[Bug] fail to get dynamic inventory for more > 50/90 VMs in module ntnx_prism_vms_inventory

[nahsiy]

Following the migration of our central prism from pc.2022.6.7 to 2024.1, our ntnx_prism_vms_inventory plugin no longer works without an explicit error, as soon as we set a lenght of more than 50/90.

Via curl or postman, it works for 500 VMs, which seems to be the limit.

our inventory code :

---

plugin: nutanix.ncp.ntnx_prism_vm_inventory
strict: false
nutanix_hostname: "{{cluster}}""
nutanix_port: "" # Make sure the port is correct
validate_certs: false
data: {"offset": 0, "length": 1000}
groups:
  vm_OFF: "'OFF' in power_state"
  vm_NGT: "'INSTALLED' in guest_tools.nutanix_guest_tools.ngt_state"
keyed_groups:
  - prefix: "vm_ZONE"
    separator: ':'
    key: "ntnx_categories.ZONE"
  - prefix: "vm_SITE"
    separator: ':'
    key: "ntnx_categories.SITE"
  - prefix: "vm_OS"
    separator: ':'
    key: "ntnx_categories.OS"
  - prefix: "vm_MCO"
    separator: ':'
    key: "ntnx_categories.MCO"

before this migration, we had no problem recovering all the VMs in our cluster (554VMS) with this configuration

[bhati-pradeep]

@nahsiy Can you share error log which you are facing ?

[nahsiy]

Hello,

(For lengh = 1000)

ansible-inventory -i 00-nutanix.yaml --list

[WARNING]: * Failed to parse /home/cmassieu/Documents/gitlab/fr-ansible/00-nutanix.yaml with auto plugin: 'Mock_Module' object has no attribute 'fail_json' [WARNING]: * Failed to parse /home/cmassieu/Documents/gitlab/fr-ansible/00-nutanix.yaml with yaml plugin: Plugin configuration YAML file, not YAML inventory [WARNING]: * Failed to parse /home/cmassieu/Documents/gitlab/fr-ansible/00-nutanix.yaml with ini plugin: Invalid host pattern '---' supplied, '---' is normally a sign this is a YAML file. [WARNING]: Unable to parse /home/cmassieu/Documents/gitlab/fr-ansible/00-nutanix.yaml as an inventory source [WARNING]: No inventory was parsed, only implicit localhost is available { "_meta": { "hostvars": {} }, "all": { "children": [ "ungrouped" ] } }

Regards

Yishan

[stahnjones]

500 entities appears to be the internal hard-coded limit for a single API call in V3 API

[bhati-pradeep]

Hi @stahnjones, for length > 500 in vms, it internally takes care of it by doing multiple api calls to fetch give number of vms.

For this ticket, issue was pc slowness causing timeout while doing list calls.
We have a ticket open for override timeout : #409

@nahsiy can we close this ticket for now and track capability of increasing timeout in above mentioned ticket ?

[nahsiy]

Hello @bhati-pradeep,

Yes, thanks a lot !