System Health details
xx
Integration version ?
1.24.1
In which region is your Petkit application registered?
No response
Device type
No response
Checklist
Describe the issue
,I updated to the latest version (1.24.1) and set DEFAULT_ALWAYS_ON_STREAM = True. After saving and restarting, my entire Home Assistant keeps restarting (about every 30 seconds) and cannot run normally. My Home Assistant version is 2026.3.4. I think I found the cause. My Home Assistant is running in a virtual machine with 64GB of disk space allocated. After enabling DEFAULT_ALWAYS_ON_STREAM = True, the disk usage kept increasing until it exceeded 96%, which caused continuous restarts and eventually prevented HA from starting.
I’m confused about what is continuously writing data—could it be saving video streams, or is another process consuming the space?
After expanding the VM disk to 128GB, I was finally able to start HA again. However, even then, it kept restarting. I immediately changed DEFAULT_ALWAYS_ON_STREAM back to False, and everything returned to normal with no more restarts。
I confirm that consuming the whep_mirror_url starts a big memory leak for me (like 100mb per sec). If DEFAULT_ALWAYS_ON_STREAM is enabled it starts on integration init。
Reproduction steps
...
Debug logs
Diagnostics dump
No response
System Health details
xx
Integration version ?
1.24.1
In which region is your Petkit application registered?
No response
Device type
No response
Checklist
Describe the issue
,I updated to the latest version (1.24.1) and set DEFAULT_ALWAYS_ON_STREAM = True. After saving and restarting, my entire Home Assistant keeps restarting (about every 30 seconds) and cannot run normally. My Home Assistant version is 2026.3.4. I think I found the cause. My Home Assistant is running in a virtual machine with 64GB of disk space allocated. After enabling DEFAULT_ALWAYS_ON_STREAM = True, the disk usage kept increasing until it exceeded 96%, which caused continuous restarts and eventually prevented HA from starting.
I’m confused about what is continuously writing data—could it be saving video streams, or is another process consuming the space?
After expanding the VM disk to 128GB, I was finally able to start HA again. However, even then, it kept restarting. I immediately changed DEFAULT_ALWAYS_ON_STREAM back to False, and everything returned to normal with no more restarts。
I confirm that consuming the whep_mirror_url starts a big memory leak for me (like 100mb per sec). If DEFAULT_ALWAYS_ON_STREAM is enabled it starts on integration init。
Reproduction steps
...
Debug logs
Diagnostics dump
No response