Hi again, it seems like the aws autoscaler is not spinning instances with the EBS configuration I configured. Here is the configuration:
resource_configurations { A100 { instance_type = "p3.2xlarge" is_spot = false availability_zone = "us-east-1b" ami_id = "ami-04c0416d6bd8e4b1f" ebs_device_name = "/dev/xvda" ebs_volume_size = 100 ebs_volume_type = "gp3" key_name = "..." security_group_ids = ["..."] extra_configurations = {"SubnetId": "..."} } }And here is the output of df -h in one of the aws instances created by the aws autoscaler:
Filesystem Size Used Avail Use% Mounted on udev 30G 0 30G 0% /dev tmpfs 6.0G 820K 6.0G 1% /run /dev/xvda1 31G 17G 15G 54% / tmpfs 30G 40K 30G 1% /dev/shm tmpfs 5.0M 0 5.0M 0% /run/lock tmpfs 30G 0 30G 0% /sys/fs/cgroup /dev/loop0 31M 31M 0 100% /snap/snapd/9721 /dev/loop1 29M 29M 0 100% /snap/amazon-ssm-agent/2333 /dev/loop2 56M 56M 0 100% /snap/core18/1932 /dev/loop3 32M 32M 0 100% /snap/snapd/10707 /dev/loop4 56M 56M 0 100% /snap/core18/1944 /dev/loop5 33M 33M 0 100% /snap/amazon-ssm-agent/2996 tmpfs 6.0G 0 6.0G 0% /run/user/1000So there is no 100Gb disk available. Note that the 100Gb volume isn't "attached but not mounted", since I don't see it either with lsblk :
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT loop0 7:0 0 31M 1 loop /snap/snapd/9721 loop1 7:1 0 28.1M 1 loop /snap/amazon-ssm-agent/2333 loop2 7:2 0 55.4M 1 loop /snap/core18/1932 loop3 7:3 0 31.1M 1 loop /snap/snapd/10707 loop4 7:4 0 55.4M 1 loop /snap/core18/1944 loop5 7:5 0 32.3M 1 loop /snap/amazon-ssm-agent/2996 xvda 202:0 0 32G 0 disk └─xvda1 202:1 0 32G 0 part /