AWS Sync: Try all AWS accounts, and raise Exception afterwards

Title: Try all AWS accounts, and raise Exception afterwards

Description: Sometimes a single AWS account’s sync could fail. In such situations, we more likely want to proceed with the rest of our accounts, but still be notified of the failure.

I propose to wrap _sync_one_account into a try-catch, invoke logger.exception for the failures, and then finally raise another exception listing the accounts that failed.

https://github.com/lyft/cartography/blob/master/cartography/intel/aws/__init__.py#L145

For example:

failed_accont_ids = []
for account_id in account_ids:
    try:
        _sync_one_account(account_id)
    except Exception as e:
        logger.exception(e)
        failed_accont_ids.append(account_id)
if failed_accont_ids:
    raise Exception(f'Syncs failed for accounts {failed_accont_ids}')

[achantavy]

I agree on this approach and wanted to add that we should ensure that if there is an exception in one of the accounts, then execution stops and does not continue on to the cleanup jobs. We cannot have an exception in the get phase cause all data to be deleted.

[ryan-lane]

It’s worth mentioning that with SyncMetadata, it’s possible to determine if subsets of data are stale, so it’s a bit less painful to let that happen 😃