From 9afc27994cb613a46c3d040742b3ea83665c9989 Mon Sep 17 00:00:00 2001
From: Jasmina Malicevic <jasmina.dustinac@gmail.com>
Date: Tue, 20 Feb 2024 09:22:09 +0100
Subject: [PATCH] fix(e2e): Reduce flakiness of TestGRPC_GetBlockResults
 (#2367)

The test is failing often due to the fact that it asks for the first
block via RPC Status then waits to be notified about the latest height
via gRPC and then requests block results at the first and last heights.
If pruning is happening fast enough, there is a high chance the results
will be pruned at the time of asking.

This fix is not making the behaviour fully deterministic but increases
the chances the block is still around because it sets the first height
to be the last reported first height + retain blocks.

<!--

Please add a reference to the issue that this PR addresses and indicate
which
files are most critical to review. If it fully addresses a particular
issue,
please include "Closes #XXX" (where "XXX" is the issue number).

If this PR is non-trivial/large/complex, please ensure that you have
either
created an issue that the team's had a chance to respond to, or had some
discussion with the team prior to submitting substantial pull requests.
The team
can be reached via GitHub Discussions or the Cosmos Network Discord
server in
the #cometbft channel. GitHub Discussions is preferred over Discord as
it
allows us to keep track of conversations topically.
https://github.com/cometbft/cometbft/discussions

If the work in this PR is not aligned with the team's current
priorities, please
be advised that it may take some time before it is merged - especially
if it has
not yet been discussed with the team.

See the project board for the team's current priorities:
https://github.com/orgs/cometbft/projects/1

-->

---

#### PR checklist

- [ ] Tests written/updated
- [ ] Changelog entry added in `.changelog` (we use
[unclog](https://github.com/informalsystems/unclog) to manage our
changelog)
- [ ] Updated relevant documentation (`docs/` or `spec/`) and code
comments
- [ ] Title follows the [Conventional
Commits](https://www.conventionalcommits.org/en/v1.0.0/) spec

(cherry picked from commit 2c6e0c160da65cc8eece50cea4cc25043365b85f)
---
 test/e2e/tests/block_test.go     |  8 +++++++-
 test/e2e/tests/grpc_test.go      | 10 ++++++++--
 test/e2e/tests/validator_test.go |  8 +++++++-
 3 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/test/e2e/tests/block_test.go b/test/e2e/tests/block_test.go
index 14741a35c9d..3c1c52675ad 100644
--- a/test/e2e/tests/block_test.go
+++ b/test/e2e/tests/block_test.go
@@ -27,7 +27,13 @@ func TestBlock_Header(t *testing.T) {
 		first := status.SyncInfo.EarliestBlockHeight
 		last := status.SyncInfo.LatestBlockHeight
 		if node.RetainBlocks > 0 {
-			first++ // avoid race conditions with block pruning
+			// This was done in case pruning is activated.
+			// As it happens in the background this lowers the chances
+			// that the block at height=first will be pruned by the time we test
+			// this. If this test starts to fail often, it is worth revisiting this logic.
+			// To reproduce this failure locally, it is advised to set the storage.pruning.interval
+			// to 1s instead of 10s.
+			first += int64(node.RetainBlocks) // avoid race conditions with block pruning
 		}
 
 		for _, block := range blocks {
diff --git a/test/e2e/tests/grpc_test.go b/test/e2e/tests/grpc_test.go
index cb797876165..5be0b1c7620 100644
--- a/test/e2e/tests/grpc_test.go
+++ b/test/e2e/tests/grpc_test.go
@@ -106,7 +106,13 @@ func TestGRPC_GetBlockResults(t *testing.T) {
 		first := status.SyncInfo.EarliestBlockHeight
 		last := status.SyncInfo.LatestBlockHeight
 		if node.RetainBlocks > 0 {
-			first++
+			// This was done in case pruning is activated.
+			// As it happens in the background this lowers the chances
+			// that the block at height=first will be pruned by the time we test
+			// this. If this test starts to fail often, it is worth revisiting this logic.
+			// To reproduce this failure locally, it is advised to set the storage.pruning.interval
+			// to 1s instead of 10s.
+			first += int64(node.RetainBlocks)
 		}
 
 		ctx, ctxCancel := context.WithTimeout(context.Background(), time.Minute)
@@ -136,7 +142,7 @@ func TestGRPC_GetBlockResults(t *testing.T) {
 		errorCases := []struct {
 			requestHeight int64
 		}{
-			{first - 2},
+			{first - int64(node.RetainBlocks) - 2},
 			{last + 100000},
 		}
 
diff --git a/test/e2e/tests/validator_test.go b/test/e2e/tests/validator_test.go
index 1d75bc20060..3b7887b03cf 100644
--- a/test/e2e/tests/validator_test.go
+++ b/test/e2e/tests/validator_test.go
@@ -30,7 +30,13 @@ func TestValidator_Sets(t *testing.T) {
 
 		// skip first block if node is pruning blocks, to avoid race conditions
 		if node.RetainBlocks > 0 {
-			first++
+			// This was done in case pruning is activated.
+			// As it happens in the background this lowers the chances
+			// that the block at height=first will be pruned by the time we test
+			// this. If this test starts to fail often, it is worth revisiting this logic.
+			// To reproduce this failure locally, it is advised to set the storage.pruning.interval
+			// to 1s instead of 10s.
+			first += int64(node.RetainBlocks)
 		}
 
 		valSchedule := newValidatorSchedule(*node.Testnet)