elasticache_info: ignore CacheClusterNotFound exeption during tag col…

…lect (#1777) (#1778)

[PR #1777/5461b2e2 backport][stable-5] elasticache_info: ignore CacheClusterNotFound exeption during tag collect

This is a backport of PR #1777 as merged into main (5461b2e).
If we call get_elasticache_tags_with_backoff() on a cluster with an invalid
state (eg: deleting), AWS will trigger an CacheClusterNotFound.
With this change, elasticache_info will ignore the cluster and continue with
the next one.

Reviewed-by: Mark Chappell

changelogs/fragments/elasticache_info-ignore-CacheClusterNotFound-when-reading-tags.yaml

@@ -0,0 +1,3 @@
+---
+bugfixes:
+- "elasticache_info - Ignore the CacheClusterNotFound exception when collecting tags (https://github.com/ansible-collections/community.aws/pull/1777)."

plugins/modules/elasticache_info.py

@@ -472,6 +472,9 @@ def get_elasticache_clusters(client, module):
         arn = "arn:aws:elasticache:%s:%s:cluster:%s" % (region, account_id, cluster['cache_cluster_id'])
         try:
             tags = get_elasticache_tags_with_backoff(client, arn)
+        except is_boto3_error_code("CacheClusterNotFound"):
+            # e.g: Cluster was listed but is in deleting state
+            continue
         except (botocore.exceptions.ClientError, botocore.exceptions.BotoCoreError) as e:
             module.fail_json_aws(e, msg="Couldn't get tags for cluster %s")