From c88f5a36d0f651b72f09294b8b35500172e52590 Mon Sep 17 00:00:00 2001
From: Ben Ye <benye@amazon.com>
Date: Fri, 10 Mar 2023 08:16:45 -0800
Subject: [PATCH] disable absent and absent_over_time for vertical sharding
 (#6201)

---
 CHANGELOG.md                       |  1 +
 pkg/querysharding/analyzer.go      | 18 +++++++++++++++++-
 pkg/querysharding/analyzer_test.go |  8 ++++++++
 3 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 194a6dc29b..964a3fd328 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -24,6 +24,7 @@ We use *breaking :warning:* to mark changes that are not backward compatible (re
 
 ### Changed
 - [#6168](https://github.com/thanos-io/thanos/pull/6168) Receiver: Make ketama hashring fail early when configured with number of nodes lower than the replication factor.
+- [#6201](https://github.com/thanos-io/thanos/pull/6201) Query-Frontend: Disable absent and absent_over_time for vertical sharding.
 
 ### Removed
 
diff --git a/pkg/querysharding/analyzer.go b/pkg/querysharding/analyzer.go
index dc8b16ca72..37a0d30296 100644
--- a/pkg/querysharding/analyzer.go
+++ b/pkg/querysharding/analyzer.go
@@ -17,11 +17,17 @@
 package querysharding
 
 import (
+	"fmt"
+
 	lru "github.com/hashicorp/golang-lru"
 	"github.com/prometheus/common/model"
 	"github.com/prometheus/prometheus/promql/parser"
 )
 
+var (
+	notShardableErr = fmt.Errorf("expressions are not shardable")
+)
+
 type Analyzer interface {
 	Analyze(string) (QueryAnalysis, error)
 }
@@ -72,7 +78,9 @@ func (a *CachedQueryAnalyzer) Analyze(query string) (QueryAnalysis, error) {
 
 // Analyze uses the following algorithm:
 //   - if a query has functions which cannot be sharded such as
-//     label_join or label_replace, then treat the query as non shardable.
+//     absent or absent_over_time, then treat the query as non shardable.
+//   - if a query has functions `label_join` or `label_replace`,
+//     calculate the shard labels based on grouping labels.
 //   - Walk the query and find the least common labelset
 //     used in grouping expressions. If non-empty, treat the query
 //     as shardable by those labels.
@@ -89,6 +97,7 @@ func (a *QueryAnalyzer) Analyze(query string) (QueryAnalysis, error) {
 		analysis      QueryAnalysis
 		dynamicLabels []string
 	)
+	isShardable := true
 	parser.Inspect(expr, func(node parser.Node, nodes []parser.Node) error {
 		switch n := node.(type) {
 		case *parser.Call:
@@ -96,6 +105,9 @@ func (a *QueryAnalyzer) Analyze(query string) (QueryAnalysis, error) {
 				if n.Func.Name == "label_join" || n.Func.Name == "label_replace" {
 					dstLabel := stringFromArg(n.Args[1])
 					dynamicLabels = append(dynamicLabels, dstLabel)
+				} else if n.Func.Name == "absent_over_time" || n.Func.Name == "absent" {
+					isShardable = false
+					return notShardableErr
 				}
 			}
 		case *parser.BinaryExpr:
@@ -117,6 +129,10 @@ func (a *QueryAnalyzer) Analyze(query string) (QueryAnalysis, error) {
 		return nil
 	})
 
+	if !isShardable {
+		return nonShardableQuery(), nil
+	}
+
 	// If currently it is shard by, it is still shardable if there is
 	// any label left after removing the dynamic labels.
 	// If currently it is shard without, it is still shardable if we
diff --git a/pkg/querysharding/analyzer_test.go b/pkg/querysharding/analyzer_test.go
index b13e805667..6faa6e839f 100644
--- a/pkg/querysharding/analyzer_test.go
+++ b/pkg/querysharding/analyzer_test.go
@@ -67,6 +67,14 @@ http_requests_total`,
 			name:       "aggregate by expression with label_join, sharding label is dynamic",
 			expression: `sum by (dst_label) (label_join(metric, "dst_label", ",", "src_label"))`,
 		},
+		{
+			name:       "absent_over_time is not shardable",
+			expression: `sum by (url) (absent_over_time(http_requests_total{code="400"}[5m]))`,
+		},
+		{
+			name:       "absent is not shardable",
+			expression: `sum by (url) (absent(http_requests_total{code="400"}))`,
+		},
 	}
 
 	shardableByLabels := []testCase{