How do I debug failed healthcheck errors on ECS.
The question is very vague because I don't get much from AWS in terms of error logs.
The only information I have is when I run my Fargate service task, I get:
Task failed ELB health checks in (target-group TARGET_GROUP_NAME)
In the target group the status is: Request timed out.
The logs in the container aren't much more useful:
4/10/2023, 7:04:52 PM [2023-04-10 15:04:52 +0000] [1] [INFO] Handling signal: term backend
4/10/2023, 7:04:28 PM [2023-04-10 15:04:28 +0000] [67] [INFO] Booting worker with pid: 67 backend
4/10/2023, 7:04:28 PM [2023-04-10 15:04:28 +0000] [1] [WARNING] Worker with pid 63 was terminated due to signal 9 backend
4/10/2023, 7:04:27 PM [2023-04-10 15:04:27 +0000] [1] [CRITICAL] WORKER TIMEOUT (pid:63) backend
4/10/2023, 7:03:57 PM [2023-04-10 15:03:57 +0000] [63] [INFO] Booting worker with pid: 63
I'm certain of my ports/ALB configuration/Target group configuration since my task was running perfectly fine before I deployed my latest changes for my docker container.
My docker works on my local machine. There weren't even any major changes.
I am also unable to ping my endpoint.
I understand this is vague, I'm not looking for a resolution just looking for how do I even start to debug this as there are absolutely no useful logs anywhere.
AWS OFFICIALUpdated 21 days ago
Based on the container logs you provided, I don't believe this is an ECS or ELB issue. The ELB health checks will fail if the application doesn't respond. It looks to me like your container itself is failing which then cascades up to the ELB. You say your container changes were not major. First, I would try reverting and make sure the last version of your container still works. A common cause of those signals is running out of memory.