # export DEPTHAI_DEVICE_NAME_LIST=10.10.XX.XX
~ # ping 10.10.XX.XX
PING 10.10.XX.XX (10.10.XX.XX): 56 data bytes64 bytes from 10.10.XX.XX: seq=0 ttl=64 time=0.037 ms
--- 10.10.XX.XX ping statistics ---
1 packets transmitted, 1 packets received, 0% packet loss
round-trip min/avg/max = 0.037/0.037/0.037 ms
~ # ssh root@10.10.XX.XX
root@10.10.XX.XX's password:
~ # systemctl status depthai_gate
● depthai_gate.service - DepthAI Gate
Loaded: loaded (file://kalama/lib/systemd/system/depthai_gate.serviceG/lib/systemd/system/depthai_gate.service[]8;;G; enabled; vendor preset: enabled)
Active: mactive (running) since Fri 1970-01-02 00:57:23 CET; 55 years 1 month ago
Main PID: 843 (python3)
Memory: 272.8M
CGroup: /system.slice/depthai_gate.service
├─[[0;38;5;245m 843 /usr/bin/python3 /usr/bin/depthai_gate[[0m
└─[[0;38;5;245m 1477 /tmp/gate_fw_t35vy56i/e6cfebb5814a2095c4d882259b140e294a22a19aa56150a7dc802524725b9f90/depthai-device-rvc4[[0m
Jan 02 00:57:24 kalama depthai_gate[843]: WARNING: This is a development server. Do not use it in a production deployment.
Jan 02 00:57:24 kalama depthai_gate[843]: * Running on http://127.0.0.1:11492
Jan 02 00:57:24 kalama depthai_gate[843]: * Running on http://127.0.0.1:11492 (Press CTRL+C to quit)
Jan 02 00:57:26 kalama depthai_gate[843]: INFO:werkzeug:127.0.0.1 - - [02/Jan/1970 00:57:26] "POST /api/v1/sessions HTTP/1.1" 200 -
Jan 02 00:57:27 kalama depthai_gate[843]: INFO:werkzeug:127.0.0.1 - - [02/Jan/1970 00:57:27] "POST /api/v1/sessions/1/fwp HTTP/1.1" 200 -
Jan 02 00:57:27 kalama depthai_gate[1471]: modprobe: FATAL: Module kmb_flash not found in directory /lib/modules/5.15.144-qki-consolidate-android13-8-g5ce1e4b1db15-dirty
Jan 02 00:57:27 kalama depthai_gate[1473]: modprobe: FATAL: Module kmb_lens not found in directory /lib/modules/5.15.144-qki-consolidate-android13-8-g5ce1e4b1db15-dirty
Jan 02 00:57:27 kalama depthai_gate[1474]: modprobe: FATAL: Module kmb_cam not found in directory /lib/modules/5.15.144-qki-consolidate-android13-8-g5ce1e4b1db15-dirty
Jan 02 00:57:27 kalama depthai_gate[1475]: modprobe: FATAL: Module pwm_keembay not found in directory /lib/modules/5.15.144-qki-consolidate-android13-8-g5ce1e4b1db15-dirty
Jan 02 00:57:27 kalama depthai_gate[843]: INFO:werkzeug:127.0.0.1 - - [02/Jan/1970 00:57:27] "POST /api/v1/sessions/1/start HTTP/1.1" 200 -

root@0961345efa71:~# oakctl app list
Connecting to agent ...
No devices found.
No devices found.
Failed to connect to agent: "\"There are no available connections.\""
root@0961345efa71:~#

jakaskerl

Hi @isk
systemctl --plain --legend=false | grep oak-agent | grep "/" | awk "{print \$NF}" | xargs -I {} systemctl stop {};
Try this. Then reboot. If it still doesn't work, factory reset.
There seems to be a nasty bug...

Thanks,
Jaka

isk

Hi,
Thank you for the answer. But nothing changed. It was also after a factory reset, the device had to be configured manually ( root passwd and ip). systemctl status oak-agent gives:

systemctl status oak-agent

● oak-agent.service - OAK Agent
     Loaded: loaded ([]8;;file://kalama/etc/systemd/system/oak-agent.serviceG/etc/systemd/system/oak-agent.service[]8;;G; enabled; vendor preset: ena
bled)
     Active: activating (auto-restart) (Result: exit-code) since Thu 1970-01-01 00:58:03 UTC; 3s ago
    Process: 1655 ExecStartPre=/usr/bin/test -f /etc/agentconfd/.setup_complete (code=exited, status=0/SUCCESS)
    Process: 1656 ExecStart=/usr/bin/oak-agent /etc/oak-agent/oak-agent.toml [[0;1;31m(code=exited, status=1/FAILURE)[[0m
   Main PID: 1656 (code=exited, status=1/FAILURE)

systemctl status depthai_gate

 [[0;1;32m●[[0m depthai_gate.service - DepthAI Gate
     Loaded: loaded ([]8;;file://kalama/lib/systemd/system/depthai_gate.serviceG/lib/systemd/system/depthai_gate.service[]8;;G; enabled; vendor prese
t: enabled)
     Active: [[0;1;32mactive (running)[[0m since Thu 1970-01-01 00:57:35 UTC; 55 years 1 month ago
   Main PID: 909 (python3)
     Memory: 40.9M
     CGroup: /system.slice/depthai_gate.service
             └─[[0;38;5;245m 909 /usr/bin/python3 /usr/bin/depthai_gate[[0m

