Merge branch 'main' into neilalexander/cfg

docs/administration/5_troubleshooting.md

@@ -0,0 +1,81 @@
+---
+title: Troubleshooting
+parent: Administration
+permalink: /administration/troubleshooting
+---
+
+# Troubleshooting
+
+If your Dendrite installation is acting strangely, there are a few things you should
+check before seeking help.
+
+## 1. Logs
+
+Dendrite, by default, will log all warnings and errors to stdout, in addition to any
+other locations configured in the `dendrite.yaml` configuration file. Often there will
+be clues in the logs.
+
+You can increase this log level to the more verbose `debug` level if necessary by adding
+this to the config and restarting Dendrite:
+
+```
+logging:
+- type: std
+  level: debug
+```
+
+Look specifically for lines that contain `level=error` or `level=warning`.
+
+## 2. Federation tester
+
+If you are experiencing problems federating with other homeservers, you should check
+that the [Federation Tester](https://federationtester.matrix.org) is passing for your
+server.
+
+Common reasons that it may not pass include:
+
+1. Incorrect DNS configuration;
+2. Misconfigured DNS SRV entries or well-known files;
+3. Invalid TLS/SSL certificates;
+4. Reverse proxy configuration issues (if applicable).
+
+Correct any errors if shown and re-run the federation tester to check the results.
+
+## 3. System time
+
+Matrix relies heavily on TLS which requires the system time to be correct. If the clock
+drifts then you may find that federation no works reliably (or at all) and clients may
+struggle to connect to your Dendrite server.
+
+Ensure that your system time is correct and consider syncing to a reliable NTP source.
+
+## 4. Database connections
+
+If you are using the PostgreSQL database, you should ensure that Dendrite's configured
+number of database connections does not exceed the maximum allowed by PostgreSQL.
+
+Open your `postgresql.conf` configuration file and check the value of `max_connections`
+(which is typically `100` by default). Then open your `dendrite.yaml` configuration file
+and ensure that:
+
+1. If you are using the `global.database` section, that `max_open_conns` does not exceed
+   that number;
+2. If you are **not** using the `global.database` section, that the sum total of all
+   `max_open_conns` across all `database` blocks does not exceed that number.
+
+## 5. File descriptors
+
+Dendrite requires a sufficient number of file descriptors for every connection it makes
+to a remote server, every connection to the database engine and every file it is reading
+or writing to at a given time (media, logs etc). We recommend ensuring that the limit is
+no lower than 65535 for Dendrite.
+
+Dendrite will check at startup if there are a sufficient number of available descriptors.
+If there aren't, you will see a log lines like this:
+
+```
+level=warning msg="IMPORTANT: Process file descriptor limit is currently 65535, it is recommended to raise the limit for Dendrite to at least 65535 to avoid issues"
+```
+
+Follow the [Optimisation](../installation/10_optimisation.md) instructions to correct the
+available number of file descriptors.

go.mod

@@ -25,7 +25,7 @@ require (
 	github.com/matrix-org/dugong v0.0.0-20210921133753-66e6b1c67e2e
 	github.com/matrix-org/go-sqlite3-js v0.0.0-20220419092513-28aa791a1c91
 	github.com/matrix-org/gomatrix v0.0.0-20210324163249-be2af5ef2e16
-	github.com/matrix-org/gomatrixserverlib v0.0.0-20220711125303-3bb2e997a44c
+	github.com/matrix-org/gomatrixserverlib v0.0.0-20220713083127-fc2ea1e62e46
 	github.com/matrix-org/pinecone v0.0.0-20220708135211-1ce778fcde6a
 	github.com/matrix-org/util v0.0.0-20200807132607-55161520e1d4
 	github.com/mattn/go-sqlite3 v1.14.13

go.sum

@@ -341,8 +341,8 @@ github.com/matrix-org/go-sqlite3-js v0.0.0-20220419092513-28aa791a1c91/go.mod h1
 github.com/matrix-org/gomatrix v0.0.0-20190528120928-7df988a63f26/go.mod h1:3fxX6gUjWyI/2Bt7J1OLhpCzOfO/bB3AiX0cJtEKud0=
 github.com/matrix-org/gomatrix v0.0.0-20210324163249-be2af5ef2e16 h1:ZtO5uywdd5dLDCud4r0r55eP4j9FuUNpl60Gmntcop4=
 github.com/matrix-org/gomatrix v0.0.0-20210324163249-be2af5ef2e16/go.mod h1:/gBX06Kw0exX1HrwmoBibFA98yBk/jxKpGVeyQbff+s=
-github.com/matrix-org/gomatrixserverlib v0.0.0-20220711125303-3bb2e997a44c h1:mt30TDK8kXKV+nCmVfnqoXsh842N+74kvZw7DXuS/JQ=
-github.com/matrix-org/gomatrixserverlib v0.0.0-20220711125303-3bb2e997a44c/go.mod h1:jX38yp3SSLJNftBg3PXU1ayd0PCLIiDHQ4xAc9DIixk=
+github.com/matrix-org/gomatrixserverlib v0.0.0-20220713083127-fc2ea1e62e46 h1:5X/kXY3nwqKOwwrE9tnMKrjbsi3PHigQYvrvDBSntO8=
+github.com/matrix-org/gomatrixserverlib v0.0.0-20220713083127-fc2ea1e62e46/go.mod h1:jX38yp3SSLJNftBg3PXU1ayd0PCLIiDHQ4xAc9DIixk=
 github.com/matrix-org/pinecone v0.0.0-20220708135211-1ce778fcde6a h1:DdG8vXMlZ65EAtc4V+3t7zHZ2Gqs24pSnyXS+4BRHUs=
 github.com/matrix-org/pinecone v0.0.0-20220708135211-1ce778fcde6a/go.mod h1:ulJzsVOTssIVp1j/m5eI//4VpAGDkMt5NrRuAVX7wpc=
 github.com/matrix-org/util v0.0.0-20190711121626-527ce5ddefc7/go.mod h1:vVQlW/emklohkZnOPwD3LrZUBqdfsbiyO3p1lNV8F6U=

