[TPU] tpu-v5-8 failed to create and says 'Insufficient reserved capacity' when no reservation is used

[cblmemo]

$ sky launch --gpus tpu-v5p-8 --region us-central1 -c ttpu
Missing runtime_version in accelerator_args, using default (v2-alpha-tpuv5)
Considered resources (1 node):
-----------------------------------------------------------------------------------------
 CLOUD   INSTANCE   vCPUs   Mem(GB)   ACCELERATORS   REGION/ZONE     COST ($)   CHOSEN   
-----------------------------------------------------------------------------------------
 GCP     TPU-VM     -       -         tpu-v5p-8:1    us-central1-a   16.80         ✔     
-----------------------------------------------------------------------------------------
Launching a new cluster 'ttpu'. Proceed? [Y/n]: 
⚙︎ Launching on GCP us-central1 (us-central1-a).
W 01-16 17:42:36 instance_utils.py:112] Got return code 8 in us-central1-a: 'Insufficient reserved capacity. Contact customer support to increase your reservation. [EID: 0x7adf6d58c08b0aa4]'
sky.exceptions.ResourcesUnavailableError: Failed to acquire resources in us-central1-a for {GCP({'tpu-v5p-8': 1}, accelerator_args={'runtime_version': 'v2-alpha-tpuv5'})}. 

↺ Trying other potential resources.
⨯ Failed to provision resources. View logs at: ~/sky_logs/sky-2025-01-16-17-42-20-623267/provision.log

sky.exceptions.ResourcesUnavailableError: Failed to provision all possible launchable resources. Relax the task's resource requirements: 1x GCP({'tpu-v5p-8': 1}, accelerator_args={'runtime_version': 'v2-alpha-tpuv5'})
To keep retrying until the cluster is up, use the `--retry-until-up` flag.