mirror of
https://github.com/seaweedfs/seaweedfs.git
synced 2026-09-10 16:40:46 +02:00
* expose whether a volume replica is backed by remote storage
Volume locations returned by lookups do not indicate whether a replica
has been tiered to remote storage. Readers cannot distinguish a local
replica from a remote-backed one, so they may hit a remote-backed
replica first even when a local replica is available.
Add DataInRemote to the lookup location message, populate it from the
master's volume info, and carry it through the wdclient vid map so
clients can prefer local replicas when resolving chunk locations.
* wdclient: prefer local volume replicas over remote-tier replicas on lookup
LookupFileIdWithFallback (and the publicUrl variant in FilerClient)
didn't honor the DataInRemote flag when shuffling URLs, so the
DataInRemote patch only took effect in LookupVolumeServerUrl. Apply
the same ReorderToFront(localUrls) to sameDcUrls/otherDcUrls so
non-remote replicas stay at the front, matching the existing vidMap
convention.
* wdclient: propagate DataInRemote across tier transitions on existing replicas
When a volume is tiered to remote storage or a remote-backed replica is
restored locally, the cached DataInRemote on the same volume-server URL
stayed at its old value because two pieces of state never updated:
* master_grpc_server.go only split newVolumes and (already-tracked) volumes
into NewVids vs RemoteVids. ChangedVolumes went straight to NewVids, so
the broadcast announced the re-classified volume as a fresh arrival and
the client had no way to tell whether its existing cache was stale.
* vid_map.addLocationToMap early-returned when an entry already had the
same URL. A tier transition reports the same URL with DataInRemote
flipped, so the cached entry stayed at the old classification.
Wire both sides together: ChangedVolumes now go through the same IsRemote
split as newVolumes, and addLocationToMap replaces the existing entry in
place when the URL matches but DataInRemote has changed. The server
reference key only depends on URL/grpc port, so the refcount does not
move across the flip.
Adds vid_map_remote_transition_test.go covering the local->remote and
remote->local paths so the in-place update and the cache-key stability
are pinned by tests.
* wdclient: prefer local replicas across data-center boundaries
The previous local-first ordering hoisted local URLs to the front of each
data-center bucket separately, then concatenated same-DC before other-DC.
That meant a same-DC remote replica could still be tried before an
other-DC local replica even though the local one would answer cheaply.
Reorder once across the full candidate list: concatenate same-DC and
other-DC first, then ReorderToFront pulls every local replica to the very
front while preserving the DC preference inside each tier. Apply the same
ordering in all four lookup paths so the cached vidMap, the
LookupFileIdWithFallback provider path, FilerClient.GetLookupFileIdFunction
(PublicUrl-preferred variant), and the deprecated filer.LookupFn all agree:
- weed/wdclient/vid_map.go (LookupVolumeServerUrl)
- weed/wdclient/vidmap_client.go (LookupFileIdWithFallback)
- weed/wdclient/filer_client.go (LookupFileId)
- weed/filer/reader_at.go (LookupFn)
Strengthen the existing local-first tests: vidmap_client_localfirst_test
now asserts both endpoints are present (not just the local one is first),
and slice_test asserts an exact match instead of accepting two orderings.
Add TestLookupFileIdWithFallbackGlobalLocalFirst to pin the cross-DC
ordering invariant: any local replica (same or other DC) precedes every
remote-tier replica; within each tier DC1 precedes DC2.
Add docstrings to ToVolumeLocations, ReorderToFront, LookupVolumeServerUrl,
LookupFileId, GetVidLocations, GetLocations, LookupFileIdWithFallback, and
updateVidMap so the touched lookup paths are described in one place.
* topology: broadcast tier transitions on existing replicas
When a volume replica is tiered to remote storage or restored locally, the
wdclient's cached DataInRemote went stale: every connected client kept
preferring a remote-backed replica over a freshly restored local one, or
demoted a freshly tiered remote replica. The fix in commit 116982595 routed
ChangedVolumes to NewVids/RemoteVids on the master, but ApplyVolumeChanges
returned only fresh arrivals and previously servable replicas. An existing
replica whose IsRemote() classification flipped was neither, so it never
reached the broadcast loop and the wdclient never learned.
