Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.

81279: storage: specify and test iterator visibility semantics r=jbowens a=erikgrinaker

`@jbowens` I split this out from the range key PR in #77417, since we'll need to audit/update existing code for the new point key batch semantics, and it seems cleaner/easier to do this separately. As you know, the test currently fails because existing batch iterators see concurrent batch writes.

This only covers point keys, but range keys should have identical semantics, and the tests will be extended for range keys in #77417.

---

This patch specifies and tests iterator visibility semantics. See
comment on `Engine.NewMVCCIterator` for details.

Release note: None

81403: kvserver: address assorted comments from previous refactor of `TransferLeaseTarget()` r=aayushshah15 a=aayushshah15

This commit addresses miscelleneous review comments from #80834.

Release note: None

81679: sql: enforce nullability when evaluating expressions in index backfills r=ajwerner a=ajwerner

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until #76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in #81675.

81777: util/metric: delete Rate r=andreimatei a=andreimatei

This exponentially moving Rate was unused. It also was exporting to
Prometheus as a gauge. It seems unlikely that we'll use this in the
future; I think we're happy enough with counters and computing rates in
the monitoring system. The Rate seems to have come from a time when we
liked moving measurements more; since then, we've adapted the Histogram,
for example, to include non-windowed state.

Release note: None

81798: sql: enable streamer for lookup joins r=yuzefovich a=yuzefovich

We've just fixed an issue with incorrectly using leaf txns for mutations
because of the streamer, so I think it is ok to fully enable the
streamer. I expect that there might be some fallout, so I'm curious to
see the run of nightlies.

Release note: None

81804: build: use bazel for roachtest stress r=rickystewart a=nicktrav

Currently, the Roachtest Stress TeamCity job uses `make` (via the
`mkrelease.sh` script) for building the required binaries. Recently, the
pipeline started hanging while building the binary. Rather than
investigating and fixing the existing `make`-based pipeline, convert the
job to using Bazel for building the binaries.

Retain the existing entrypoint to the job, converting it to use [the
existing pattern][1] of calling `run_bazel` with the required
environment variables, invoking an `_impl.sh` script. The existing logic
is moved into the new `_impl.sh` script, with some minor additions to
source some additional scripts.

Additionally, the TeamCity job itself was updated to set pass in the
`TARGET_CLOUD` environment variable (set to `gce`), and remove the need
to build within a container (instead deferring to Bazel to manage the
build environment). The diff for the TeamCity pipeline is as follows
(for posterity):

```diff
18c18
<
---
>       <param name="env.TARGET_CLOUD" value="gce" />
24c24
<           <param name="plugin.docker.imageId" value="%builder.dockerImage%" />
---
>
28,29c28,29
< export BUILD_TAG="$(git describe --abbrev=0 --tags --match=v[0-9]*)"
< ./build/teamcity-roachtest-stress.sh]]></param>
---
> build_tag="$(git describe --abbrev=0 --tags --match=v[0-9]*)"
> CLOUD=${TARGET_CLOUD} BUILD_TAG="${build_tag}" ./build/teamcity-roachtest-stress.sh]]></param>
```

Release note: None.

[1]: https://github.com/cockroachdb/cockroach/blob/12198a51408e7333cd4f96b221e6734239479765/build/teamcity/cockroach/nightlies/roachtest_nightly_gce.sh#L10-L11

81813: liveness: run sync disk write in a stopper task r=stevendanna,nicktrav a=erikgrinaker

**liveness: move stopper to `NodeLivenessOptions`**

Release note: None

**liveness: run sync disk write in a stopper task**

This patch runs the sync disk write during node heartbeats in a stopper
task. The write is done in a goroutine, so that we can respect the
caller's context cancellation (even though the write itself won't).
However, this could race with engine shutdown when stopping the node,
violating the Pebble contract and triggering the race detector.  Running
it as a stopper task will cause the node to wait for the disk write to
complete before closing the engine.

Of course, if the disk stalls then node shutdown will now never
complete. This is very unfortunate, since stopping the node is often
the only mitigation to recover stuck ranges with stalled disks. This is
mitigated by Pebble panic'ing the node on stalled disks, and Kubernetes
and other orchestration tools killing the process after some time.

Touches #81786.
Resolves #81511.
Resolves #81827.

Release note: None

81828: sql: deflake TestPrimaryKeyDropIndexNotCancelable r=chengxiong-ruan a=ajwerner

The problem was that we weren't atomically setting the boolean, at least, I
think. I repro'd another flake of it pretty quickly.

Release note: None

Co-authored-by: Erik Grinaker <grinaker@cockroachlabs.com>
Co-authored-by: Aayush Shah <aayush.shah15@gmail.com>
Co-authored-by: Andrew Werner <awerner32@gmail.com>
Co-authored-by: Andrei Matei <andrei@cockroachlabs.com>
Co-authored-by: Yahor Yuzefovich <yahor@cockroachlabs.com>
Co-authored-by: Nick Travers <travers@cockroachlabs.com>
Co-authored-by: Andrew Werner <ajwerner@cockroachlabs.com>

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.

Prior to this change, we never enforced nullability of values when performing
index backfills. Up until cockroachdb#76983, all column additions used the legacy schema
changer and its column backfill protocol. The column backfiller indeed checks
on nullability violations for columns [here](https://github.com/cockroachdb/cockroach/blob/d5313438aaa61a71a5a66c5718963bbb0fd7268b/pkg/sql/backfill/backfill.go#L355-L357).

Release note (bug fix): Prior to this change, virtual computed columns which
were marked as NOT NULL could be added to new secondary index. After this
change, attempts to add such columns to a secondary index will result in an
error. Note that such invalid columns can still be added to tables. Work to
resolve that bug is tracked in cockroachdb#81675.