featurebase/cmd/roaring-migrate/main.go
Seebs ad30a926f4 Giant Commit: drop a bunch of stuff we don't use.
These commits are hard to disentagle, and doing them separately means
re-modifying the same chunks of code several times before removing it,
and similar things.

Basically:
(1) Drop the bolt backend storage.
(2) Drop the blue-green wrapper that compares two backends.
(3) Drop unused or barely-used Tx API components from all the
remaining backends.
(4) Minor related cleanup to simplify things related to these.

The boltdb backend existed only to verify RBF. The blue-green wrapper
was mostly used to verify RBF, but in practice we had to do a lot
of working around that, and it introduced a lot of special cases.

Types removed:

IteratorFinder: Used only to implement the roaring iterator
on top of boltdb, and to complicate the way it worked in roaring.
Reverted the complications. Also unexport NewSliceContainers
which is used only for that outside of roaring's internals.

PortMapper from cluster_internal_test.go: Used only for a test
we removed early this year. Never used for anything else.

RawRoaringData: Totally unused.

TxStore: Totally unused.

Functions removed from Tx API, and sometimes corresponding
members were removed from structs:

* Dump: debugging code, I don't think I found any actually reachable
  paths to it.
* Group: only used for debugging TxGroup stuff
* IncrementOpN: only used by fragment, fragment can increment its
  own opN.
* Options: unused?
* Pointer: debugging only
* Readonly: used only to decide how to handle Tx in a TxGrp,
  but we never add a non-readonly Tx to a TxGrp. Removed also all
  the corresponding write-aware stuff.
* RoaringBitmapReader: Used exactly once, can just be a bm.WriteTo.
* Sn (and OpenSnList): Unused
* UnionInPlace: unused and conceptually-invalid; it didn't write
  to storage and shouldn't have, and was just "create a bitmap
  then call union-in-place", which we can do directly.
* UseRowCache: just checked storage.UseRowCache.

Other things removed:

The SetRequiredForAtomicWriteTx and ClearRequiredForAtomicWriteTx
functions go away, since nothing now seems to be using them? Same
for holder_internal_test's `testHasBit` and `testMustNotHaveBit`,
which were unused.

The DBPerShard "DeleteDBPath" and "HasData" functions and related
parts were mostly unused; took out the parts that were never
actually being reached.

Changed the API of one function to simplify special cases and
remove things:
* ImportRoaringBits had a special "data" argument which gave it
  subtly different semantics for RBF and roaring (for roaring, it
  could produce a roaring bitmap *with ops log*), didn't seem to
  be adding much. Removed corresponding "readStorageFromArchive"
  which is not otherwise used.

Also took out various debugging/dumping functions that were unused
and may have bitrotted.

Dropped a test from txfactory_internal_test, and the "pjobs"
code, because those two were the only things that needed Barrier
and thus idem, which lets us drop two more dependencies. We already
have errgroup for grouping things which want to terminate as
soon as one of them errors, approximately. To do better we'd have
to have context-threading, really.

Unbroke the WriteFragment test for non-roaring tests and made it
not roaring-only.
2021-10-26 12:30:25 -05:00

444 lines
11 KiB
Go

// Copyright 2017 Pilosa Corp.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
package main
import (
"encoding/json"
"fmt"
"io/ioutil"
"os"
"path/filepath"
"sort"
"strconv"
"strings"
"syscall"
pilosa "github.com/molecula/featurebase/v2"
"github.com/molecula/featurebase/v2/rbf"
"github.com/molecula/featurebase/v2/rbf/cfg"
"github.com/molecula/featurebase/v2/roaring"
txkey "github.com/molecula/featurebase/v2/short_txkey"
"github.com/spf13/cobra"
)
func main() {
var dataDir, backupPath string
cmdMigrate := &cobra.Command{
Use: "roaring-migrate",
Short: "convert roaring pilosa backup to rbf",
Long: `roaring-migrate uses the pilosa data-dir for each node, and produces a new backup that is able to be restored from utilizing the new pilosa restore tool.`,
Run: func(cmd *cobra.Command, args []string) {
nodes := strings.Split(dataDir, ",")
for _, nodePath := range nodes {
err := Migrate(nodePath, backupPath)
if err != nil {
fmt.Println("Error", err)
return
}
}
},
}
cmdMigrate.Flags().StringVarP(&dataDir, "data-dir", "d", "", "source directories for each node seperated by commas")
cmdMigrate.Flags().StringVarP(&backupPath, "backup-dir", "b", "", "location of backup directory")
err := cmdMigrate.MarkFlagRequired("data-dir")
if err != nil {
fmt.Println("Error setting flag data-dir")
os.Exit(1)
return
}
err = cmdMigrate.MarkFlagRequired("backup-dir")
if err != nil {
fmt.Println("Error setting flag backup-dir")
os.Exit(1)
return
}
err = cmdMigrate.Execute()
if err != nil {
fmt.Println("exec error", err)
os.Exit(1)
}
}
func FetchFragments(base string) []string {
var fragments []string
ff := func(pathX string, infoX os.FileInfo, errX error) error {
// first thing to do, check error. and decide what to do about it
if errX != nil {
fmt.Printf("error 「%v」 at a path 「%q」\n", errX, pathX)
return errX
}
pathX = pathX[len(base):]
if !infoX.IsDir() {
if strings.Contains(pathX, "fragment") && !strings.Contains(pathX, "cache") {
fragments = append(fragments, pathX)
}
}
return nil
}
err := filepath.Walk(base, ff)
if err != nil {
fmt.Printf("error walking the path %q: %v\n", base, err)
}
return fragments
}
type local struct {
Name string `json:"name,omitempty"`
CreatedAt uint64 `json:"createdAt,omitempty"`
Options pilosa.IndexOptions `json:"options,omitempty"`
Fields []*pilosa.FieldInfo `json:"fields,omitempty"`
}
func BuildSchema(dataDir string) ([]byte, error) {
//need to find all the ".meta" files and load as field options
schemaSerializer := struct {
Indexes []*local `json:"indexes,omitempty"`
}{Indexes: make([]*local, 0)}
var l *local
ff := func(pathX string, infoX os.FileInfo, errX error) error {
// first thing to do, check error. and decide what to do about it
if errX != nil {
fmt.Printf("error 「%v」 at a path 「%q」\n", errX, pathX)
return errX
}
pathX = pathX[len(dataDir):]
if infoX.IsDir() {
//filepath.Walk(pathX, ff)
} else {
if strings.Contains(pathX, ".meta") {
//convert the file to a fieldOptions
// ex: metaPath /trait_store/aba/.meta
fmt.Println("PATHX", pathX)
t := strings.Split(pathX, "/")
index := t[1]
src := dataDir + pathX
content, err := ioutil.ReadFile(src)
if err != nil {
return err
}
fstat, err := os.Stat(src)
stat := fstat.Sys().(*syscall.Stat_t)
if err != nil {
return err
}
if len(t) == 3 {
io, err := pilosa.UnmarshalIndexOptions(index, 0, content)
if err != nil {
return err
}
l = &local{
Name: index,
Fields: make([]*pilosa.FieldInfo, 0),
CreatedAt: uint64(CTimeNano(stat)),
Options: *io,
}
//index options
schemaSerializer.Indexes = append(schemaSerializer.Indexes, l)
return nil
}
field := t[2]
if field != "_exists" {
fi, err := pilosa.UnmarshalFieldOptions(field, CTimeNano(stat), content)
if err != nil {
return err
}
l.Fields = append(l.Fields, fi)
}
}
}
return nil
}
err := filepath.Walk(dataDir, ff)
if err != nil {
fmt.Printf("error walking the path %q: %v\n", dataDir, err)
}
return json.MarshalIndent(schemaSerializer, "", " ")
}
func Extract(filename string) (index, field, view string, shard uint64) {
//trait_store/aba/views/standard/fragments
parts := strings.Split(filename, "/")
shard, _ = strconv.ParseUint(parts[6], 10, 64)
return parts[1], parts[2], parts[4], shard
}
//just a way to collect all the open dbs
type rbfFile struct {
working *rbf.DB
last string
temp string
}
func (d *rbfFile) getDB(path, index string, shard uint64) (*rbf.DB, error) {
src := fmt.Sprintf("%v/indexes/%v/shards/%04d", path, index, shard)
if d.last != src {
d.Close()
d.last = src
fmt.Println("RBF:", src)
c := cfg.NewDefaultConfig()
c.FsyncEnabled = false
c.MinWALCheckpointSize = 0
db := rbf.NewDB(d.temp, c)
err := db.Open()
if err != nil {
return nil, err
}
d.working = db
}
return d.working, nil
}
func (d *rbfFile) Close() error {
if d.last != "" {
d.working.Close()
//if d.last exists only keep the biggest
err := os.MkdirAll(filepath.Dir(d.last), 0777)
if err != nil {
return err
}
// move the datafile backup shard
err = os.Rename(filepath.Join(d.temp, "data"), d.last)
if err != nil {
return err
}
//cleanup the tempdirectory
err = os.RemoveAll(d.temp)
if err != nil {
return err
}
}
return nil
}
func copyFile(src, dest string) error {
input, err := ioutil.ReadFile(src)
if err != nil {
return err
}
err = ioutil.WriteFile(dest, input, 0644)
if err != nil {
return err
}
return nil
}
func Migrate(dataDir, backupPath string) error {
err := os.MkdirAll(backupPath, 0777)
if err != nil {
return err
}
err = copyFile(filepath.Join(dataDir, "idalloc.db"), filepath.Join(backupPath, "idalloc"))
if err != nil {
return err
}
schema, err := BuildSchema(dataDir)
if err != nil {
return err
}
err = ioutil.WriteFile(filepath.Join(backupPath, "schema"), schema, 0644)
if err != nil {
return err
}
raw := FetchFragments(dataDir)
sort.Slice(raw, func(i, j int) bool {
//trait_store/zip_code/views/standard/fragments/
ti := strings.LastIndex(raw[i], "/") + 1
shardi, err := strconv.ParseUint(raw[i][ti:], 10, 16)
if err != nil {
panic(err)
}
tj := strings.LastIndex(raw[j], "/") + 1
shardj, err := strconv.ParseUint(raw[j][tj:], 10, 16)
if err != nil {
panic(err)
}
if shardi < shardj {
return true
} else if shardi == shardj {
return strings.Compare(raw[i][:ti], raw[j][:tj]) < 0
}
return false
})
//raw is now sorted by shard
// need index/field/shard
// make rbf file in backup
rowSize := uint64(0) //?
clear := false
log := false
cache := &rbfFile{
temp: filepath.Join(backupPath, "_SCRATCH"),
}
for _, filename := range raw {
index, field, view, shard := