Starting using weaveworks common fork (#6386)

* Starting using weaveworks common fork

Signed-off-by: alanprot <[email protected]>

* update weavewrworks

Signed-off-by: alanprot <[email protected]>

* bring weaveworks common with the workers change

Signed-off-by: alanprot <[email protected]>

* Changelog

Signed-off-by: alanprot <[email protected]>

---------

Signed-off-by: alanprot <[email protected]>

Author: alanprot

CHANGELOG.md

@@ -41,6 +41,7 @@
 * [ENHANCEMENT] StoreGateway: Add new `cortex_bucket_store_chunk_pool_inuse_bytes` metric to track the usage in chunk pool. #6310
 * [ENHANCEMENT] Distributor: Add new `cortex_distributor_inflight_client_requests` metric to track number of ingester client inflight requests. #6358
 * [ENHANCEMENT] Distributor: Expose `cortex_label_size_bytes` native histogram metric. #6372
+* [ENHANCEMENT] Add new option `-server.grpc_server-num-stream-workers` to configure the number of worker goroutines that should be used to process incoming streams. #6386
 * [BUGFIX] Runtime-config: Handle absolute file paths when working directory is not / #6224
 * [BUGFIX] Ruler: Allow rule evaluation to complete during shutdown. #6326
 * [BUGFIX] Ring: update ring with new ip address when instance is lost, rejoins, but heartbeat is disabled  #6271

docs/configuration/config-file-reference.md

@@ -5417,6 +5417,12 @@ grpc_tls_config:
 # CLI flag: -server.grpc-max-concurrent-streams
 [grpc_server_max_concurrent_streams: <int> | default = 100]
 
+# Number of worker goroutines that should be used to process incoming
+# streams.Setting this 0 (default) will disable workers and spawn a new
+# goroutine for each stream.
+# CLI flag: -server.grpc_server-num-stream-workers
+[grpc_server_num_stream_workers: <int> | default = 0]
+
 # The duration after which an idle connection should be closed. Default:
 # infinity
 # CLI flag: -server.grpc.keepalive.max-connection-idle

go.mod

@@ -80,7 +80,7 @@ require (
 	github.com/bboreham/go-loser v0.0.0-20230920113527-fcc2c21820a3
 	github.com/cespare/xxhash/v2 v2.3.0
 	github.com/google/go-cmp v0.6.0
-	github.com/sercand/kuberesolver/v4 v4.0.0
+	github.com/sercand/kuberesolver/v5 v5.1.1
 	go.opentelemetry.io/collector/pdata v1.20.0
 	golang.org/x/exp v0.0.0-20240613232115-7f521ea00fb8
 	google.golang.org/protobuf v1.35.2
@@ -246,6 +246,9 @@ require (
 	k8s.io/utils v0.0.0-20240711033017-18e509b52bc8 // indirect
 )
 
+// Using cortex fork of weaveworks/common
+replace github.com/weaveworks/common => github.com/cortexproject/weaveworks-common v0.0.0-20241129212437-96019edf21f1
+
 // Override since git.apache.org is down.  The docs say to fetch from github.
 replace git.apache.org/thrift.git => github.com/apache/thrift v0.0.0-20180902110319-2566ecd5d999
 
@@ -270,8 +273,5 @@ replace github.com/google/gnostic => github.com/googleapis/gnostic v0.6.9
 // https://github.com/thanos-io/thanos/blob/fdeea3917591fc363a329cbe23af37c6fff0b5f0/go.mod#L265
 replace gopkg.in/alecthomas/kingpin.v2 => github.com/alecthomas/kingpin v1.3.8-0.20210301060133-17f40c25f497
 
-// Pin kuberesolver/v5 to support new grpc version. Need to upgrade kuberesolver version on weaveworks/common.
-replace github.com/sercand/kuberesolver/v4 => github.com/sercand/kuberesolver/v5 v5.1.1
-
 // gRPC 1.66 introduced memory pooling which breaks Cortex queries. Pin 1.65.0 until we have a fix.
 replace google.golang.org/grpc => google.golang.org/grpc v1.65.0

go.sum

@@ -8,7 +8,7 @@ import (
 
 	"github.com/go-kit/log/level"
 	"github.com/pkg/errors"
-	"github.com/sercand/kuberesolver/v4"
+	"github.com/sercand/kuberesolver/v5"
 	"github.com/weaveworks/common/tracing"
 	"go.opentelemetry.io/contrib/propagators/aws/xray"
 	"google.golang.org/grpc/credentials"