You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Random error on startup - Could not determine target lifecycle of instance: operation error ec2imds: GetMetadata, http response error StatusCode: 404, request to EC2 IMDS failed" module=agent.go
#4487
Open
mello7tre opened this issue
Feb 4, 2025
· 0 comments
Hi, after update to 1.89.3 i am experiencing random error on startup.
I am running the official aws ecs image al2023-ami-ecs-hvm-2023.0.20250129-kernel-6.1-x86_64.
Docker container ecs-agent do not start and looking at log file /var/log/ecs/ecs-agent.log last lines are:
level=info time=2025-02-04T07:02:16Z msg="Loading state!" module=state_manager.go
level=info time=2025-02-04T07:02:16Z msg="Event stream ContainerChange start listening..." module=eventstream.go
level=info time=2025-02-04T07:02:16Z msg="Waiting for instance to go InService" module=agent.go
level=critical time=2025-02-04T07:02:21Z msg="Could not determine target lifecycle of instance: operation error ec2imds: GetMetadata, http response error StatusCode: 404, request to EC2 IMDS failed" module=agent.go
After i discover the problem, usually if i restart the service ecs, the second time it start up without any error.
But this is a problem because when happen, cluster ecs instance start up but do not connect to ECS cluster, and there is no watchdog service to mark it as bad and terminate the instance.
Regards.
The text was updated successfully, but these errors were encountered:
Hi, after update to
1.89.3
i am experiencing random error on startup.I am running the official aws ecs image
al2023-ami-ecs-hvm-2023.0.20250129-kernel-6.1-x86_64
.Docker container ecs-agent do not start and looking at log file
/var/log/ecs/ecs-agent.log
last lines are:After i discover the problem, usually if i restart the service ecs, the second time it start up without any error.
But this is a problem because when happen, cluster ecs instance start up but do not connect to ECS cluster, and there is no watchdog service to mark it as bad and terminate the instance.
Regards.
The text was updated successfully, but these errors were encountered: