minio/cmd/storage-datatypes.go

474 lines
14 KiB
Go
Raw Normal View History

// Copyright (c) 2015-2021 MinIO, Inc.
//
// This file is part of MinIO Object Storage stack
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package cmd
2016-04-08 02:01:15 +00:00
import (
"time"
"github.com/minio/minio/internal/crypto"
2016-04-08 02:01:15 +00:00
)
//go:generate msgp -file=$GOFILE
// DeleteOptions represents the disk level delete options available for the APIs
type DeleteOptions struct {
BaseOptions
perf: websocket grid connectivity for all internode communication (#18461) This PR adds a WebSocket grid feature that allows servers to communicate via a single two-way connection. There are two request types: * Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small roundtrips with small payloads. * Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`, which allows for different combinations of full two-way streams with an initial payload. Only a single stream is created between two machines - and there is, as such, no server/client relation since both sides can initiate and handle requests. Which server initiates the request is decided deterministically on the server names. Requests are made through a mux client and server, which handles message passing, congestion, cancelation, timeouts, etc. If a connection is lost, all requests are canceled, and the calling server will try to reconnect. Registered handlers can operate directly on byte slices or use a higher-level generics abstraction. There is no versioning of handlers/clients, and incompatible changes should be handled by adding new handlers. The request path can be changed to a new one for any protocol changes. First, all servers create a "Manager." The manager must know its address as well as all remote addresses. This will manage all connections. To get a connection to any remote, ask the manager to provide it given the remote address using. ``` func (m *Manager) Connection(host string) *Connection ``` All serverside handlers must also be registered on the manager. This will make sure that all incoming requests are served. The number of in-flight requests and responses must also be given for streaming requests. The "Connection" returned manages the mux-clients. Requests issued to the connection will be sent to the remote. * `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)` performs a single request and returns the result. Any deadline provided on the request is forwarded to the server, and canceling the context will make the function return at once. * `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)` will initiate a remote call and send the initial payload. ```Go // A Stream is a two-way stream. // All responses *must* be read by the caller. // If the call is canceled through the context, //The appropriate error will be returned. type Stream struct { // Responses from the remote server. // Channel will be closed after an error or when the remote closes. // All responses *must* be read by the caller until either an error is returned or the channel is closed. // Canceling the context will cause the context cancellation error to be returned. Responses <-chan Response // Requests sent to the server. // If the handler is defined with 0 incoming capacity this will be nil. // Channel *must* be closed to signal the end of the stream. // If the request context is canceled, the stream will no longer process requests. Requests chan<- []byte } type Response struct { Msg []byte Err error } ``` There are generic versions of the server/client handlers that allow the use of type safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-21 01:09:35 +00:00
Recursive bool `msg:"r"`
Immediate bool `msg:"i"`
UndoWrite bool `msg:"u"`
}
// BaseOptions represents common options for all Storage API calls
type BaseOptions struct{}
// RenameOptions represents rename API options, currently its same as BaseOptions
type RenameOptions struct {
BaseOptions
}
// DiskInfoOptions options for requesting custom results.
type DiskInfoOptions struct {
DiskID string `msg:"id"`
Metrics bool `msg:"m"`
NoOp bool `msg:"np"`
}
// DiskInfo is an extended type which returns current
// disk usage per path.
// The above means that any added/deleted fields are incompatible.
//
// The above means that any added/deleted fields are incompatible.
//
//msgp:tuple DiskInfo
type DiskInfo struct {
Total uint64
Free uint64
Used uint64
UsedInodes uint64
FreeInodes uint64
Major uint32
Minor uint32
NRRequests uint64
FSType string
RootDisk bool
Healing bool
Scanning bool
Endpoint string
MountPath string
ID string
Rotational bool
Metrics DiskMetrics
Error string // carries the error over the network
}
// DiskMetrics has the information about XL Storage APIs
// the number of calls of each API and the moving average of
// the duration of each API.
type DiskMetrics struct {
LastMinute map[string]AccElem `json:"apiLatencies,omitempty"`
APICalls map[string]uint64 `json:"apiCalls,omitempty"`
TotalWaiting uint32 `json:"totalWaiting,omitempty"`
TotalErrorsAvailability uint64 `json:"totalErrsAvailability"`
TotalErrorsTimeout uint64 `json:"totalErrsTimeout"`
TotalWrites uint64 `json:"totalWrites"`
TotalDeletes uint64 `json:"totalDeletes"`
}
// VolsInfo is a collection of volume(bucket) information
type VolsInfo []VolInfo
// VolInfo - represents volume stat information.
// The above means that any added/deleted fields are incompatible.
//
// The above means that any added/deleted fields are incompatible.
//
//msgp:tuple VolInfo
type VolInfo struct {
// Name of the volume.
Name string
// Date and time when the volume was created.
Created time.Time
}
// FilesInfo represent a list of files, additionally
// indicates if the list is last.
//
//msgp:tuple FileInfo
type FilesInfo struct {
Files []FileInfo
IsTruncated bool
}
feat: decommission feature for pools (#14012) ``` λ mc admin decommission start alias/ http://minio{1...2}/data{1...4} ``` ``` λ mc admin decommission status alias/ ┌─────┬─────────────────────────────────┬──────────────────────────────────┬────────┐ │ ID │ Pools │ Capacity │ Status │ │ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Active │ │ 2nd │ http://minio{3...4}/data{1...4} │ 329 GiB (used) / 421 GiB (total) │ Active │ └─────┴─────────────────────────────────┴──────────────────────────────────┴────────┘ ``` ``` λ mc admin decommission status alias/ http://minio{1...2}/data{1...4} Progress: ===================> [1GiB/sec] [15%] [4TiB/50TiB] Time Remaining: 4 hours (started 3 hours ago) ``` ``` λ mc admin decommission status alias/ http://minio{1...2}/data{1...4} ERROR: This pool is not scheduled for decommissioning currently. ``` ``` λ mc admin decommission cancel alias/ ┌─────┬─────────────────────────────────┬──────────────────────────────────┬──────────┐ │ ID │ Pools │ Capacity │ Status │ │ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Draining │ └─────┴─────────────────────────────────┴──────────────────────────────────┴──────────┘ ``` > NOTE: Canceled decommission will not make the pool active again, since we might have > Potentially partial duplicate content on the other pools, to avoid this scenario be > very sure to start decommissioning as a planned activity. ``` λ mc admin decommission cancel alias/ http://minio{1...2}/data{1...4} ┌─────┬─────────────────────────────────┬──────────────────────────────────┬────────────────────┐ │ ID │ Pools │ Capacity │ Status │ │ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Draining(Canceled) │ └─────┴─────────────────────────────────┴──────────────────────────────────┴────────────────────┘ ```
2022-01-10 17:07:49 +00:00
// Size returns size of all versions for the object 'Name'
func (f FileInfoVersions) Size() (size int64) {
for _, v := range f.Versions {
size += v.Size
}
return size
}
// FileInfoVersions represent a list of versions for a given file.
// The above means that any added/deleted fields are incompatible.
//
// The above means that any added/deleted fields are incompatible.
//
//msgp:tuple FileInfoVersions
type FileInfoVersions struct {
// Name of the volume.
Volume string `msg:"v,omitempty"`
// Name of the file.
Name string `msg:"n,omitempty"`
// Represents the latest mod time of the
// latest version.
LatestModTime time.Time `msg:"lm"`
Versions []FileInfo `msg:"vs"`
FreeVersions []FileInfo `msg:"fvs"`
}
// findVersionIndex will return the version index where the version
// was found. Returns -1 if not found.
func (f *FileInfoVersions) findVersionIndex(v string) int {
if f == nil || v == "" {
return -1
}
for i, ver := range f.Versions {
if ver.VersionID == v {
return i
}
}
return -1
}
// RawFileInfo - represents raw file stat information as byte array.
// The above means that any added/deleted fields are incompatible.
// Make sure to bump the internode version at storage-rest-common.go
type RawFileInfo struct {
// Content of entire xl.meta (may contain data depending on what was requested by the caller.
perf: websocket grid connectivity for all internode communication (#18461) This PR adds a WebSocket grid feature that allows servers to communicate via a single two-way connection. There are two request types: * Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small roundtrips with small payloads. * Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`, which allows for different combinations of full two-way streams with an initial payload. Only a single stream is created between two machines - and there is, as such, no server/client relation since both sides can initiate and handle requests. Which server initiates the request is decided deterministically on the server names. Requests are made through a mux client and server, which handles message passing, congestion, cancelation, timeouts, etc. If a connection is lost, all requests are canceled, and the calling server will try to reconnect. Registered handlers can operate directly on byte slices or use a higher-level generics abstraction. There is no versioning of handlers/clients, and incompatible changes should be handled by adding new handlers. The request path can be changed to a new one for any protocol changes. First, all servers create a "Manager." The manager must know its address as well as all remote addresses. This will manage all connections. To get a connection to any remote, ask the manager to provide it given the remote address using. ``` func (m *Manager) Connection(host string) *Connection ``` All serverside handlers must also be registered on the manager. This will make sure that all incoming requests are served. The number of in-flight requests and responses must also be given for streaming requests. The "Connection" returned manages the mux-clients. Requests issued to the connection will be sent to the remote. * `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)` performs a single request and returns the result. Any deadline provided on the request is forwarded to the server, and canceling the context will make the function return at once. * `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)` will initiate a remote call and send the initial payload. ```Go // A Stream is a two-way stream. // All responses *must* be read by the caller. // If the call is canceled through the context, //The appropriate error will be returned. type Stream struct { // Responses from the remote server. // Channel will be closed after an error or when the remote closes. // All responses *must* be read by the caller until either an error is returned or the channel is closed. // Canceling the context will cause the context cancellation error to be returned. Responses <-chan Response // Requests sent to the server. // If the handler is defined with 0 incoming capacity this will be nil. // Channel *must* be closed to signal the end of the stream. // If the request context is canceled, the stream will no longer process requests. Requests chan<- []byte } type Response struct { Msg []byte Err error } ``` There are generic versions of the server/client handlers that allow the use of type safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-21 01:09:35 +00:00
Buf []byte `msg:"b,allownil"`
}
// FileInfo - represents file stat information.
// The above means that any added/deleted fields are incompatible.
// Make sure to bump the internode version at storage-rest-common.go
2016-04-08 02:01:15 +00:00
type FileInfo struct {
// Name of the volume.
Volume string `msg:"v,omitempty"`
// Name of the file.
Name string `msg:"n,omitempty"`
// Version of the file.
VersionID string `msg:"vid,omitempty"`
// Indicates if the version is the latest
IsLatest bool `msg:"is"`
// Deleted is set when this FileInfo represents
// a deleted marker for a versioned bucket.
Deleted bool `msg:"del"`
// TransitionStatus is set to Pending/Complete for transitioned
// entries based on state of transition
TransitionStatus string `msg:"ts"`
// TransitionedObjName is the object name on the remote tier corresponding
// to object (version) on the source tier.
TransitionedObjName string `msg:"to"`
// TransitionTier is the storage class label assigned to remote tier.
TransitionTier string `msg:"tt"`
// TransitionVersionID stores a version ID of the object associate
// with the remote tier.
TransitionVersionID string `msg:"tv"`
// ExpireRestored indicates that the restored object is to be expired.
ExpireRestored bool `msg:"exp"`
// DataDir of the file
DataDir string `msg:"dd"`
// Indicates if this object is still in V1 format.
XLV1 bool `msg:"v1"`
// Date and time when the file was last modified, if Deleted
// is 'true' this value represents when while was deleted.
ModTime time.Time `msg:"mt"`
// Total file size.
Size int64 `msg:"sz"`
// File mode bits.
Mode uint32 `msg:"m"`
// WrittenByVersion is the unix time stamp of the MinIO
// version that created this version of the object.
WrittenByVersion uint64 `msg:"wv"`
// File metadata
Metadata map[string]string `msg:"meta"`
// All the parts per object.
Parts []ObjectPartInfo `msg:"parts"`
// Erasure info for all objects.
Erasure ErasureInfo `msg:"ei"`
MarkDeleted bool `msg:"md"` // mark this version as deleted
ReplicationState ReplicationState `msg:"rs"` // Internal replication state to be passed back in ObjectInfo
Data []byte `msg:"d,allownil"` // optionally carries object data
NumVersions int `msg:"nv"`
SuccessorModTime time.Time `msg:"smt"`
Fresh bool `msg:"fr"` // indicates this is a first time call to write FileInfo.
// Position of this version or object in a multi-object delete call,
// no other caller must set this value other than multi-object delete call.
// usage in other calls in undefined please avoid.
Idx int `msg:"i"`
2022-08-29 23:57:16 +00:00
// Combined checksum when object was uploaded.
Checksum []byte `msg:"cs,allownil"`
// Versioned - indicates if this file is versioned or not.
Versioned bool `msg:"vs"`
}
// ShallowCopy - copies minimal information for READ MRF checks.
func (fi FileInfo) ShallowCopy() (n FileInfo) {
n.Volume = fi.Volume
n.Name = fi.Name
n.VersionID = fi.VersionID
n.Deleted = fi.Deleted
n.Erasure = fi.Erasure
return
}
// WriteQuorum returns expected write quorum for this FileInfo
func (fi FileInfo) WriteQuorum(dquorum int) int {
if fi.Deleted {
return dquorum
}
quorum := fi.Erasure.DataBlocks
if fi.Erasure.DataBlocks == fi.Erasure.ParityBlocks {
quorum++
}
return quorum
}
// ReadQuorum returns expected read quorum for this FileInfo
func (fi FileInfo) ReadQuorum(dquorum int) int {
if fi.Deleted {
return dquorum
}
return fi.Erasure.DataBlocks
}
// Equals checks if fi(FileInfo) matches ofi(FileInfo)
func (fi FileInfo) Equals(ofi FileInfo) (ok bool) {
typ1, ok1 := crypto.IsEncrypted(fi.Metadata)
typ2, ok2 := crypto.IsEncrypted(ofi.Metadata)
if ok1 != ok2 {
return false
}
if typ1 != typ2 {
return false
}
if fi.IsCompressed() != ofi.IsCompressed() {
return false
}
if !fi.TransitionInfoEquals(ofi) {
return false
}
if !fi.ModTime.Equal(ofi.ModTime) {
return false
}
return fi.Erasure.Equal(ofi.Erasure)
}
// GetDataDir returns an expected dataDir given FileInfo
// - deleteMarker returns "delete-marker"
// - returns "legacy" if FileInfo is XLV1 and DataDir is
// empty, returns DataDir otherwise
// - returns "dataDir"
func (fi FileInfo) GetDataDir() string {
if fi.Deleted {
return "delete-marker"
}
if fi.XLV1 && fi.DataDir == "" {
return "legacy"
}
return fi.DataDir
}
// IsCompressed returns true if the object is marked as compressed.
func (fi FileInfo) IsCompressed() bool {
_, ok := fi.Metadata[ReservedMetadataPrefix+"compression"]
return ok
}
// InlineData returns true if object contents are inlined alongside its metadata.
func (fi FileInfo) InlineData() bool {
_, ok := fi.Metadata[ReservedMetadataPrefixLower+"inline-data"]
// Earlier MinIO versions didn't reset "x-minio-internal-inline-data"
// from fi.Metadata when the object was tiered. So, tiered objects
// would return true for InlineData() in these versions even though the
// object isn't inlined in xl.meta
return ok && !fi.IsRemote()
}
// SetInlineData marks object (version) as inline.
func (fi *FileInfo) SetInlineData() {
if fi.Metadata == nil {
fi.Metadata = make(map[string]string, 1)
}
fi.Metadata[ReservedMetadataPrefixLower+"inline-data"] = "true"
}
// VersionPurgeStatusKey denotes purge status in metadata
feat: decommission feature for pools (#14012) ``` λ mc admin decommission start alias/ http://minio{1...2}/data{1...4} ``` ``` λ mc admin decommission status alias/ ┌─────┬─────────────────────────────────┬──────────────────────────────────┬────────┐ │ ID │ Pools │ Capacity │ Status │ │ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Active │ │ 2nd │ http://minio{3...4}/data{1...4} │ 329 GiB (used) / 421 GiB (total) │ Active │ └─────┴─────────────────────────────────┴──────────────────────────────────┴────────┘ ``` ``` λ mc admin decommission status alias/ http://minio{1...2}/data{1...4} Progress: ===================> [1GiB/sec] [15%] [4TiB/50TiB] Time Remaining: 4 hours (started 3 hours ago) ``` ``` λ mc admin decommission status alias/ http://minio{1...2}/data{1...4} ERROR: This pool is not scheduled for decommissioning currently. ``` ``` λ mc admin decommission cancel alias/ ┌─────┬─────────────────────────────────┬──────────────────────────────────┬──────────┐ │ ID │ Pools │ Capacity │ Status │ │ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Draining │ └─────┴─────────────────────────────────┴──────────────────────────────────┴──────────┘ ``` > NOTE: Canceled decommission will not make the pool active again, since we might have > Potentially partial duplicate content on the other pools, to avoid this scenario be > very sure to start decommissioning as a planned activity. ``` λ mc admin decommission cancel alias/ http://minio{1...2}/data{1...4} ┌─────┬─────────────────────────────────┬──────────────────────────────────┬────────────────────┐ │ ID │ Pools │ Capacity │ Status │ │ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Draining(Canceled) │ └─────┴─────────────────────────────────┴──────────────────────────────────┴────────────────────┘ ```
2022-01-10 17:07:49 +00:00
const (
VersionPurgeStatusKey = ReservedMetadataPrefixLower + "purgestatus"
)
// newFileInfo - initializes new FileInfo, allocates a fresh erasure info.
func newFileInfo(object string, dataBlocks, parityBlocks int) (fi FileInfo) {
fi.Erasure = ErasureInfo{
Algorithm: erasureAlgorithm,
DataBlocks: dataBlocks,
ParityBlocks: parityBlocks,
BlockSize: blockSizeV2,
Distribution: hashOrder(object, dataBlocks+parityBlocks),
}
return fi
}
// ReadMultipleReq contains information of multiple files to read from disk.
type ReadMultipleReq struct {
Bucket string // Bucket. Can be empty if multiple buckets.
Prefix string // Shared prefix of all files. Can be empty. Will be joined to filename without modification.
Files []string // Individual files to read.
MaxSize int64 // Return error if size is exceed.
MetadataOnly bool // Read as XL meta and truncate data.
AbortOn404 bool // Stop reading after first file not found.
MaxResults int // Stop after this many successful results. <= 0 means all.
}
// ReadMultipleResp contains a single response from a ReadMultipleReq.
type ReadMultipleResp struct {
Bucket string // Bucket as given by request.
Prefix string // Prefix as given by request.
File string // File name as given in request.
Exists bool // Returns whether the file existed on disk.
Error string // Returns any error when reading.
Data []byte // Contains all data of file.
Modtime time.Time // Modtime of file on disk.
}
perf: websocket grid connectivity for all internode communication (#18461) This PR adds a WebSocket grid feature that allows servers to communicate via a single two-way connection. There are two request types: * Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small roundtrips with small payloads. * Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`, which allows for different combinations of full two-way streams with an initial payload. Only a single stream is created between two machines - and there is, as such, no server/client relation since both sides can initiate and handle requests. Which server initiates the request is decided deterministically on the server names. Requests are made through a mux client and server, which handles message passing, congestion, cancelation, timeouts, etc. If a connection is lost, all requests are canceled, and the calling server will try to reconnect. Registered handlers can operate directly on byte slices or use a higher-level generics abstraction. There is no versioning of handlers/clients, and incompatible changes should be handled by adding new handlers. The request path can be changed to a new one for any protocol changes. First, all servers create a "Manager." The manager must know its address as well as all remote addresses. This will manage all connections. To get a connection to any remote, ask the manager to provide it given the remote address using. ``` func (m *Manager) Connection(host string) *Connection ``` All serverside handlers must also be registered on the manager. This will make sure that all incoming requests are served. The number of in-flight requests and responses must also be given for streaming requests. The "Connection" returned manages the mux-clients. Requests issued to the connection will be sent to the remote. * `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)` performs a single request and returns the result. Any deadline provided on the request is forwarded to the server, and canceling the context will make the function return at once. * `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)` will initiate a remote call and send the initial payload. ```Go // A Stream is a two-way stream. // All responses *must* be read by the caller. // If the call is canceled through the context, //The appropriate error will be returned. type Stream struct { // Responses from the remote server. // Channel will be closed after an error or when the remote closes. // All responses *must* be read by the caller until either an error is returned or the channel is closed. // Canceling the context will cause the context cancellation error to be returned. Responses <-chan Response // Requests sent to the server. // If the handler is defined with 0 incoming capacity this will be nil. // Channel *must* be closed to signal the end of the stream. // If the request context is canceled, the stream will no longer process requests. Requests chan<- []byte } type Response struct { Msg []byte Err error } ``` There are generic versions of the server/client handlers that allow the use of type safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-21 01:09:35 +00:00
// DeleteVersionHandlerParams are parameters for DeleteVersionHandler
type DeleteVersionHandlerParams struct {
DiskID string `msg:"id"`
Volume string `msg:"v"`
FilePath string `msg:"fp"`
ForceDelMarker bool `msg:"fdm"`
Opts DeleteOptions `msg:"do"`
FI FileInfo `msg:"fi"`
perf: websocket grid connectivity for all internode communication (#18461) This PR adds a WebSocket grid feature that allows servers to communicate via a single two-way connection. There are two request types: * Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small roundtrips with small payloads. * Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`, which allows for different combinations of full two-way streams with an initial payload. Only a single stream is created between two machines - and there is, as such, no server/client relation since both sides can initiate and handle requests. Which server initiates the request is decided deterministically on the server names. Requests are made through a mux client and server, which handles message passing, congestion, cancelation, timeouts, etc. If a connection is lost, all requests are canceled, and the calling server will try to reconnect. Registered handlers can operate directly on byte slices or use a higher-level generics abstraction. There is no versioning of handlers/clients, and incompatible changes should be handled by adding new handlers. The request path can be changed to a new one for any protocol changes. First, all servers create a "Manager." The manager must know its address as well as all remote addresses. This will manage all connections. To get a connection to any remote, ask the manager to provide it given the remote address using. ``` func (m *Manager) Connection(host string) *Connection ``` All serverside handlers must also be registered on the manager. This will make sure that all incoming requests are served. The number of in-flight requests and responses must also be given for streaming requests. The "Connection" returned manages the mux-clients. Requests issued to the connection will be sent to the remote. * `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)` performs a single request and returns the result. Any deadline provided on the request is forwarded to the server, and canceling the context will make the function return at once. * `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)` will initiate a remote call and send the initial payload. ```Go // A Stream is a two-way stream. // All responses *must* be read by the caller. // If the call is canceled through the context, //The appropriate error will be returned. type Stream struct { // Responses from the remote server. // Channel will be closed after an error or when the remote closes. // All responses *must* be read by the caller until either an error is returned or the channel is closed. // Canceling the context will cause the context cancellation error to be returned. Responses <-chan Response // Requests sent to the server. // If the handler is defined with 0 incoming capacity this will be nil. // Channel *must* be closed to signal the end of the stream. // If the request context is canceled, the stream will no longer process requests. Requests chan<- []byte } type Response struct { Msg []byte Err error } ``` There are generic versions of the server/client handlers that allow the use of type safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-21 01:09:35 +00:00
}
// MetadataHandlerParams is request info for UpdateMetadataHandle and WriteMetadataHandler.
