[BUG] CPU gauge in dashboard overview shows only CPU usage of the first container instead of the sum of all containers

[Moo2Mee]

Description

CPU gauge in dashboard overview shows only CPU usage of the first container instead of the sum of all containers

Steps to reproduce

1. Run a few docker containers
2. Produce some load on one container. Leave the first container in the list idle.
3. Look at the CPU in the dashboard overview and at CPU in the details.

Logs

UI issue, no logs

Browser logs

Dockhand version

v1.0.24

Hawser version (if used)

No response

Connection mode

socket or direct ip

Please confirm the following

• I have already searched for relevant existing issues and discussions before opening this report.
• I have updated the title field above with a concise description.

[LordRofl]

same

[dancekid]

got same error on 2 / 3 machines.
OS: debian 12.13
docker on all 3 machines: 29.40
hawser: 0.2.39

One machine shows total CPU, two machines show first container cpu only

[Benares1]

Same here, Memory -and CPU-Summary in Dashboard are completely wrong.

[MeesHoogenboom]

Am experiencing this as well on v1.0.20. Using socket-proxy.

[jotka]

can you test fnsys/dockhand:dc8b47b1 please?

[dancekid]

Hello jotka,
its getting better.
dockhand shows 50% and btop 70%

greetings michael.

[Moo2Mee]

Now it shows reasonable values for me. The "Top containers by CPU" value shows up to 600% using 6 cores. Is this intended? Also in the container overview in the environment. Or should these values divided by the number of cores?