Internal hardware watchdog time out??

AMcIntosh

This morning I had a complaint from a user saying that his WiFi dropped out whilst on a Teams call.

Checking the Nebula dashboard I could see that there was 1min where the connection dropped. On further investigation of the even log, I could see that at the very time of the drop, the AP (NWA50AX Pro) had reported "device is rebooted by internal hardware watchdog timeout"

Any idea what this is or what causes it??

Zyxel_Melen

Hi @AMcIntosh

Please help to collect diagnostic file and send it with us so we can help to clarify the root cause.

https://community.zyxel.com/en/discussion/17314/how-to-collect-the-nebula-ap-diagnostic-info-by-gui

AMcIntosh

@Zyxel_Melen

I have sent the Diagnostic file via message (rather than post here)

Zyxel_Melen

Hi @AMcIntosh

After checking, we need your help to enable Zyxel support access for us to check if this issue happens before and monitor from event log.

https://community.zyxel.com/en/discussion/14234/nebula-how-to-turn-on-zyxel-support-access

Additionally, device is rebooted by internal hardware watchdog timeout is a protect feature that prevent the AP becomes unresponsive or encounters a deadlock.

rosariodellib

Hi @AMcIntosh and @ZyXEL _Melen,

I am experiencing the exact same issue on two NWA50AX Pro units.Log entry: Device is rebooted by internal hardware watchdog timeout!

To troubleshoot this effectively and rule out any Cloud synchronization issues, I factory reset both APs and switched them from Nebula Cloud to Standalone mode. Unfortunately, the random reboots persist even in Standalone mode with a very conservative configuration.

** Troubleshooting steps already taken:**

1. Hardware: Replaced the PoE+ switch (New Zyxel GS1005HP)
2. Config: Reduced 5GHz width to 80MHz, disabled Smart Mesh/DCS, and set Fixed Multicast Rate to offload the CPU.
3. Management: Switched to Standalone to isolate the issue.

It seems like a firmware/driver instability under constant load (I have IP cameras streaming).

Two questions for support:

1. Is there a stable or Beta firmware available that addresses this specific Watchdog/CPU freeze issue? I have tried every possible configuration change without success.
2. To provide you with the necessary logs/access for checking, do I need to onboard them back to Nebula Cloud, or can I simply send you the diag-info file downloaded from the Standalone GUI?

Looking forward to a solution.