Add proptest for testing RBAC permissions

[fabubaker]

What changes were proposed in this pull request?

Why are the changes needed?

Does this PR introduce any user-facing change? If yes is this documented?

How was this patch tested?

Are there any further changes required?

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Rust Benchmark'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 2.

Benchmark suite	Current: 03853c7	Previous: 9823ef7	Ratio
lotr_graph/num_edges	5 ns/iter (± 0)	0 ns/iter (± 0)	+∞
lotr_graph/num_nodes	126 ns/iter (± 32)	1 ns/iter (± 0)	126
lotr_graph/graph_latest	3 ns/iter (± 0)	0 ns/iter (± 0)	+∞
lotr_graph_materialise/materialize	7039156 ns/iter (± 23250)	1564816 ns/iter (± 35303)	4.50
lotr_graph_window_100/num_edges	29 ns/iter (± 0)	8 ns/iter (± 0)	3.63
lotr_graph_window_100/num_nodes	140 ns/iter (± 34)	5 ns/iter (± 0)	28
lotr_graph_window_100_materialise/materialize	7427428 ns/iter (± 26556)	1669150 ns/iter (± 10700)	4.45
lotr_graph_window_10/has_node_existing	147 ns/iter (± 9)	62 ns/iter (± 11)	2.37
lotr_graph_window_10/iterate nodes	36569 ns/iter (± 172)	11339 ns/iter (± 40)	3.23
lotr_graph_window_10/iterate edges	101521 ns/iter (± 189)	48684 ns/iter (± 211)	2.09
lotr_graph_window_10_materialise/materialize	3088293 ns/iter (± 18110)	971980 ns/iter (± 4278)	3.18
lotr_graph_subgraph_10pc/has_edge_existing	261 ns/iter (± 4)	93 ns/iter (± 1)	2.81
lotr_graph_subgraph_10pc/num_nodes	123 ns/iter (± 27)	4 ns/iter (± 0)	30.75
lotr_graph_subgraph_10pc/has_node_existing	118 ns/iter (± 0)	34 ns/iter (± 0)	3.47
lotr_graph_subgraph_10pc/iterate nodes	2613 ns/iter (± 88)	839 ns/iter (± 5)	3.11
lotr_graph_subgraph_10pc_materialise/materialize	1568204 ns/iter (± 13207)	334634 ns/iter (± 1287)	4.69
lotr_graph_subgraph_10pc_windowed/has_node_existing	139 ns/iter (± 7)	62 ns/iter (± 14)	2.24
lotr_graph_subgraph_10pc_windowed/iterate nodes	4930 ns/iter (± 97)	1365 ns/iter (± 3)	3.61
lotr_graph_subgraph_10pc_windowed_materialise/materialize	945183 ns/iter (± 10472)	230399 ns/iter (± 2617)	4.10
lotr_graph_window_50_layered/num_edges_temporal	154077 ns/iter (± 2292)	70121 ns/iter (± 7586)	2.20
lotr_graph_window_50_layered/num_nodes	58418 ns/iter (± 1169)	21435 ns/iter (± 536)	2.73
lotr_graph_window_50_layered/has_node_existing	786 ns/iter (± 74)	129 ns/iter (± 12)	6.09
lotr_graph_window_50_layered/max_id	65171 ns/iter (± 2846)	25556 ns/iter (± 252)	2.55
lotr_graph_window_50_layered/iterate nodes	122429 ns/iter (± 414)	19308 ns/iter (± 47)	6.34
lotr_graph_window_50_layered/iterate edges	197698 ns/iter (± 421)	83616 ns/iter (± 1318)	2.36
lotr_graph_window_50_layered/graph_latest	95150 ns/iter (± 1272)	36649 ns/iter (± 916)	2.60
lotr_graph_window_50_layered_materialise/materialize	27991179 ns/iter (± 69479)	3488825 ns/iter (± 24948)	8.02
lotr_graph_persistent_window_50_layered/num_edges_temporal	675615 ns/iter (± 13087)	192686 ns/iter (± 1569)	3.51
lotr_graph_persistent_window_50_layered/num_nodes	74334 ns/iter (± 1664)	31517 ns/iter (± 779)	2.36
lotr_graph_persistent_window_50_layered/has_node_existing	934 ns/iter (± 198)	174 ns/iter (± 83)	5.37
lotr_graph_persistent_window_50_layered/max_id	82114 ns/iter (± 1304)	38024 ns/iter (± 490)	2.16
lotr_graph_persistent_window_50_layered/iterate nodes	159392 ns/iter (± 415)	35886 ns/iter (± 191)	4.44
lotr_graph_persistent_window_50_layered/iterate edges	190222 ns/iter (± 580)	84161 ns/iter (± 596)	2.26
lotr_graph_persistent_window_50_layered/iterate_exploded_edges	4415761 ns/iter (± 9181)	1659940 ns/iter (± 19402)	2.66
lotr_graph_persistent_window_50_layered/graph_latest	141621 ns/iter (± 821)	57549 ns/iter (± 4809)	2.46
lotr_graph_persistent_window_50_layered_materialise/materialize	47382628 ns/iter (± 63586)	5298035 ns/iter (± 147912)	8.94
lotr_graph/proto_encode	8658464 ns/iter (± 64508)	1157897 ns/iter (± 73709)	7.48

This comment was automatically generated by workflow using github-action-benchmark.

[shivamka1]

test_permissions is moved to pometry_storage

[shivamka1]

Probably also makes sense to move the rbac prop tests as well to pometry-storage

[shivamka1]

The server is never explicitly stopped between iterations. _server is just dropped, which aborts the background task without sending a shutdown signal. The OS socket is left open, so the next iteration's start_server call might fail to bind port 43871 with "address already in use". The abort via drop is fast enough that the OS usually reclaims the socket before the next iteration tries to bind but it's a race condition. Under load, a slow CI machine, or with OS scheduling delays, the socket won't be released in time and the test will fail.

[shivamka1]

Probably, we reuse the same server for all iterations, cleaning up before next iteration or use OS-assigned port per iteration?

[shivamka1]

Also we don't seem to be cleaning up the _tempdir. Leaves directories on disk permanently.

[shivamka1]

missing:

 assert!(
    !can_introspect_namespace(path, children, client).unwrap(),
    "namespace {path} should not be introspectable"
);
assert!(
    !can_write_namespace(path, client).unwrap(),
    "namespace {path} should not be writable"
);

[shivamka1]

missing:

assert!(
    can_discover_namespace(path, client).unwrap(),
    "namespace {path} should be discoverable"
);

[shivamka1]

missing:

assert!(
    can_discover_namespace(path, client).unwrap(),
    "namespace {path} should be discoverable"
);

[shivamka1]

missing:

assert!(
    can_discover_namespace(path, client).unwrap(),
    "namespace {path} should be discoverable"
);

[shivamka1]

This makes debugging harder. When a test fails you no longer know which query caused the error. Worth keeping the query in the message, even if reformatted.