Jan 01 00:57:36 kalama depthai_gate[909]: ERROR:root:[(<class 'depthai_gate.exceptions.PublicError'>, <function main.<locals>.handle_exception at 0x7
f9a54be20>)]
Jan 01 00:57:36 kalama depthai_gate[909]:  * Serving Flask app 'depthai_gate' (lazy loading)
Jan 01 00:57:36 kalama depthai_gate[909]:  * Environment: production
Jan 01 00:57:36 kalama depthai_gate[909]:    WARNING: This is a development server. Do not use it in a production deployment.
Jan 01 00:57:36 kalama depthai_gate[909]:    Use a production WSGI server instead.
Jan 01 00:57:36 kalama depthai_gate[909]:  * Debug mode: off
Jan 01 00:57:36 kalama depthai_gate[909]: INFO:werkzeug: * Running on all addresses (0.0.0.0)
Jan 01 00:57:36 kalama depthai_gate[909]:    WARNING: This is a development server. Do not use it in a production deployment.
Jan 01 00:57:36 kalama depthai_gate[909]:  * Running on http://127.0.0.1:11492
Jan 01 00:57:36 kalama depthai_gate[909]:  * Running on http://127.0.0.1:11492 (Press CTRL+C to quit)

Switched to dhcp. Thus the ip address seems not to be correct.

Ales

From this part of your output I suppose that the oakctl commands did not work correctly:

root@0961345efa71:~# oakctl app listConnecting to agent ...
No devices found.
No devices found.
Failed to connect to agent: "\"There are no available connections.\""

To disable+stop the default app you do this before factory-reset:

systemctl --plain --legend=false | grep oak-agent | grep "/" | awk "{print \$NF}" | xargs -I {} systemctl stop {};

Then do the factory reset - after that, then do following oakctl commands from your PC (I'm also pasting expected output):

# 1. Self-update oakctl
oakctl self-update
# Output:
# Already up to date

# 2. Connect to the device
oakctl connect <ip> --with-password
# Output:
# Password is not going to be saved securely!
# Use at your own risk
# > Camera password ********
# Initiating storing connection to 10.0.1.24:9998
# Connection to 10.0.1.24:9998 successfully stored

# 3. Check list of apps
oakctl app list
# Output:
# +---+--------------------------------------+---------------------------+---------+
# | E | Container Id                         | Identifier                | Status  |
# +================================================================================+
# | * | 2d272224-82e5-414d-b7a3-4cdd344fe04d | com.luxonis.default:0.4.1 | running |
# +---+--------------------------------------+---------------------------+---------+

# 4. Disable app
oakctl app disable <ID> # e.g. 2d272224-82e5-414d-b7a3-4cdd344fe04d
# Output:
# Disabling app 2d272224-82e5-414d-b7a3-4cdd344fe04d ...
#   App has been disabled! App will not be started automatically on the next device boot.

# 5. Stop app
oakctl app stop <ID>
# Stopping app 2d272224-82e5-414d-b7a3-4cdd344fe04d ...
#   App has been stopped!

# 6. Reboot
reboot

# 7. Check list of apps again
oakctl app list
# Output:
# +---+--------------------------------------+---------------------------+---------+
# | E | Container Id                         | Identifier                | Status  |
# +================================================================================+
# |   | 2d272224-82e5-414d-b7a3-4cdd344fe04d | com.luxonis.default:0.4.1 | stopped |
# +---+--------------------------------------+---------------------------+---------+

Then - to be sure - try this:

systemctl restart agentconfd 
systemctl restart oak-agent

Now - wait for a while and send us ouput from this two commands if something is still wrong:

oakctl device info
oakctl list

Then you can proceed to run your app.

isk

Thank you for the answer. But after

systemctl --plain --legend=false | grep oak-agent | grep "/" | awk "{print \$NF}" | xargs -I {} systemctl stop {};

and factory reset, nothing changed.
I tried to do a full os recovery (https://rvc4.docs.luxonis.com/software/depthai-components/luxonis-os/). I got confused
with the sentence: "You also need to replace the [DEVICE_ID] with the correct xml file for your device".

Where to replace it?
Where get the code for a OAK4-D PRO AF CR1?