internal/caching/cache_roominfo.go

@@ -16,18 +16,18 @@ import (
 // a room Info cache. It must only be used from the roomserver only
 // It is not safe for use from other components.
 type RoomInfoCache interface {
-	GetRoomInfo(roomID string) (roomInfo types.RoomInfo, ok bool)
-	StoreRoomInfo(roomID string, roomInfo types.RoomInfo)
+	GetRoomInfo(roomID string) (roomInfo *types.RoomInfo, ok bool)
+	StoreRoomInfo(roomID string, roomInfo *types.RoomInfo)
 }
 
 // GetRoomInfo must only be called from the roomserver only. It is not
 // safe for use from other components.
-func (c Caches) GetRoomInfo(roomID string) (types.RoomInfo, bool) {
+func (c Caches) GetRoomInfo(roomID string) (*types.RoomInfo, bool) {
 	return c.RoomInfos.Get(roomID)
 }
 
 // StoreRoomInfo must only be called from the roomserver only. It is not
 // safe for use from other components.
-func (c Caches) StoreRoomInfo(roomID string, roomInfo types.RoomInfo) {
+func (c Caches) StoreRoomInfo(roomID string, roomInfo *types.RoomInfo) {
 	c.RoomInfos.Set(roomID, roomInfo)
 }

internal/caching/caches.go

@@ -28,7 +28,7 @@ type Caches struct {
 	RoomServerRoomNIDs Cache[string, types.RoomNID]                           // room ID -> room NID
 	RoomServerRoomIDs  Cache[int64, string]                                   // room NID -> room ID
 	RoomServerEvents   Cache[int64, *gomatrixserverlib.Event]                 // event NID -> event
-	RoomInfos          Cache[string, types.RoomInfo]                          // room ID -> room info
+	RoomInfos          Cache[string, *types.RoomInfo]                         // room ID -> room info
 	FederationPDUs     Cache[int64, *gomatrixserverlib.HeaderedEvent]         // queue NID -> PDU
 	FederationEDUs     Cache[int64, *gomatrixserverlib.EDU]                   // queue NID -> EDU
 	SpaceSummaryRooms  Cache[string, gomatrixserverlib.MSC2946SpacesResponse] // room ID -> space response

internal/caching/impl_ristretto.go

@@ -100,7 +100,7 @@ func NewRistrettoCache(maxCost config.DataUnit, maxAge time.Duration, enableProm
 				MaxAge: maxAge,
 			},
 		},
-		RoomInfos: &RistrettoCachePartition[string, types.RoomInfo]{ // room ID -> room info
+		RoomInfos: &RistrettoCachePartition[string, *types.RoomInfo]{ // room ID -> room info
 			cache:   cache,
 			Prefix:  roomInfosCache,
 			Mutable: true,

roomserver/storage/shared/room_updater.go

@@ -225,13 +225,12 @@ func (u *RoomUpdater) SetLatestEvents(
 		if err := u.d.RoomsTable.UpdateLatestEventNIDs(u.ctx, txn, roomNID, eventNIDs, lastEventNIDSent, currentStateSnapshotNID); err != nil {
 			return fmt.Errorf("u.d.RoomsTable.updateLatestEventNIDs: %w", err)
 		}
-		if roomID, ok := u.d.Cache.GetRoomServerRoomID(roomNID); ok {
-			if roomInfo, ok := u.d.Cache.GetRoomInfo(roomID); ok {
-				roomInfo.StateSnapshotNID = currentStateSnapshotNID
-				roomInfo.IsStub = false
-				u.d.Cache.StoreRoomInfo(roomID, roomInfo)
-			}
-		}
+
+		// Since it's entirely possible that this types.RoomInfo came from the
+		// cache, we should make sure to update that entry so that the next run
+		// works from live data.
+		u.roomInfo.StateSnapshotNID = currentStateSnapshotNID
+		u.roomInfo.IsStub = false
 		return nil
 	})
 }

roomserver/storage/shared/storage.go

@@ -139,13 +139,13 @@ func (d *Database) RoomInfo(ctx context.Context, roomID string) (*types.RoomInfo
 }
 
 func (d *Database) roomInfo(ctx context.Context, txn *sql.Tx, roomID string) (*types.RoomInfo, error) {
-	if roomInfo, ok := d.Cache.GetRoomInfo(roomID); ok {
-		return &roomInfo, nil
+	if roomInfo, ok := d.Cache.GetRoomInfo(roomID); ok && roomInfo != nil {
+		return roomInfo, nil
 	}
 	roomInfo, err := d.RoomsTable.SelectRoomInfo(ctx, txn, roomID)
 	if err == nil && roomInfo != nil {
 		d.Cache.StoreRoomServerRoomID(roomInfo.RoomNID, roomID)
-		d.Cache.StoreRoomInfo(roomID, *roomInfo)
+		d.Cache.StoreRoomInfo(roomID, roomInfo)
 	}
 	return roomInfo, err
 }

sytest-whitelist

@@ -719,4 +719,5 @@ registration is idempotent, with username specified
 Setting state twice is idempotent
 Joining room twice is idempotent
 Inbound federation can return missing events for shared visibility
-Inbound federation ignores redactions from invalid servers room > v3
+Inbound federation ignores redactions from invalid servers room > v3
+Newly joined room includes presence in incremental sync