We have a CodeBuild jobs that connect to our RDS databases using TLS. For that we need to grab the AWS RDS CA Bundle file, which we download using curl
.
curl https://s3.amazonaws.com/rds-downloads/rds-combined-ca-bundle.pem --create-dirs -o $MYSQL_SSLCA
Everything works fine most of the time, but every once in a while we get a SSL_ERROR_ZERO_RETURN
error after it tried to download the file for 60 seconds. If we rerun the job everything works fine.
Dload Upload Total Spent Left Speed
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- 0:00:01 --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- 0:00:02 --:--:-- 0
....
0 0 0 0 0 0 0 0 --:--:-- 0:00:52 --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- 0:00:53 --:--:-- 0
0 0 0 0 0 0 0 0 --:--:-- 0:00:53 --:--:-- 0
curl: (35) OpenSSL SSL_connect: SSL_ERROR_ZERO_RETURN in connection to s3.amazonaws.com:443
My question is, why is this happening? I was under the impression that S3 was more robust than this. Is there some problem with this request? Is anyone else seeing this problem? We're not even running that many CodeBuild jobs. maybe 100 / week, and I've seen this problem enough to recognize it.