Make Disk.doAddOrUpdateVolume return a third signal -- tierTransition --
true exactly when an existing replica's IsRemote() flips. ApplyVolumeChanges
treats that as an arrival so the existing SendHeartbeat routing loop now
sees it. Add a master-side end-to-end test covering local->remote,
remote->local, no-op re-reports, and a mixed heartbeat that only announces
the tier transition.
Also add docstrings to LookupFileId, wdclientLocationsToPb, and
LookupVolume where the prior change touched their bodies.
* topology: broadcast tier transitions received through full reconciliation
The previous commit added tier-transition routing on the ChangedVolumes
delta path, but that is not the only way a re-tiered replica reaches the
master. After a digest mismatch the volume server resends a full Volumes
list, and SyncDataNodeRegistration applies the new IsRemote() classification
silently -- the changedVolumes return value was being thrown away. The
master therefore never broadcast NewVids/RemoteVids, and a wdclient connected
during the recovery kept the stale DataInRemote until it lost contact with
the master.
Surface the changed set through UpdateVolumes.changedVolumes (now covering
both ReadOnly flips and tier flips) and SyncDataNodeRegistration, then route
it through NewVids/RemoteVids in SendHeartbeat the same way the delta path
already does. Add an end-to-end test for the full reconciliation path.
* master: keep an EC volume's locations in the volume lookup
The nodes that answer for an EC volume hold shards, not a volume record,
so asking them for one fails. Dropping the location on that failure
emptied the result and turned every EC read through the master's HTTP
lookup and fid redirect into a 404.
Treat an absent volume record as a local read and keep the node in the
answer. The per-node conversion moves into topologyLocation so the EC
case is covered by a test.
Claude-Session: https://claude.ai/code/session_01FcSp6quCaxQ1cb3Vx7o9fW
* wdclient: replace a tier-flipped location without writing under a reader
GetLocations hands back the entry's own slice and the caller walks it
after the read lock is dropped, which is why every other mutation here
builds a new slice. Writing the flipped replica into the array in place
raced LookupVolumeServerUrl, reported by -race.
Copy the slice, swap the one element, and publish it.
Claude-Session: https://claude.ai/code/session_01FcSp6quCaxQ1cb3Vx7o9fW
* master: keep a remote volume on NewVids for older clients
Moving remote-tier volumes out of NewVids and into RemoteVids alone is a
wire break in the wrong direction. A master upgraded ahead of its filers
and mounts -- the usual order -- announces a tiered volume only on a
field the older client ignores, so the volume drops out of that client's
vid map entirely and reads for it fail.
Announce every volume on NewVids and repeat the remote-tier subset on
RemoteVids, so a new client still learns the tier and an old one keeps
the location. The routing moves into announceVolume, which the heartbeat
paths and their tests now share instead of each restating it.
On the client, RemoteVids no longer needs a second write per volume: the
tier is settled before anything is added.
Claude-Session: https://claude.ai/code/session_01FcSp6quCaxQ1cb3Vx7o9fW
* topology: split the volume snapshot by tier without copying the records
ToVolumeLocations runs on every KeepConnected, so a filer or mount
connecting made the master allocate a full VolumeInfo per volume per node
just to read four bytes of id off each one. AppendVolumeIds exists to
avoid exactly that.
Extend it to fill the remote-tier list alongside the full one, and use it
again in the snapshot.
Claude-Session: https://claude.ai/code/session_01FcSp6quCaxQ1cb3Vx7o9fW
* wdclient: keep the data-center preference ahead of the local-first ordering
Hoisting every local replica to the very front puts an other-DC local
read ahead of a same-DC remote one. When the remote tier sits in the same
region as the replicas -- the common arrangement -- that trades an
in-region GET for a WAN round trip and costs more than the remote read it
avoids.
Reorder inside each data-center bucket instead, so local still wins among
equals and the data-center preference still wins overall.
Claude-Session: https://claude.ai/code/session_01FcSp6quCaxQ1cb3Vx7o9fW
* operation: pick the read replica from one list
The local-preferring lookup built a list of local URLs and then branched
on whether it was empty, duplicating the random pick. Fall back by
filling the same list with every replica instead.
Claude-Session: https://claude.ai/code/session_01FcSp6quCaxQ1cb3Vx7o9fW
---------
Co-authored-by: Bruce Zou <gift_secondst@msn.com>
Co-authored-by: bruce-zzz <bruce.zou@hhy-data.com>
330 lines
11 KiB
Go
330 lines
11 KiB
Go
package wdclient
|
|
|
|
import (
|
|
"context"
|
|
"errors"
|
|
"fmt"
|
|
"math/rand"
|
|
"sort"
|
|
"strconv"
|
|
"strings"
|
|
|
|
"golang.org/x/sync/singleflight"
|
|
|
|
"github.com/seaweedfs/seaweedfs/weed/glog"
|
|
"github.com/seaweedfs/seaweedfs/weed/pb"
|
|
"github.com/seaweedfs/seaweedfs/weed/util"
|
|
)
|
|
|
|
// VolumeLocationProvider is the interface for looking up volume locations
|
|
// This allows different implementations (master subscription, filer queries, etc.)
|
|
type VolumeLocationProvider interface {
|
|
// LookupVolumeIds looks up volume locations for the given volume IDs
|
|
// Returns a map of volume ID to locations
|
|
LookupVolumeIds(ctx context.Context, volumeIds []string) (map[string][]Location, error)
|
|
}
|
|
|
|
// vidMapClient provides volume location caching with pluggable lookup
|
|
// It wraps the battle-tested vidMap with customizable volume lookup strategies
|
|
type vidMapClient struct {
|
|
vidMap *vidMap
|
|
provider VolumeLocationProvider
|
|
vidLookupGroup singleflight.Group
|
|
}
|
|
|
|
const (
|
|
// DefaultVidMapCacheSize is the default number of resets a volume location
|
|
// survives without being relearned. This provides cache history when
|
|
// volumes move between servers.
|
|
DefaultVidMapCacheSize = 5
|
|
)
|
|
|
|
// newVidMapClient creates a new client with the given provider and data center
|
|
func newVidMapClient(provider VolumeLocationProvider, dataCenter string, cacheSize int) *vidMapClient {
|
|
return &vidMapClient{
|
|
vidMap: newVidMap(dataCenter, cacheSize),
|
|
provider: provider,
|
|
}
|
|
}
|
|
|
|
// GetLookupFileIdFunction returns a function that can be used to lookup file IDs
|
|
func (vc *vidMapClient) GetLookupFileIdFunction() LookupFileIdFunctionType {
|
|
return vc.LookupFileIdWithFallback
|
|
}
|
|
|
|
// LookupFileIdWithFallback resolves a "<vid>,<cookie>" file id to a list of
|
|
// HTTP read URLs, using the cached vidMap when populated and falling back to
|
|
// the provider for a fresh lookup on miss. URLs are returned in preference
|
|
// order: same-DC local, same-DC remote-tier, then the other data centers on
|
|
// the same footing -- mirroring the cached vidMap path so both routes agree.
|
|
// Concurrent misses for the same vid are coalesced via the singleflight group
|
|
// on LookupVolumeIdsWithFallback.
