You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
while it is great to grab all inactive targets and connect them the process to actually get there is missing.
Situation:
We deploy a new cluster node and configure the necessary things for it on the iscsi server side.
Now we switch to the node-to-be and add the target portal. -> stage one
after that we log into it and see every target there is to have. -> stage two
With that setup we are able to connect every single target with the proper iqn and chap secret and what else you need to place. -> stage three
currently we can only setup these things if we already KNOW the nodeaddress.
This is a very static thing. with the staged process we are able to keep a minimum of configuration and make it work.
I am not fluent in iSCSI but that way it works just fine with Nutanix Volumes (Server) and Windows Server 2016-2022 (Client)
The server side here is non-windows so I do not need any dsc thingy on that part.
Proposed properties
Property
Type qualifier
Data type
Description
Default value
Allowed values
PropertyName
Key
String
Detailed description
None
None
We do not need any new properties related to iSCSIinitiator, we just need to drop the nodeaddress requirement to actually discover the targets before looping through them and use iSCSIInitiator after that.
That way we can error out early on and make sure to have the targetportal setup before the targets.
the current iSCSIinitiator does everything to setup a target but it is very repetitive to configure.
Special considerations or limitations
This might not be the correct process for every iSCSI setup so it might need to be tested for other configurations besides Nutanix Volumes.
The text was updated successfully, but these errors were encountered:
Resource proposal
while it is great to grab all inactive targets and connect them the process to actually get there is missing.
Situation:
We deploy a new cluster node and configure the necessary things for it on the iscsi server side.
Now we switch to the node-to-be and add the target portal. -> stage one
after that we log into it and see every target there is to have. -> stage two
With that setup we are able to connect every single target with the proper iqn and chap secret and what else you need to place. -> stage three
currently we can only setup these things if we already KNOW the nodeaddress.
This is a very static thing. with the staged process we are able to keep a minimum of configuration and make it work.
I am not fluent in iSCSI but that way it works just fine with Nutanix Volumes (Server) and Windows Server 2016-2022 (Client)
The server side here is non-windows so I do not need any dsc thingy on that part.
Proposed properties
We do not need any new properties related to iSCSIinitiator, we just need to drop the nodeaddress requirement to actually discover the targets before looping through them and use iSCSIInitiator after that.
That way we can error out early on and make sure to have the targetportal setup before the targets.
the current iSCSIinitiator does everything to setup a target but it is very repetitive to configure.
Special considerations or limitations
This might not be the correct process for every iSCSI setup so it might need to be tested for other configurations besides Nutanix Volumes.
The text was updated successfully, but these errors were encountered: