README
¶
go-ds-flatfs
A datastore implementation using sharded directories and flat files to store data
go-ds-flatfs is used by go-ipfs to store raw block contents on disk. It supports several sharding functions (prefix, suffix, next-to-last/*).
It is not a general-purpose datastore and has several important restrictions. See the restrictions section for details.
Lead Maintainer
Table of Contents
Install
go-ds-flatfs can be used like any Go module:
import "github.com/ipfs/go-ds-flatfs"
Usage
Check the GoDoc module documentation for an overview of this module's functionality.
Restrictions
FlatFS keys are severely restricted. Only keys that match /[0-9A-Z+-_=]\+ are
allowed. That is, keys may only contain upper-case alpha-numeric characters,
'-', '+', '_', and '='. This is because values are written directly to the
filesystem without encoding.
Importantly, this means namespaced keys (e.g., /FOO/BAR), are not allowed. Attempts to write to such keys will result in an error.
DiskUsage and Accuracy
This datastore implements the PersistentDatastore interface. It offers a DiskUsage() method which strives to find a balance between accuracy and performance. This implies:
- The total disk usage of a datastore is calculated when opening the datastore
- The current disk usage is cached frequently in a file in the datastore root (
diskUsage.cacheby default). This file is also written when the datastore is closed. - If this file is not present when the datastore is opened:
- The disk usage will be calculated by walking the datastore's directory tree and estimating the size of each folder.
- This may be a very slow operation for huge datastores or datastores with slow disks
- The operation is time-limited (5 minutes by default).
- Upon timeout, the remaining folders will be assumed to have the average of the previously processed ones.
- After opening, the disk usage is updated in every write/delete operation.
This means that for certain datastores (huge ones, those with very slow disks or special content), the values reported by
DiskUsage() might be reduced accuracy and the first startup (without a diskUsage.cache file present), might be slow.
If you need increased accuracy or a fast start from the first time, you can manually create or update the
diskUsage.cache file.
The file diskUsage.cache is a JSON file with two fields diskUsage and accuracy. For example the JSON file for a
small repo might be:
{"diskUsage":6357,"accuracy":"initial-exact"}
diskUsage is the calculated disk usage and accuracy is a note on the accuracy of the initial calculation. If the
initial calculation was accurate the file will contain the value initial-exact. If some of the directories have too
many entries and the disk usage for that directory was estimated based on the first 2000 entries, the file will contain
initial-approximate. If the calculation took too long and timed out as indicated above, the file will contain
initial-timed-out.
If the initial calculation timed out the JSON file might be:
{"diskUsage":7589482442898,"accuracy":"initial-timed-out"}
To fix this with a more accurate value you could do (in the datastore root):
$ du -sb .
7536515831332 .
$ echo -n '{"diskUsage":7536515831332,"accuracy":"initial-exact"}' > diskUsage.cache
Contribute
PRs accepted.
Small note: If editing the README, please conform to the standard-readme specification.
License
MIT © Protocol Labs, Inc.
Documentation
¶
Overview ¶
Package flatfs is a Datastore implementation that stores all objects in a two-level directory structure in the local file system, regardless of the hierarchy of the keys.
Package flatfs is a Datastore implementation that stores all objects in a two-level directory structure in the local file system, regardless of the hierarchy of the keys.
