[LEGACY] [CI]: TestRestartWithSignal does not seem to (always?) work correctly (EL8 only?)

[apostasie]

Description

Seen on #4056

See attached trace (note that PR #4056 has the new timeout logic).

Signal has been sent and received, but the container has not restarted (and was instead terminated by timeout).

@AkihiroSuda the good news here is that the new timeout logic is working as expected, and this test would have previously hung-up on EL8.

Now we just have to figure out what is happening here.

Steps to reproduce the issue

Describe the results you received and expected

=== FAIL: cmd/nerdctl/container TestRestartWithSignal (21.97s)
    container_restart_linux_test.go:163: 
    container_restart_linux_test.go:163: ======================== Pre-test cleanup ========================
    container_restart_linux_test.go:163: 
    container_restart_linux_test.go:163: ======================== Test setup ========================
    container_restart_linux_test.go:163: 
    container_restart_linux_test.go:163: ======================== Test Run ========================
    command.go:435: [2025-04-01T16:46:36Z] [command=/usr/local/bin/nerdctl --namespace=nerdctl-test run --name testrestartwithsignal-27bfb0e4 ghcr.io/stargz-containers/alpine:3.13-org sh -c #!/bin/sh
        	set -eu
        
        	sig_msg () {
        		printf "received\n"
        		[ "false" != true ] || exit 0
        	}
        
        	trap sig_msg 10
        	printf "trap ready\n"
        	while true; do
        		printf "waiting...\n"
        		sleep 0.5
        	done
        ] command cancelled
    command.go:115: expected: 137  - to be equal to: -1
    command.go:115: Expected exit code: 137
        
    command.go:115: 
        =================================
        | Command:	/usr/local/bin/nerdctl run --name testrestartwithsignal-27bfb0e4 ghcr.io/stargz-containers/alpine:3.13-org sh -c #!/bin/sh
        	set -eu
        
        	sig_msg () {
        		printf "received\n"
        		[ "false" != true ] || exit 0
        	}
        
        	trap sig_msg 10
        	printf "trap ready\n"
        	while true; do
        		printf "waiting...\n"
        		sleep 0.5
        	done
        
        | Working Dir:	/tmp/TestRestartWithSignal2256824605/001
        | Timeout:	20s
        =================================
        	SHELL=/bin/bash
        	LOGNAME=rootless
        	XDG_SESSION_TYPE=tty
        	HOME=/home/rootless
        	LANG=C.UTF-8
        	SSH_CONNECTION=::1 41538 ::1 22
        	XDG_SESSION_CLASS=user
        	IPFS_PATH=/home/rootless/.local/share/ipfs
        	USER=rootless
        	SHLVL=1
        	XDG_SESSION_ID=4
        	XDG_RUNTIME_DIR=/run/user/1001
        	SSH_CLIENT=::1 41538 22
        	DBUS_SESSION_BUS_ADDRESS=unix:path=/run/user/1001/bus
        	OLDPWD=/home/rootless
        	_=/usr/local/bin/gotestsum
        	PATH=/usr/local/go/bin:/usr/local/go/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin
        	***
        	DOCKER_CONFIG=/tmp/TestRestartWithSignal2256824605/001
        	NERDCTL_TOML=/tmp/TestRestartWithSignal2256824605/001/nerdctl.toml
        =================================
        | Stderr:
        =================================
        
        =================================
        | Stdout:
        =================================
        trap ready
        waiting...
        waiting...
        waiting...
        received
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        waiting...
        
        =================================
        | Exit Code: -1
        | Signaled: killed
        | Err: command timed out
        =================================
    case.go:180: 
    case.go:181: ======================== Post-test cleanup ========================

What version of nerdctl are you using?

main

Are you using a variant of nerdctl? (e.g., Rancher Desktop)

None

Host information

I WANT A GREEN CI AND WILL NOT STOP BEING ENRAGED UNTIL WE GET IT 😡🤣

[apostasie]

Maybe EL8 is just really really really slow and we need to extend the timeout from 20 seconds to 30?

[AkihiroSuda]

Maybe EL8 is just really really really slow and we need to extend the timeout from 20 seconds to 30?

Maybe yes, as EL 8 is executed on a nested virt

[apostasie]

Ok, let's test the hypothesis.

We can merge it - it is rather innocuous. Worse case scenario, it will still fail on EL8, just slower...

#4130