testcluster: assign unique ClusterName

[pav-kv]

This commit makes TestCluster by default assign a unique cluster name in the TestServerArgs. The cluster name is shared by all participating or added nodes, unless overridden in the per-node args.

This helps preventing accidental message exchange between TestClusters in the same environment that use the same TCP port in close proximity from each other.

Addresses #157838

[cockroach-teamcity]

This change is 

[pav-kv]

Fixed all the pkg/cli tests, but TestPartialZip/partial2 still resists. It decommissions node 2 and expects this node to not pop up in the debug.zip, but for some reason it still does. So far I don't understand how the change in this PR makes it behave differently.

Here, this node is visible as NODE_STATUS_DEAD, and I take it the test waits until it becomes NODE_STATUS_DECOMMISSIONED and doesn't pop up in the debug.zip.

cockroach/pkg/cli/zip_per_node.go

Line 62 in 21eed7b

if liveness == livenesspb.NodeLivenessStatus_DECOMMISSIONED {

Which must eventually happen because of the decommission command:

cockroach/pkg/cli/zip_test.go

Lines 766 to 775 in 21eed7b

	// Now mark the stopped node as decommissioned, and check that zip
	// skips over it automatically. We specifically use --wait=none because
	// we're decommissioning a node in a 3-node cluster, so there's no node to
	// up-replicate the under-replicated ranges to.
	{
	_, err := c.RunWithCapture(fmt.Sprintf("node decommission --checks=skip --wait=none %d", 2))
	if err != nil {
	t.Fatal(err)
	}
	}

[pav-kv]

Ah, probably need to set --cluster-name on the decommission command too.
Yes, fixed.

[pav-kv]

TFTRs!

bors r=RaduBerinde,stevendanna

[craig]

Build succeeded:

• linux_amd64_build
• lint
• docker_image_amd64
• acceptance
• examples_orms
• check_generated_code
• linux_amd64_fips_build
• local_roachtest
• local_roachtest_fips
• unit_tests

[blathers-crl]

Encountered an error creating backports. Some common things that can go wrong:

1. The backport branch might have already existed.
2. There was a merge conflict.
3. The backport branch contained merge commits.

You might need to create your backport manually using the backport tool.

---

💡 Consider backporting to the fork repo instead of the main repo. See instructions for more details.

error creating merge commit from 1757367 to blathers/backport-release-24.3-157868: POST https://api.github.com/repos/cockroachdb/cockroach/merges: 409 Merge conflict []

you may need to manually resolve merge conflicts with the backport tool.

Backport to branch release-24.3 failed. See errors above.

---

💡 Consider backporting to the fork repo instead of the main repo. See instructions for more details.

error creating merge commit from 1757367 to blathers/backport-release-25.2-157868: POST https://api.github.com/repos/cockroachdb/cockroach/merges: 409 Merge conflict []

you may need to manually resolve merge conflicts with the backport tool.

Backport to branch release-25.2 failed. See errors above.

---

💡 Consider backporting to the fork repo instead of the main repo. See instructions for more details.

💡 Consider backporting to the fork repo instead of the main repo. See instructions for more details.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is dev-inf.}

[srosenberg]

There is recent evidence to suggest rand.Uint32 isn't sufficiently unique; collision probability is high after thousands of iterations [1].

[1] #157919 (comment)

[pav-kv]

Good point. I'll fix up.