Box display bug - response timeout from local instance via Gladys Plus

[Tlse-vins]

I’ve recently been experiencing 2 issues.

The first:
My Gladys shuts down and doesn’t restart properly; I can’t connect to Gladys. Also, when I notice it and I’m not on site, I don’t know if it’s due to a network outage (not power, since the oven didn’t reset) or an internal bug.
I don’t know if I can view the logs after a reboot; I haven’t tried.

The second:

• The temperature graphs don’t refresh when I select my Temperature tab.
  I have to refresh the tab a second or even a third time to get the correct display.
• The humidity box shows the box that is on my main tab, i.e. a connected outlet that is no longer connected at the moment.

I think I’ve had this problem since the bathroom temperature sensor started acting up with the connection (it’s not the battery because I just checked on the zigbee2mqtt dashboard), but I’ll replace it to be sure.
The box tries to display values that don’t exist and keeps spinning.

[pierre-gilles]

Hello @Tlse-vins !

For the first one, did you check the Gladys logs? Did you check what was happening on your hardware?

@Tlse-vins

The temperature graphs don’t refresh when I select my Temperature tab.

Do you have any errors in the browser logs?

[Tlse-vins]

First click:

After the second refresh, I no longer see the message.

[pierre-gilles]

Your local instance is having problems contacting Gladys Plus.

In my opinion there is an issue with your machine, a hardware issue for example

[Tlse-vins]

Can you confirm:
I’ll check the disk status with the command:

df

and the CPU usage at that time with the command:

cat/proc/loadavg

[pierre-gilles]

Yes for df

Otherwise, I recommend htop to monitor CPU/RAM usage / load average:

htop

If it’s not installed, you can install it with apt-get install

[VonOx]

Same here today, no instance.

I’ll check when I get home tonight

[Tlse-vins]

Here is the output from my end.
I see that the swap partition is full, could that be the cause?

For the df command, my hard disk is at 6%

[pierre-gilles]

You have 600Mb of RAM available so I don’t think so, but I see you only have 1 day of uptime — did you do that reboot or did your Pi reboot by itself?

And the Gladys logs, anything unusual?

Anything else running on this Pi besides Gladys?

[Tlse-vins]

Hi,

No, it wasn’t me who did the reboots of my Raspberry Pi, I don’t know why it crashed.
For the logs, I have a lot of warn and info entries. Apart from that, nothing in particular.

No, I just installed Node-RED in Docker but I’m not using it for now in Gladys.

[pierre-gilles]

Ok! Hard to say, so it’s hard to know if everything’s okay ^^

Are you still having the problem or is it better?

[Tlse-vins]

I just tested and it’s the same.
Had to refresh the tab twice.

edit:
I think it’s coming from the graph — there are 3 temperatures in one and 3 humidity percentages in the other.

On another tab where I only have the temperature, I don’t have this problem.
Has anyone else encountered this problem when displaying multiple data on the same graph?

[pierre-gilles]

And locally, is everything okay? (on your local network)

Can you take a screenshot of the size/speed of the requests locally:

[Tlse-vins]

I’ll take a look at that tonight, if I can.

[pierre-gilles]

@Tlse-vins Could you re-display that page, I added error logs on Gladys Plus to better understand why your instance is not properly connected

[Tlse-vins]

OK I did several refreshes but I didn’t have the problem.

[pierre-gilles]

Ah! In a way, that’s for the best, aha

As soon as you have the issue again, tell me here when it happened

[Tlse-vins]

ok that works.
Yes but we’d need to understand.

[Tlse-vins]

It’s 11:42.

[pierre-gilles]

No luck, a small issue in the logs I had added, I didn’t notice anything

I’ve modified the logs, as soon as you see the issue again let me know