|
|
func (vc *vidMapClient) LookupFileIdWithFallback(ctx context.Context, fileId string) (fullUrls []string, err error) {
|
|
// Try cache first
|
|
dataCenter := vc.vidMap.DataCenter
|
|
fullUrls, err = vc.vidMap.LookupFileId(ctx, fileId)
|
|
|
|
// Cache hit - return immediately
|
|
if err == nil && len(fullUrls) > 0 {
|
|
return
|
|
}
|
|
|
|
// Cache miss - extract volume ID from file ID (format: "volumeId,needle_id_cookie")
|
|
if fileId == "" {
|
|
return nil, fmt.Errorf("empty fileId")
|
|
}
|
|
parts := strings.Split(fileId, ",")
|
|
if len(parts) != 2 {
|
|
return nil, fmt.Errorf("invalid fileId %s", fileId)
|
|
}
|
|
volumeId := parts[0]
|
|
|
|
// Use shared lookup logic with batching and singleflight
|
|
vidLocations, err := vc.LookupVolumeIdsWithFallback(ctx, []string{volumeId})
|
|
|
|
// Check for partial results first (important for multi-volume batched lookups)
|
|
locations, found := vidLocations[volumeId]
|
|
if !found || len(locations) == 0 {
|
|
// Volume not found - return specific error with context from lookup if available
|
|
if err != nil {
|
|
return nil, fmt.Errorf("volume %s not found for fileId %s: %w", volumeId, fileId, err)
|
|
}
|
|
return nil, fmt.Errorf("volume %s not found for fileId %s", volumeId, fileId)
|
|
}
|
|
|
|
// Volume found successfully - ignore any errors about other volumes
|
|
// (not relevant for single-volume lookup, but defensive for future batching)
|
|
|
|
// Build HTTP URLs from locations, preferring same data center
|
|
var sameDcUrls, otherDcUrls []string
|
|
localUrls := make(map[string]bool)
|
|
for _, loc := range locations {
|
|
httpUrl := "http://" + loc.Url + "/" + fileId
|
|
glog.V(4).Infof("lookup %s => %s, data in remote storage tier: %v", fileId, loc.Url, loc.DataInRemote)
|
|
if !loc.DataInRemote {
|
|
localUrls[httpUrl] = true
|
|
}
|
|
if dataCenter != "" && dataCenter == loc.DataCenter {
|
|
sameDcUrls = append(sameDcUrls, httpUrl)
|
|
} else {
|
|
otherDcUrls = append(otherDcUrls, httpUrl)
|
|
}
|
|
}
|
|
|
|
// Shuffle to distribute load across volume servers
|
|
rand.Shuffle(len(sameDcUrls), func(i, j int) { sameDcUrls[i], sameDcUrls[j] = sameDcUrls[j], sameDcUrls[i] })
|
|
rand.Shuffle(len(otherDcUrls), func(i, j int) { otherDcUrls[i], otherDcUrls[j] = otherDcUrls[j], otherDcUrls[i] })
|
|
|
|
// Local replicas go first inside each data center, but never ahead of the
|
|
// data-center preference itself. Mirrors vidMap.LookupVolumeServerUrl so
|
|
// all client lookup paths agree.
|
|
if len(localUrls) > 0 {
|
|
sameDcUrls = util.ReorderToFront(localUrls, sameDcUrls)
|
|
otherDcUrls = util.ReorderToFront(localUrls, otherDcUrls)
|
|
}
|
|
fullUrls = append(sameDcUrls, otherDcUrls...)
|
|
return fullUrls, nil
|
|
}
|
|
|
|
// LookupVolumeIdsWithFallback looks up volume locations, querying provider if not in cache.
|
|
// Uses singleflight to coalesce concurrent requests for the same batch of volumes.
|
|
//
|
|
// IMPORTANT: This function may return PARTIAL results with a non-nil error.
|
|
// The result map contains successfully looked up volumes, while the error aggregates
|
|
// failures for volumes that couldn't be found or had lookup errors.
|
|
//
|
|
// Callers MUST check both the result map AND the error:
|
|
// - result != nil && err == nil: All volumes found successfully
|
|
// - result != nil && err != nil: Some volumes found, some failed (check both)
|
|
// - result == nil && err != nil: Complete failure (connection error, etc.)
|
|
//
|
|
// Example usage:
|
|
//
|
|
// locs, err := mc.LookupVolumeIdsWithFallback(ctx, []string{"1", "2", "999"})
|
|
// if len(locs) > 0 {
|
|
// // Process successfully found volumes
|
|
// }
|
|
// if err != nil {
|
|
// // Log/handle failed volumes
|
|
// }
|
|
func (vc *vidMapClient) LookupVolumeIdsWithFallback(ctx context.Context, volumeIds []string) (map[string][]Location, error) {
|
|
result := make(map[string][]Location)
|
|
var needsLookup []string
|
|
var lookupErrors []error
|
|
|
|
// Check cache first and parse volume IDs once
|
|
vidStringToUint := make(map[string]uint32, len(volumeIds))
|
|
|
|
for _, vidString := range volumeIds {
|
|
vid, err := strconv.ParseUint(vidString, 10, 32)
|
|
if err != nil {
|
|
return nil, fmt.Errorf("invalid volume id %s: %v", vidString, err)
|
|
}
|
|
vidStringToUint[vidString] = uint32(vid)
|
|
|
|
locations, found := vc.vidMap.GetLocations(uint32(vid))
|
|
if found && len(locations) > 0 {
|
|
result[vidString] = locations
|
|
} else {
|
|
needsLookup = append(needsLookup, vidString)
|
|
}
|
|
}
|
|
|
|
if len(needsLookup) == 0 {
|
|
return result, nil
|
|
}
|
|
|
|
// Batch query all missing volumes using singleflight on the batch key
|
|
// Sort for stable key to coalesce identical batches
|
|
sort.Strings(needsLookup)
|
|
batchKey := strings.Join(needsLookup, ",")
|
|
|
|
sfResult, err, _ := vc.vidLookupGroup.Do(batchKey, func() (interface{}, error) {
|
|
// Double-check cache for volumes that might have been populated while waiting
|
|
stillNeedLookup := make([]string, 0, len(needsLookup))
|
|
batchResult := make(map[string][]Location)
|
|
|
|
for _, vidString := range needsLookup {
|
|
vid := vidStringToUint[vidString] // Use pre-parsed value
|
|
if locations, found := vc.vidMap.GetLocations(vid); found && len(locations) > 0 {
|
|
batchResult[vidString] = locations
|
|
} else {
|
|
stillNeedLookup = append(stillNeedLookup, vidString)
|
|
}
|
|
}
|
|
|
|
if len(stillNeedLookup) == 0 {
|
|
return batchResult, nil
|
|
}
|
|
|
|
// Query provider with batched volume IDs
|
|
glog.V(2).Infof("Looking up %d volumes from provider: %v", len(stillNeedLookup), stillNeedLookup)
|
|
|
|
providerResults, err := vc.provider.LookupVolumeIds(ctx, stillNeedLookup)
|
|
if err != nil {
|
|
return batchResult, fmt.Errorf("provider lookup failed: %w", err)
|
|
}
|
|
|
|
// Update cache with results
|
|
for vidString, locations := range providerResults {
|
|
vid, err := strconv.ParseUint(vidString, 10, 32)
|
|
if err != nil {
|
|
glog.Warningf("Failed to parse volume id '%s': %v", vidString, err)
|
|
continue
|
|
}
|
|
|
|
for _, loc := range locations {
|
|
vc.addLocation(uint32(vid), loc)
|
|
}
|
|
|
|
if len(locations) > 0 {
|
|
batchResult[vidString] = locations
|
|
}
|
|
}
|
|
|
|
return batchResult, nil
|
|
})
|
|
|
|
if err != nil {
|
|
lookupErrors = append(lookupErrors, err)
|
|
}
|
|
|
|
// Merge singleflight batch results
|
|
if batchLocations, ok := sfResult.(map[string][]Location); ok {
|
|
for vid, locs := range batchLocations {
|
|
result[vid] = locs
|
|
}
|
|
}
|
|
|
|
// Check for volumes that still weren't found
|
|
for _, vidString := range needsLookup {
|
|
if _, found := result[vidString]; !found {
|
|
lookupErrors = append(lookupErrors, fmt.Errorf("volume %s not found", vidString))
|
|
}
|
|
}
|
|
|
|
// Return aggregated errors
|
|
return result, errors.Join(lookupErrors...)