type MetadataHandlerParams struct {
DiskID string `msg:"id"`
Volume string `msg:"v"`
OrigVolume string `msg:"ov"`
perf: websocket grid connectivity for all internode communication (#18461) This PR adds a WebSocket grid feature that allows servers to communicate via a single two-way connection. There are two request types: * Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small roundtrips with small payloads. * Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`, which allows for different combinations of full two-way streams with an initial payload. Only a single stream is created between two machines - and there is, as such, no server/client relation since both sides can initiate and handle requests. Which server initiates the request is decided deterministically on the server names. Requests are made through a mux client and server, which handles message passing, congestion, cancelation, timeouts, etc. If a connection is lost, all requests are canceled, and the calling server will try to reconnect. Registered handlers can operate directly on byte slices or use a higher-level generics abstraction. There is no versioning of handlers/clients, and incompatible changes should be handled by adding new handlers. The request path can be changed to a new one for any protocol changes. First, all servers create a "Manager." The manager must know its address as well as all remote addresses. This will manage all connections. To get a connection to any remote, ask the manager to provide it given the remote address using. ``` func (m *Manager) Connection(host string) *Connection ``` All serverside handlers must also be registered on the manager. This will make sure that all incoming requests are served. The number of in-flight requests and responses must also be given for streaming requests. The "Connection" returned manages the mux-clients. Requests issued to the connection will be sent to the remote. * `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)` performs a single request and returns the result. Any deadline provided on the request is forwarded to the server, and canceling the context will make the function return at once. * `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)` will initiate a remote call and send the initial payload. ```Go // A Stream is a two-way stream. // All responses *must* be read by the caller. // If the call is canceled through the context, //The appropriate error will be returned. type Stream struct { // Responses from the remote server. // Channel will be closed after an error or when the remote closes. // All responses *must* be read by the caller until either an error is returned or the channel is closed. // Canceling the context will cause the context cancellation error to be returned. Responses <-chan Response // Requests sent to the server. // If the handler is defined with 0 incoming capacity this will be nil. // Channel *must* be closed to signal the end of the stream. // If the request context is canceled, the stream will no longer process requests. Requests chan<- []byte } type Response struct { Msg []byte Err error } ``` There are generic versions of the server/client handlers that allow the use of type safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-21 01:09:35 +00:00
FilePath string `msg:"fp"`
UpdateOpts UpdateMetadataOpts `msg:"uo"`
FI FileInfo `msg:"fi"`
}
// UpdateMetadataOpts provides an optional input to indicate if xl.meta updates need to be fully synced to disk.
type UpdateMetadataOpts struct {
NoPersistence bool `msg:"np"`
}
// CheckPartsHandlerParams are parameters for CheckPartsHandler
type CheckPartsHandlerParams struct {
DiskID string `msg:"id"`
Volume string `msg:"v"`
FilePath string `msg:"fp"`
FI FileInfo `msg:"fi"`
}
// DeleteFileHandlerParams are parameters for DeleteFileHandler
type DeleteFileHandlerParams struct {
DiskID string `msg:"id"`
Volume string `msg:"v"`
FilePath string `msg:"fp"`
Opts DeleteOptions `msg:"do"`
}
// RenameDataHandlerParams are parameters for RenameDataHandler.
type RenameDataHandlerParams struct {
DiskID string `msg:"id"`
SrcVolume string `msg:"sv"`
SrcPath string `msg:"sp"`
DstVolume string `msg:"dv"`
DstPath string `msg:"dp"`
FI FileInfo `msg:"fi"`
Opts RenameOptions `msg:"ro"`
perf: websocket grid connectivity for all internode communication (#18461) This PR adds a WebSocket grid feature that allows servers to communicate via a single two-way connection. There are two request types: * Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small roundtrips with small payloads. * Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`, which allows for different combinations of full two-way streams with an initial payload. Only a single stream is created between two machines - and there is, as such, no server/client relation since both sides can initiate and handle requests. Which server initiates the request is decided deterministically on the server names. Requests are made through a mux client and server, which handles message passing, congestion, cancelation, timeouts, etc. If a connection is lost, all requests are canceled, and the calling server will try to reconnect. Registered handlers can operate directly on byte slices or use a higher-level generics abstraction. There is no versioning of handlers/clients, and incompatible changes should be handled by adding new handlers. The request path can be changed to a new one for any protocol changes. First, all servers create a "Manager." The manager must know its address as well as all remote addresses. This will manage all connections. To get a connection to any remote, ask the manager to provide it given the remote address using. ``` func (m *Manager) Connection(host string) *Connection ``` All serverside handlers must also be registered on the manager. This will make sure that all incoming requests are served. The number of in-flight requests and responses must also be given for streaming requests. The "Connection" returned manages the mux-clients. Requests issued to the connection will be sent to the remote. * `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)` performs a single request and returns the result. Any deadline provided on the request is forwarded to the server, and canceling the context will make the function return at once. * `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)` will initiate a remote call and send the initial payload. ```Go // A Stream is a two-way stream. // All responses *must* be read by the caller. // If the call is canceled through the context, //The appropriate error will be returned. type Stream struct { // Responses from the remote server. // Channel will be closed after an error or when the remote closes. // All responses *must* be read by the caller until either an error is returned or the channel is closed. // Canceling the context will cause the context cancellation error to be returned. Responses <-chan Response // Requests sent to the server. // If the handler is defined with 0 incoming capacity this will be nil. // Channel *must* be closed to signal the end of the stream. // If the request context is canceled, the stream will no longer process requests. Requests chan<- []byte } type Response struct { Msg []byte Err error } ``` There are generic versions of the server/client handlers that allow the use of type safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-21 01:09:35 +00:00
}
// RenameFileHandlerParams are parameters for RenameFileHandler.
