执行到 docker 重新启动容器

Question

I have a docker container which was working well at one point. I have changed some files inside this container and now it wont stop restarting due to the changes I made. I would like to enter this container but I keep getting this error:

Error response from daemon: Container c571b334990f0f2b3fc61f5f922b9a5eacf5347fb8720d004d22c80d8569fceb is restarting, wait until the container is running

Is there anything I can do to enter this container and fix the changes I made? Is there a feature/option where I start the container and enter it before it runs any entry script or does something?

Thanks for the help.

The errors in the logs:

goroutine 1 [running]:
jfrog.com/xray/pkg/utils/crypt.EncryptDecryptConnString(0x0, 0x0, 0x0, 0x0, 0x5d0, 0x5d, 0xc002fbb650)
    /go/src/jfrog.com/xray/pkg/utils/crypt/encrypt_util.go:37 +0x263
jfrog.com/xray/internal/dbaccess.OpenConnection()
    /go/src/jfrog.com/xray/internal/dbaccess/sql_layer.go:25 +0x65
jfrog.com/xray/internal/dbaccess.Connect()
    /go/src/jfrog.com/xray/internal/dbaccess/dbutil.go:10 +0x22
jfrog.com/xray/internal/startup.InitXrayServer()
    /go/src/jfrog.com/xray/internal/startup/start_xray_server.go:93 +0x4ef
main.main()
    /go/src/jfrog.com/xray/cmd/server/server.go:9 +0x20
Serverxray-server service container is starting...
Exec: /opt/jfrog/xray-server/server
panic: runtime error: index out of range [1] with length 1

goroutine 1 [running]:
jfrog.com/xray/pkg/utils/crypt.EncryptDecryptConnString(0x0, 0x0, 0x0, 0x0, 0x5d0, 0x5d, 0xc002f8f4d0)
    /go/src/jfrog.com/xray/pkg/utils/crypt/encrypt_util.go:37 +0x263
jfrog.com/xray/internal/dbaccess.OpenConnection()
    /go/src/jfrog.com/xray/internal/dbaccess/sql_layer.go:25 +0x65
jfrog.com/xray/internal/dbaccess.Connect()
    /go/src/jfrog.com/xray/internal/dbaccess/dbutil.go:10 +0x22
jfrog.com/xray/internal/startup.InitXrayServer()
    /go/src/jfrog.com/xray/internal/startup/start_xray_server.go:93 +0x4ef
main.main()
    /go/src/jfrog.com/xray/cmd/server/server.go:9 +0x20

Answer 1

Best way to handle situations like this is to override the entrypoint, by mounting it on the host. Find the entrypoint, and add a delay in the entrypoint so it gives you enough time to clean-up. The logs appears to be of jfrog xray and the entrypoint for this is /opt/jfrog/xray/app/bin/wrapper.sh (if the version of xray is 3.x)

Also be aware, if the changes you have done are persisted, the changes are actually on the volumes or mount-point. So you do not need container to start to undo the changes.