Browser error message: Page unresponsive.

[306936930]

Current Behavior

Browser error message: Page unresponsive.

In my case, I used 10 meters to display real-time data. Initially, it displayed correctly, but after about 30 minutes, the meter data stopped updating. Clicking the page with the mouse displayed the "Page unresponsive" message. Refreshing the tab had no effect; I had to open a new tab for it to display correctly again. I believe this is a bug. I checked my computer and browser settings, avoiding putting the page in power-saving mode, but that didn't work.

Environment: Edge browser, Node-Red 5.0, Dashboard2 1.30.2

Expected Behavior

No response

Steps To Reproduce

No response

Environment

• Dashboard version:1.30.2
• Node-RED version:5.0.0
• Node.js version:v24.16.0
• npm version:
• Platform/OS:win11
• Browser: Edge 149.0.4022.98

Have you provided an initial effort estimate for this issue?

I can not provide an initial effort estimate

[colinl]

Are node-red and the browser running on the same PC?

I had to open a new tab for it to display correctly again

Do you mean that if you leave the stalled tab open but open a new one too that the new one works while the first one remains stalled?

If you open the system monitor does it show anything using excessive CPU?

How often are the gauges being updated? Add debug nodes showing what you are sending to the gauges to make sure that you are not accidentally flooding one of them with messages.

Do you have other dashboard pages that continue to work?

Is it only gauges on that page?

Please make sure to answer all questions.

[306936930]

Thank you for your help.

1. Yes, the existing tabs are still stuck, but newly opened tabs display correctly.

2. When opening a new tab, the server-side CPU and memory usage barely fluctuates, and hard drive storage is sufficient. The browser-side CPU usage increases from 10% to 20%, then returns to 10% after loading is complete. 3. Node-Red runs in a Docker container on a NAS device with a quad-core A53 ARM architecture. Several other containers also run, such as InfluxDB, Mosquitto, and Grafana. The average CPU utilization is around 25%. The browser runs on a PC. The dashboard data comes from MQTT messages, refreshed once per second. The MQTT payload is: {"temperature":30.92,"humidity":55.8,"pressure":99.01,"v1":4.6338,"v2":1.6475,"v3":1.646,"v4":1.6498,"v5":1.6516,"v6":1.6444,"v7":1.6466,"v8":1.6464}.

3. Charts have also been added to the dashboard.

4. Node-Red logs:

1 Jul 03:41:41 - [info] [ui-base:projetc name] Disconnected y8O7DfMnMFWQS1UNAADj due to ping timeout

1 Jul 03:44:45 - [info] [ui-base:projetc name] Disconnected tBi44_jIx1vKt016AADl due to ping timeout

1 Jul 03:47:54 - [info] [ui-base:projetc name] Disconnected whPQRFCW0lpF1qAqAADn due to ping timeout

1 Jul 03:51:07 - [info] [ui-base:projetc name] Disconnected wCpECb5uTZFvCtSlAADp due to ping timeout

1 Jul 03:54:11 - [info] [ui-base:projetc name] Disconnected cyv1IbFvrQRMF-AkAADr due to ping timeout

1 Jul 03:57:18 - [info] [ui-base:projetc name] Disconnected I5Z8ykv2g51wAn2AAADt due to ping timeout

1 Jul 04:00:34 - [info] [ui-base:projetc name] Disconnected jDSoMdS52VhckZMZAADv due to ping timeout

error

1 Jul 16:22:41 - [info] Stopping flows

1 Jul 16:22:41 - [info] [mqtt-broker:0de9ce2b00ef78b8] Disconnected from broker: mqtts://172.20.0.6:8883

1 Jul 16:22:41 - [info] Stopped flows

1 Jul 16:22:41 - [info] Updated flows

1 Jul 16:22:41 - [info] Starting flows
1 Jul 16:22:41 - [info] Started flows
1 Jul 16:22:41 - [info] [mqtt-broker:0de9ce2b00ef78b8] Connected to broker: mqtts://172.20.0.6:8883
1 Jul 16:22:42 - [info] [debug:debug 1]

[colinl]

When I go to the password login page it is asking for a phone number and password and when I enter admin admin it says the phone number is not long enough.

Sometimes charts can cause the sort of problems you are seeing. Please disable all the charts and see if the problem remains.

2. MQTT messages, refreshed once per second. The MQTT payload is: {"temperature":30.92,"humidity":55.8,"pressure":99.01,"v1":4.6338,"v2":1.6475,"v3":1.646,"v4":1.6498,"v5":1.6516,"v6":1.6444,"v7":1.6466,"v8":1.6464}.

Is that data driving all the gauges? Could you build a simple test flow using an Inject node to trigger the data injection and feeding the charts with typical data, then Export the test flow and paste it here please.

[colinl]

Another question, is it specifically that page that hangs and not others? If you have a simple page with just a Text In node, for example, does it continue to function or does it also stop after a time?

[Edit] Is you login authorisation expiring possibly?

[306936930]

1. For testing, I built a brand new node-red container, installed only the dashboard 2.0 node, and used a very simple process to test a dashboard. I tested it multiple times over 10 hours, and the page unresponsive issue was always reproduced.
2. I don't think it's due to expired permissions, because reopening the tab didn't prompt for a username and password. Of course, I tested the default settings.xml, which didn't have login permissions or HTTPS configured, and the page unresponsive issue was still reproduced.
3. If I returned to the browser to view the page after an hour, there was a scenario that should be the scenario before the unresponsiveness occurred: the data seemed to be in a congested state. At this point, the dashboard updated about 5 data points within 2 seconds, instead of the normal refresh rate of 1 per second. I believe this is a crucial clue to a bug.
4. I'm currently testing dashboard 1.0. If the problem persists, I will install Docker and various containers on a new device and then test it.

[colinl]

The next time it happens, open the browser developer tools console and refresh the page. See what errors are shown in the console and check the network tab to see if there are issues there.

[306936930]

After two days of testing, no issues occurred in dashboard1 (node-red-dashboard 3.6.6).

[colinl]

How about the console and network tabs in the developer tools (after hitting F5)?

[306936930]

I recall there were no error messages when the problem occurred, since the tab had already crashed. I'm sorry, but after the dashboard 2 update, I don't want to install this node; the experience is really bad. I suggest you try it and see if you can reproduce it. Thank you.

[colinl]

I recall there were no error messages when the problem occurred, since the tab had already crashed.

A page refresh should have sorted a crash, that is why looking at the browser network tab would be of great interest.

I don't want to install this node

You already had it installed. There was no need to uninstall it to use D1, they will happily co-exist.

I suggest you try it and see if you can reproduce it.

I have tried. Also many are using this node and no similar problem has been reported. Obviously you are seeing a problem and ideally we would like to find what is different about your system that triggers it.

[306936930]

I just tested dashboard2 again, running a node-red container on the new hardware environment, and encountered the same problem. I specifically checked the browser console, and there were no error messages; it just crashed at the same time as the tab. I'm exhausted. I'll use dashboard1 for now; dashboard1's CPU usage is much lower. Thank you.

[colinl]

there were no error messages; it just crashed at the same time as the tab.

What about when you refreshed the page?

You haven't said what you see in the Network tab when you refresh the page after it crashes? I don't know how many times I have asked that, it feels like dozens.