使用 CF 模板更新 AWS 堆栈

Question

Hope everyone is keeping safe!

I have a Ruby on Rails application hosted on AWS Beanstalk. I am using CloudFormation template, to update any stack for eg, Ruby version, Linux Platform upgrade etc.

I was trying to upgrade, Linux box to 2.11.7 and Ruby to 2.6.6 and then ElasticSearch to 7.4

I was doing these changes in CloudFormation YML template and then I ran aws cloudformation update-stack command to apply these changes.

While the changes took time, I accidentally clicked on Rebuild Environment from Web AWS Console as a result, all the previously configured settings like SQS , Load balancer etc., were replaced by new settings.

Now, whenever I am trying to execute the update-stack command, it fails with below errors:

2020-06-09 15:25:44 UTC+0530
WARN
Environment health has transitioned from Info to Degraded. Command failed on all instances. 
Incorrect application version found on all instances. Expected version "code-pipeline-xxxxxxxxxx" (deployment 2377). Application update failed 40 seconds ago and took 79 seconds.

2020-06-09 15:25:03 UTC+0530    
INFO
The environment was reverted to the previous configuration setting.

2020-06-09 15:24:44 UTC+0530    
INFO
Environment health has transitioned from Ok to Info. Application update in progress on 1 instance. 0 out of 1 instance completed (running for 39 seconds).

2020-06-09 15:24:30 UTC+0530    
ERROR
During an aborted deployment, some instances may have deployed the new application version. 
To ensure all instances are running the same version, re-deploy the appropriate application version.

2020-06-09 15:24:30 UTC+0530    
ERROR
Failed to deploy application.

2020-06-09 15:24:30 UTC+0530    
ERROR
Unsuccessful command execution on instance id(s) 'i-xxxxxxxxxx'. Aborting the operation.

2020-06-09 15:24:30 UTC+0530    
INFO
Command execution completed on all instances. Summary: [Successful: 0, Failed: 1].

2020-06-09 15:24:30 UTC+0530    
ERROR
[Instance: i-xxxxxxxxxx] Command failed on instance. Return code: 18 Output: (TRUNCATED)...g: the running version of Bundler (1.16.0) is older than the version that created the lockfile (1.17.3). We suggest you upgrade to the latest version of Bundler by running `gem install bundler`. Your Ruby version is 2.6.6, but your Gemfile specified 2.6.5. Hook /opt/elasticbeanstalk/hooks/appdeploy/pre/10_bundle_install.sh failed. For more detail, check /var/log/eb-activity.log using console or EB CLI.

2020-06-09 15:24:19 UTC+0530    
INFO
Deploying new version to instance(s).

2020-06-09 15:23:45 UTC+0530    
INFO
Updating environment developWeb's configuration settings.

2020-06-09 15:23:36 UTC+0530    
INFO
Environment update is starting.

I can confirm that I have Ruby-2.6.6 set. I am not sure from where it is picking up the old version of Ruby?

Is there any way I can fix this? OR forcefully apply template changes?

Any help on this would be highly appreciated.

[UPDATE]: When I try to connect to ElasticSearch from Rails console, I get:

Faraday::ConnectionFailed: Failed to open TCP connection to old-elasticsearch-host-name.es.amazonaws.com:80 (Hostname not known: old-elasticsearch-host-name.es.amazonaws.com)
from /opt/rubies/ruby-2.6.6/lib/ruby/2.6.0/net/http.rb:949:in `rescue in block in connect'
Caused by SocketError: Failed to open TCP connection to old-elasticsearch-host-name.es.amazonaws.com:80 (Hostname not known: old-elasticsearch-host-name.es.amazonaws.com)
from /opt/rubies/ruby-2.6.6/lib/ruby/2.6.0/net/http.rb:949:in `rescue in block in connect'
Caused by Resolv::ResolvError: no address for old-elasticsearch-host-name.es.amazonaws.com
from /opt/rubies/ruby-2.6.6/lib/ruby/2.6.0/resolv.rb:94:in `getaddress'

The new URL of elasticsearch instance is different but it is still picking up the old URL from ELASTICSEARCH_HOST ENV variable.

Information from my CF template:

I can now provide info as per request. please tag me to see what I have in CF template

Answer 1

This was a config issue.

Whenever I was running aws update-stack command, it was going to s3 and pulling the zip (of the source code) code and in Gemfile of that zip code ruby version was set to 2.6.5.

So, I'd uploaded the fresh copy of the source code and then executed the update-stack command and it worked