Pynetbox - get clusters of a specific type and tenant

[nomad-cyanide]

Hi

I am trying to get a list of the clusters of a specific type with a specific tenant set.

Type I get like this:
nb_clu_type=nb.virtualization.cluster_types.get(name='Check Point VSX')

Then I want to get the list of clusters, but nothin I have tried so far has worked, like
nb_clu=nb.virtualization.cluster.get(cluster_type_id=nb_clu_type.id)
nb_clu=nb.virtualization.cluster.get(type=nb_clu_type.id)
nb_clu=nb.virtualization.cluster.get(cluster_type=nb_clu_type.name)

The end goal is to get all the virtual machines in a list, from the clusters of type "Check Point VSX" assigned to a specific tenant.

What am I not understanding correctly?

Best regards
Ian

[candlerb]

Since you are trying to get a list of things, rather than a single thing, I think you want filter(...) rather than get(...)

>>> list(nb.virtualization.virtual_machines.filter(cluster_type='ganeti'))
[elk]
>>>

[nomad-cyanide]

But this will not limit the search to only the specific cluster type within a specific tenant.

[candlerb]

It's up to you to modify the filter as appropriate; you can include multiple conditions in the query. Note that you can either assign tenant to the cluster or to individual virtual machines so it depends exactly what you're trying to filter on.

At worst, you may have to do one query to find a list of clusters of interest, and then provide the list of clusters to the VM search:

list(nb.virtualization.virtual_machines.filter(cluster_id=[5,6]))

[nomad-cyanide]

I don't think I have been clear enough about my problem. First, I get that I should use filter, to get multiple items. That was a brainfart.
But really, I need to get the a list of the cluster of type X within tenant Z, not just any cluster with type X. This is the part that is troubling me.
I thought something like this might work:
nb_clu_tenant = list(nb.virtualization.cluster.filter(cluster_type_id=nb_clu_type.id,tenant_id=666))
but no such luck.

After I get that list of clusters, I need to list all of the VM's in those clusters, which likely would be something like:

for list in nb_clu_tenant:
   vms = nb.virtualization.virtual_machines.filter(cluster_id=cluster.id)

[candlerb]

Going back to the original query:

I am trying to get a list of the clusters of a specific type with a specific tenant set

>>> nb.tenancy.tenants.get(name="blah").id
4
>>> list(nb.virtualization.clusters.filter(tenant_id=4))
[XXX, YYY, ZZZ]
>>> nb.virtualization.cluster_types.get(name="Proxmox VE").id
8
>>> list(nb.virtualization.clusters.filter(tenant_id=4,type_id=8))
[XXX, ZZZ]

The end goal is to get all the virtual machines in a list, from the clusters of type "Check Point VSX" assigned to a specific tenant.

>>> cluster_ids = [c.id for c in nb.virtualization.clusters.filter(tenant_id=4,type_id=8)]
>>> cluster_ids
[68, 70]
>>> for v in nb.virtualization.virtual_machines.filter(cluster_id=cluster_ids): print(v)
...
VM1
VM2
...

NOTE: you can currently filter directly using slugs: here with Netbox v3.1.9 and pynetbox 6.6.1

>>> list(nb.virtualization.clusters.filter(type="proxmox-ve"))
[XXX, YYY, ZZZ, ...]

However my current understanding is that (sadly) slugs are going to be completely removed from Netbox in the near future, so you would be best advised not to do this.

EDIT: correction, they're not going away completely, but they may be weakened to the point where they are no longer usable as unique identifiers. See #8113.

[candlerb]

I thought something like this might work:
nb_clu_tenant = list(nb.virtualization.cluster.filter(cluster_type_id=nb_clu_type.id,tenant_id=666))
but no such luck.

Since the query is on cluster, you need type_id not cluster_type_id

[markkuleinio]

My VMs:

>>> for vm in nb.virtualization.virtual_machines.filter("VM"):
...     print(vm.name, vm.cluster)
...
TestVM Host1
VM1 Cluster1
VM2 Cluster2

My clusters and cluster types:

>>> for cluster in nb.virtualization.clusters.filter("Cluster"):
...     print(cluster.name, cluster.type)
...
Cluster1 Type1
Cluster2 Type2

VMs in clusters of Type1:

>>> nb_clu_type = nb.virtualization.cluster_types.get(name="Type1")
>>> clusters = nb.virtualization.clusters.filter(type_id=nb_clu_type.id)
>>> for cluster in clusters:
...     print("Cluster:", cluster)
...     print("VMs:", list(nb.virtualization.virtual_machines.filter(cluster_id=cluster.id)))
...
Cluster: Cluster1
VMs: [VM1]

[markkuleinio]

With tenant:

>>> for cluster in nb.virtualization.clusters.filter("Cluster"):
...     print(cluster.name, cluster.type, cluster.tenant)
...
Cluster1 Type1 None
Cluster1-Tenant1 Type1 Tenant1
Cluster2 Type2 None
>>> nb_clu_type = nb.virtualization.cluster_types.get(name="Type1")
>>> tenant = nb.tenancy.tenants.get(name="Tenant1")
>>> clusters = nb.virtualization.clusters.filter(type_id=nb_clu_type.id, tenant_id=tenant.id)
>>> for cluster in clusters:
...     print("Cluster:", cluster)
...     print("VMs:", list(nb.virtualization.virtual_machines.filter(cluster_id=cluster.id)))
...
Cluster: Cluster1-Tenant1
VMs: [VM3]

[nomad-cyanide]

clusters = nb.virtualization.clusters.filter(type_id=nb_clu_type.id, tenant_id=tenant.id)
Thank you. I think I have tried this already, but I'll try again. I may have made a mess in something else, since it didn't work for me yesterday. I'll get back to you.