13 Posts
0
822
December 23rd, 2020 17:00
Operating specifications of the Ansible module
Hi
We are currently using Ansible Modules for Dell EMC PowerMax to verify VMAX configuration changes and other operations.
Since the following events occurred, we would like to confirm the operation specifications of the Ansible module.
Implementation details
Two 10GB volumes and one 20GB volume are created using the following Ansible volume module.
→Assumed result: Three new Symdevs will be created, and they will join the StorageGroup specified in the Playbook.
Result
・A new 10GB Symdev has been created as "00067" and "00068".
・A 20GB Symdev has been allocated as "000B3", but an existing Symdev has been allocated.
※"000B3" existed on VMAX at 20GB before Ansible was run.
it did not belong to any StorageGroup.
Question
As a behavior of VMAX using Unisphere and SolutionEnabler, when a volume is created, the available young number is automatically selected and a new Symdev is created.
In this case where Ansible was executed, only 20GB was not newly created and "000B3" was overwritten.
Can you please tell me the working specification of Ansible module?
The following is the information about the Ansible module we used.
module
Ansible Modules for Dell EMC PowerMax Version 1.1
・dellemc_powermax_volume.py
ran in playbook
- name: Create volume
dellemc_powermax_volume:
unispherehost: "{ {unispherehost}}"
universion: "{ {universion}}"
verifycert: "{ {verifycert}}"
user: "{ {user}}"
password: "{ {password}}"
serial_no: "{ {serial_no}}"
vol_name: "{ {vol_name}}"
sg_name: "{ {sg_name}}"
size: ""
cap_unit: "{ {cap_unit}}"
state: 'present'
bmcfeeters
1 Rookie
•
72 Posts
0
January 5th, 2021 08:00
Hello,
It appears that the 1.0 version of the modules exhibit this behavior as I was able recreate the problem with the volumes module from that release. After upgrading to the latest version 1.3 the problem went away and a new volume was created every time as appropriate.
Thanks
Bryan
FoolInTheRain
123 Posts
0
January 4th, 2021 08:00
I have made engineering aware of this issue, and have asked that they look into it deeper. When there is more info from them, you should see another reply.
v_koji.i_s
13 Posts
0
January 8th, 2021 03:00
Thank you for your reply.
The same content is reproduced,
It turns out that this content has been improved in version1.3.
FoolInTheRain
123 Posts
0
January 13th, 2021 09:00
To add further info to the device selection process in the newer release configuration, the following was an explanation given to engineering:
*****
To fully explain the device selection on the API if a new device is not being created the API will seek to reuse existing devices. This is advantageous in that we don’t create any new device and consume more meta data un-necessarily. If a matching device that is not in use can not be found the API will create a new device.
The criteria for automatic device selection is as follows:
With the criteria above you can safely assume that the ansible modules, even if they do select an existing device will not have an impact on any customer data. Re-Use of devices does not wipe any existing data, if a customer wishes to do this then as part of their reclaim operations they should do a free on the devices or simply delete when they are being decommissioned.
****