sync users not running about terraform-aws-bastion

[Guimove]

mmh I don't see any issue with your code, I think the issue is in the user data script execution.

Are all the scripts created on the server ?

I mean are these script present on the bastion :

• /usr/bin/bastion/sync_users
• /usr/bin/bastion/sync_s3
• /usr/bin/bastion/shell

The crontab of the root user is empty ?

[brunelloriserva]

The scripts are there, no crontab entry.

…

On Mon, Apr 25, 2022, at 12:29 AM, Guillaume wrote: mmh I don't see any issue with your code, I think the issue is in the user data script execution. Are all the scripts created on the server ? I mean are these script present on the bastion : * /usr/bin/bastion/sync_users * /usr/bin/bastion/sync_s3 * /usr/bin/bastion/shell The crontab of the root user is empty ? — Reply to this email directly, view it on GitHub <#137 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAZJ4SEWVBVDJ5NFCD7MQYTVGYNQZANCNFSM5UHJYKIA>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[matloob-smartmimic]

I am getting the same issue as well, however I see crontab entries present on my bastion box, also files sync_users is also present at /usr/bin/bastion/

[root@ip-10-122-22-79 home]# crontab -l
*/5 * * * * /usr/bin/bastion/sync_users
0 0 * * * yum -y update --security
*/5 * * * * /usr/bin/bastion/sync_s3

@brunelloriserva did you get your problem sorted?
@Guimove - can you please help us with this.

AWS- Looks like on this bastion aws cli version is 1.2x

NOTE- I am not sure, but looks like it does not respond at following script line
aws s3api list-objects --bucket

[kcasas]

I got the same issue and it has something to do with the s3 bucket name.
I renamed the s3 bucket after the second apply. The s3 bucket was replaced just fine but the scripts inside the ec2 instance was not updated and it still points to the old bucket.

[kcasas]

@matloob-smartmimic @brunelloriserva I resolved my issue by making sure that the bastion EC2 instance is updated with the latest launch template template version. You can compare the EC2 tag aws:ec2launchtemplate:version with the launch template tag.

TLDR: Triggering instance refresh on EC2 > Auto Scaling Groups > ${bastion ASG} > Instance Refresh resolved my issue. You can also do this using awscli.

@Guimove I guess the fix would be utilizing https://registry.terraform.io/providers/hashicorp/aws/latest/docs/resources/autoscaling_group#instance_refresh.

[josescuderoh]

I had this problem when I was setting the auto_scaling_group_subnets to private subnets just like @brunelloriserva. The fix was using public subnets for the ASG and triggering an Instance Refresh as indicated by @kcasas

[ruiwei]

@josescuderoh I'm having the same problem. how do you trigger an Instance Refresh?