type RenameFileHandlerParams struct {
DiskID string `msg:"id"`
SrcVolume string `msg:"sv"`
SrcFilePath string `msg:"sp"`
DstVolume string `msg:"dv"`
DstFilePath string `msg:"dp"`
}
// ReadAllHandlerParams are parameters for ReadAllHandler.
type ReadAllHandlerParams struct {
DiskID string `msg:"id"`
Volume string `msg:"v"`
FilePath string `msg:"fp"`
}
// WriteAllHandlerParams are parameters for WriteAllHandler.
type WriteAllHandlerParams struct {
DiskID string `msg:"id"`
Volume string `msg:"v"`
FilePath string `msg:"fp"`
Buf []byte `msg:"b"`
}
perf: websocket grid connectivity for all internode communication (#18461) This PR adds a WebSocket grid feature that allows servers to communicate via a single two-way connection. There are two request types: * Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small roundtrips with small payloads. * Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`, which allows for different combinations of full two-way streams with an initial payload. Only a single stream is created between two machines - and there is, as such, no server/client relation since both sides can initiate and handle requests. Which server initiates the request is decided deterministically on the server names. Requests are made through a mux client and server, which handles message passing, congestion, cancelation, timeouts, etc. If a connection is lost, all requests are canceled, and the calling server will try to reconnect. Registered handlers can operate directly on byte slices or use a higher-level generics abstraction. There is no versioning of handlers/clients, and incompatible changes should be handled by adding new handlers. The request path can be changed to a new one for any protocol changes. First, all servers create a "Manager." The manager must know its address as well as all remote addresses. This will manage all connections. To get a connection to any remote, ask the manager to provide it given the remote address using. ``` func (m *Manager) Connection(host string) *Connection ``` All serverside handlers must also be registered on the manager. This will make sure that all incoming requests are served. The number of in-flight requests and responses must also be given for streaming requests. The "Connection" returned manages the mux-clients. Requests issued to the connection will be sent to the remote. * `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)` performs a single request and returns the result. Any deadline provided on the request is forwarded to the server, and canceling the context will make the function return at once. * `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)` will initiate a remote call and send the initial payload. ```Go // A Stream is a two-way stream. // All responses *must* be read by the caller. // If the call is canceled through the context, //The appropriate error will be returned. type Stream struct { // Responses from the remote server. // Channel will be closed after an error or when the remote closes. // All responses *must* be read by the caller until either an error is returned or the channel is closed. // Canceling the context will cause the context cancellation error to be returned. Responses <-chan Response // Requests sent to the server. // If the handler is defined with 0 incoming capacity this will be nil. // Channel *must* be closed to signal the end of the stream. // If the request context is canceled, the stream will no longer process requests. Requests chan<- []byte } type Response struct { Msg []byte Err error } ``` There are generic versions of the server/client handlers that allow the use of type safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-21 01:09:35 +00:00
// RenameDataResp - RenameData()'s response.
type RenameDataResp struct {
Signature uint64 `msg:"sig"`
}
// LocalDiskIDs - GetLocalIDs response.
type LocalDiskIDs struct {
IDs []string
}
// ListDirResult - ListDir()'s response.
type ListDirResult struct {
Entries []string `msg:"e"`
}