Docker Desktop crashing after a few weeks on windows 10

Docker Desktop
1

Hello community,

I have an application that I have been developing as multiple microservices on k8s. The application interacts with two external APIs and runs fine on Docker desktop (with kubernetes enabled) for a few weeks. After that, docker seems to crash somehow, with Docker desktop no longer running in the foreground, but docker desktop back-end usually still consuming system resources.

WSL-versjon: 2.3.26.0
Kjerneversjon: 5.15.167.4-1
WSLg-versjon: 1.0.65
MSRDC-versjon: 1.2.5620
Direct3D-versjon: 1.611.1-81528511
DXCore-versjon: 10.0.26100.1-240331-1435.ge-release
Windows-versjon: 10.0.19045.5247

I have .wslconfig setup with 15 GB memory and 12 GB swap, and also tried the autoMemoryReclaim function.

Docker desktop version 4.37.1

 kubectl get pods
E0131 13:19:24.433714    1688 memcache.go:265] couldn't get current server API g
roup list: Get "https://kubernetes.docker.internal:6443/api?timeout=32s": net/ht
tp: TLS handshake timeout

I wanted to post full log entries, but was denied by the “new users can only post 2 links block”.
Here are the relevant log entries I could find from the time of the crash:

2025-01-30 04:04:43 [W] skip listening node : docker builder "default" is already listening to the same workers: [6b078ced-ac3a-4de5-8fd8-cb4ac2ae6552]
2025-01-30 04:10:12 [E] failed to listen to docker events: unexpected EOF
2025-01-30 04:10:12 [E] reset build history listener for default/default: rpc error: code = Unavailable desc = error reading from server: EOF
2025-01-30 04:10:13 [E] failed to listen to docker events: request returned Bad Gateway for API route and version

[2025-01-31T12:28:09.642274400Z][ docker  diagnose] (8eada5dd-1) c5839d73-diagnose C<-S 734e25d4-BackendAPI GET /idle (2.553ms): {"booted":true,"busyReason":["33 container(s)","timed activities: map[/idle/make-busy:28999015600]"],"containers":33,"kubernetesEnabled":true,"timedActivities":{"/idle/make-busy":28999015600},"vmRunning":true}

[2025-01-31T12:28:09.642309200Z][docker diagnose] Triggering Linux sysrq and log flushes via a unix socket: \\.\  pipe\  dockerDiagnosticd
[2025-01-31T12:28:10.643538800Z][docker diagnose][W] /flush failed: Post "h t t p /unix/flush": context deadline exceeded
[2025-01-31T12:28:14.643526300Z][docker diagnose][W] (24d7e672-0) feec0729-com.docker.diagnose C<-S ConnectionClosed POST /diagnostics (5.0011581s): Post "http://ipc/diagnostics": context deadline exceeded

[2025-01-31T12:28:14.643526300Z][docker diagnose][W] (24d7e672-1) feec0729-com.docker.diagnose C<-S ConnectionClosed GET /ping (0s): Get "h  t t p:/  ipc/ping": context deadline exceeded

[2025-01-31T12:29:15.158804300Z][docker diagnose][W] [C:\Program Files\Docker\Docker\resources\bin\kubectl.exe --namespace docker get pods -o go-template={{range .items}}{{.metadata.name}}
{{end}}]: exit status 1
[2025-01-31T12:29:15.158804300Z][docker diagnose][W] E0131 13:28:35.015159    6176 memcache.go:265] couldn't get current server API group list: Get "h t t p s :/ kubernetes.docker.internal:6443/api?timeout=32s": net/http: TLS handshake timeout
E0131 13:28:45.094611    6176 memcache.go:265] couldn't get current server API group list: Get " 
  kubernetes.docker.internal:6443/api?timeout=32s": net/http: TLS handshake timeout

Unable to connect to the server: EOF

Could a full reinstall of Docker desktop/WSL/Ubuntu fix this or do you think the problem lies elsewhere?

Thank you for any constructive input!

Best regards,
/S

2

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.