#Trident ontap-nas/storageclass config question
1 messages · Page 1 of 1 (latest)
just trying to make this more clear:
I have storageclass A and B
I also have backends A and B
SC A should be using backend A, etc.
what i have observed, is that if presented with 2 backends, trident will use SC A and randomly choose backend A or B.
The storageClass will choose from whatever backends are available to meet the need. The name is not a binding attribute. You can see more about what attributes will change the binding here: https://docs.netapp.com/us-en/trident/trident-reference/objects.html#kubernetes-storageclass-objects
thanks but I am not using storagepools.
additionally this is a bit unclear as to how to distinguish between two different backends that are of the same type, with similar parameters.
Ah... This looks like it may be what you are looking for. https://kb.netapp.com/Advice_and_Troubleshooting/Cloud_Services/Astra_Trident/How_Trident_uses_labels_to_selects_the_correct_backend
why would that not be in the public docs
This is in the public docs under the Virtual Storage Pools chapter: https://docs.netapp.com/us-en/trident/trident-use/ontap-nas-examples.html#examples-of-backends-with-virtual-pools
the easiest is really to use the parameter storagePools. In the following example, I map a SC to a specific backend/aggregate:
ok looking at your example, this is 2 backends that run on the same SVM, yes?
i have 2 backends each on their own separate SVM
also im not using storage pools
Do the SVMs use the same aggregates? If not, then you can use the storagePool parameter to specify the aggregate name.
yes i think they are on different aggs. so the storagepool param can specify the different aggregate even if there are no pools?