Index ¶
- Constants
- Variables
- func Create(path string, fun *ShardIdV1) error
- func DirIsEmpty(name string) (bool, error)
- func DowngradeV1toV0(path string) error
- func Move(oldPath string, newPath string, out io.Writer) error
- func UpgradeV0toV1(path string, prefixLen int) error
- func WriteReadme(dir string, id *ShardIdV1) error
- func WriteShardFunc(dir string, id *ShardIdV1) error
- type Datastore
- func (fs *Datastore) Accuracy() string
- func (fs *Datastore) Batch(_ context.Context) (datastore.Batch, error)
- func (fs *Datastore) Close() error
- func (fs *Datastore) Delete(ctx context.Context, key datastore.Key) error
- func (fs *Datastore) DiskUsage(ctx context.Context) (uint64, error)
- func (fs *Datastore) Get(ctx context.Context, key datastore.Key) (value []byte, err error)
- func (fs *Datastore) GetSize(ctx context.Context, key datastore.Key) (size int, err error)
- func (fs *Datastore) Has(ctx context.Context, key datastore.Key) (exists bool, err error)
- func (fs *Datastore) Put(ctx context.Context, key datastore.Key, value []byte) error
- func (fs *Datastore) Query(ctx context.Context, q query.Query) (query.Results, error)
- func (fs *Datastore) ShardStr() string
- func (fs *Datastore) Sync(ctx context.Context, prefix datastore.Key) error
- type ShardFunc
- type ShardIdV1
Constants ¶
const PREFIX = "/repo/flatfs/shard/"
const README_FN = "_README"
const SHARDING_FN = "SHARDING"
const SyncThreadsMax = 16
don't block more than 16 threads on sync opearation 16 should be able to sataurate most RAIDs in case of two used disks per write (RAID 1, 5) and queue depth of 2, 16 concurrent Sync calls should be able to saturate 16 HDDs RAID TODO: benchmark it out, maybe provide tweak parmeter
Variables ¶
var ( // DiskUsageFile is the name of the file to cache the size of the // datastore in disk DiskUsageFile = "diskUsage.cache" // DiskUsageFilesAverage is the maximum number of files per folder // to stat in order to calculate the size of the datastore. // The size of the rest of the files in a folder will be assumed // to be the average of the values obtained. This includes // regular files and directories. DiskUsageFilesAverage = 2000 // DiskUsageCalcTimeout is the maximum time to spend // calculating the DiskUsage upon a start when no // DiskUsageFile is present. // If this period did not suffice to read the size of the datastore, // the remaining sizes will be stimated. DiskUsageCalcTimeout = 5 * time.Minute // RetryDelay is a timeout for a backoff on retrying operations // that fail due to transient errors like too many file descriptors open. RetryDelay = time.Millisecond * 200 // RetryAttempts is the maximum number of retries that will be attempted // before giving up. RetryAttempts = 6 )
var ( ErrDatastoreExists = errors.New("datastore already exists") ErrDatastoreDoesNotExist = errors.New("datastore directory does not exist") ErrShardingFileMissing = fmt.Errorf("%s file not found in datastore", SHARDING_FN) ErrClosed = errors.New("datastore closed") ErrInvalidKey = errors.New("key not supported by flatfs") )
var IPFS_DEF_SHARD = NextToLast(2)
var IPFS_DEF_SHARD_STR = IPFS_DEF_SHARD.String()
var README_IPFS_DEF_SHARD = `` /* 1123-byte string literal not displayed */
Functions ¶
func DirIsEmpty ¶
From: http://stackoverflow.com/questions/30697324/how-to-check-if-directory-on-path-is-empty
func DowngradeV1toV0 ¶
func UpgradeV0toV1 ¶
func WriteReadme ¶
func WriteShardFunc ¶
Types ¶
type Datastore ¶
type Datastore struct {
// contains filtered or unexported fields
}
Datastore implements the go-datastore Interface. Note this datastore cannot guarantee order of concurrent write operations to the same key. See the explanation in Put().
func CreateOrOpen ¶
convenience method
func (*Datastore) Accuracy ¶
Accuracy returns a string representing the accuracy of the DiskUsage() result, the value returned is implementation defined and for informational purposes only
func (*Datastore) Delete ¶
Delete removes a key/value from the Datastore. Please read the Put() explanation about the handling of concurrent write operations to the same key.
func (*Datastore) DiskUsage ¶
DiskUsage implements the PersistentDatastore interface and returns the current disk usage in bytes used by this datastore.
The size is approximative and may slightly differ from the real disk values.
func (*Datastore) Put ¶
Put stores a key/value in the datastore.
Note, that we do not guarantee order of write operations (Put or Delete) to the same key in this datastore.
For example. i.e. in the case of two concurrent Put, we only guarantee that one of them will come through, but cannot assure which one even if one arrived slightly later than the other. In the case of a concurrent Put and a Delete operation, we cannot guarantee which one will win.
type ShardIdV1 ¶
type ShardIdV1 struct {
// contains filtered or unexported fields
}
Source Files
Directories