|
|
}
|
|
|
|
// Public methods for external access
|
|
//
|
|
// The vidMap itself is never replaced, so these all read the one map under its
|
|
// own lock. Resets bump its generation instead of swapping in a fresh instance.
|
|
|
|
// GetLocations safely retrieves volume locations
|
|
func (vc *vidMapClient) GetLocations(vid uint32) (locations []Location, found bool) {
|
|
return vc.vidMap.GetLocations(vid)
|
|
}
|
|
|
|
// GetLocationsClone safely retrieves a clone of volume locations
|
|
func (vc *vidMapClient) GetLocationsClone(vid uint32) (locations []Location, found bool) {
|
|
return vc.vidMap.GetLocationsClone(vid)
|
|
}
|
|
|
|
// GetVidLocations safely retrieves volume locations by string ID
|
|
func (vc *vidMapClient) GetVidLocations(vid string) (locations []Location, err error) {
|
|
return vc.vidMap.GetVidLocations(vid)
|
|
}
|
|
|
|
// LookupFileId safely looks up URLs for a file ID
|
|
func (vc *vidMapClient) LookupFileId(ctx context.Context, fileId string) (fullUrls []string, err error) {
|
|
return vc.vidMap.LookupFileId(ctx, fileId)
|
|
}
|
|
|
|
// LookupVolumeServerUrl safely looks up volume server URLs
|
|
func (vc *vidMapClient) LookupVolumeServerUrl(vid string) (serverUrls []string, err error) {
|
|
return vc.vidMap.LookupVolumeServerUrl(vid)
|
|
}
|
|
|
|
// HasVolumeServer reports whether addr is currently a known volume server
|
|
// (hosts at least one volume or EC shard) in the cached vid map. Used by
|
|
// admission paths that must only contact peers learned from the master.
|
|
func (vc *vidMapClient) HasVolumeServer(addr pb.ServerAddress) bool {
|
|
return vc.vidMap.hasVolumeServer(addr)
|
|
}
|
|
|
|
// GetDataCenter safely retrieves the data center
|
|
func (vc *vidMapClient) GetDataCenter() string {
|
|
return vc.vidMap.DataCenter
|
|
}
|
|
|
|
// Thread-safe helpers for vidMap operations
|
|
|
|
// addLocation adds a volume location
|
|
func (vc *vidMapClient) addLocation(vid uint32, location Location) {
|
|
vc.vidMap.addLocation(vid, location)
|
|
}
|
|
|
|
// deleteLocation removes a volume location
|
|
func (vc *vidMapClient) deleteLocation(vid uint32, location Location) {
|
|
vc.vidMap.deleteLocation(vid, location)
|
|
}
|
|
|
|
// addEcLocation adds an EC volume location
|
|
func (vc *vidMapClient) addEcLocation(vid uint32, location Location) {
|
|
vc.vidMap.addEcLocation(vid, location)
|
|
}
|
|
|
|
// deleteEcLocation removes an EC volume location
|
|
func (vc *vidMapClient) deleteEcLocation(vid uint32, location Location) {
|
|
vc.vidMap.deleteEcLocation(vid, location)
|
|
}
|
|
|
|
// resetVidMap starts a new generation, as when the master changes: what the
|
|
// previous one told us stays readable until it is relearned or expires.
|
|
func (vc *vidMapClient) resetVidMap() {
|
|
vc.vidMap.reset()
|
|
}
|
|
|
|
// InvalidateCache removes all cached locations for a volume ID
|
|
func (vc *vidMapClient) InvalidateCache(fileId string) {
|
|
parts := strings.Split(fileId, ",")
|
|
vidString := parts[0]
|
|
vid, err := strconv.ParseUint(vidString, 10, 32)
|
|
if err != nil {
|
|
return
|
|
}
|
|
vc.vidMap.deleteVid(uint32(vid))
|
|
}
|