spanconfig: enable range coalescing by default

[irfansharif]

Fixes #81008.

We built the basic infrastructure to coalesce ranges across table boundaries back in 22.2 as part of #66063. We've enabled this optimization for secondary tenants since then, but hadn't for the system tenant because of two primary blockers:

• sql: SHOW RANGES broken by coalesced ranges #93617: SHOW RANGES was broken by coalesced ranges.
• kvserver,server,sql: provide efficient mechanism to retrieve data size information for span #84105: APIs to compute sizes for schema objects (used in our UI) was broken by coalesced ranges.

In both these cases we baked in assumptions about there being a minimum of one-{table,index,partition}-per-range. These blockers didn't apply to secondary tenants at the time since they didn't have access to SHOW RANGES, nor the UI pages where these schema statistics were displayed.

We've addressed both these blockers in the 23.1 cycle as part of bridging the compatibility between secondary tenants and yesteryear's system tenant.

• sql: better SHOW RANGES and changes to crdb_internal.ranges{,_no_leases} #93644 revised SHOW RANGES and crdb_internal.ranges{,_no_leases}, both internally and its external UX, to accommodate ranges straddling table/database boundaries.
• kvserver, server: new implementation of SpanStats suitable for use with coalesced ranges #96223 re-worked our SpanStats API to work in the face of coalesced ranges, addressing kvserver,server,sql: provide efficient mechanism to retrieve data size information for span #84105.

Release note (general change): CockroachDB would previously use separate ranges for each table, index, or partition. This is no longer true -- it's possible now to have multiple tables, indexes, and partitions to get packed into the same range. For users with many such "schema objects", this will reduce the total range count in their clusters. This is especially true if individual tables, indexes, or partitions are smaller than the default configured maximum range size (controlled using zone configs, specifically the range_max_bytes parameter). We made this change to improve scalability with respect to the number of schema objects, since the underlying range count now no longer a bottleneck. Users upgrading from 22.2, once finalizing their upgrade, may observe a round of range merges and snapshot transfers (to power said range merges) as a result of this change. If users want to opt-out of this optimization, they can use the following cluster setting:

  SET CLUSTER SETTING spanconfig.storage_coalesce_adjacent.enabled = false; 

[cockroach-teamcity]

This change is 

[irfansharif]

Hm, looks like #97942 is another dependency. Our data distribution page in the advanced debug page is making the same minimum-one-table-per-range assumption here (+cc @dhartunian). Womp womp.

cockroach/pkg/server/admin.go

Line 3081 in b6e0f76

_, tableID, err := keys.MakeSQLCodec(tenID).DecodeTablePrefix(rangeDesc.StartKey.AsRawKey())

[knz]

For UX we want #97858 (already being worked on) and #97861 too.

[irfansharif]

CI failure is #98555.

[irfansharif]

Discussed offline with David H. + Alex B., who are ok with just the following warning in the advanced debug page.

@arulajmani, can I get a review?

[knz]

Is there a tracking issue to fix the data distribution screen or remove it outright?

[irfansharif]

There was #97942, but the Obs-Inf folks closed it. I'll defer to them if they want something to fix this and/or removing it outright.

[arulajmani]

Reviewed 10 of 11 files at r1, all commit messages.
Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on @ajwerner and @irfansharif)

---

pkg/spanconfig/spanconfigstore/span_store.go line 159 at r1 (raw file):

			if roachpb.Key(rem).Compare(systemTableUpperBound) < 0 ||
				!(StorageCoalesceAdjacentSetting.Get(&s.settings.SV) &&
					s.settings.Version.IsActive(ctx, clusterversion.V23_2_EnableRangeCoalescingForSystemTenant)) {

Seems like the range coalescing cluster setting would be a no-op in 23.1 <-> 23.2 mixed version clusters; is the thinking that this is okay because it was a hidden cluster setting previously?

[irfansharif]

TFTR.

bors r+

Reviewable status: complete! 0 of 0 LGTMs obtained (and 1 stale) (waiting on @ajwerner, @arulajmani, and @knz)

---

pkg/spanconfig/spanconfigstore/span_store.go line 36 at r1 (raw file):

Previously, knz (Raphael 'kena' Poss) wrote…

Can you make this one public too then thanks

Done.

---

pkg/spanconfig/spanconfigstore/span_store.go line 159 at r1 (raw file):

Previously, arulajmani (Arul Ajmani) wrote…

Seems like the range coalescing cluster setting would be a no-op in 23.1 <-> 23.2 mixed version clusters; is the thinking that this is okay because it was a hidden cluster setting previously?

Yes.

[craig]

Build succeeded:

• Bazel Essential CI (Cockroach)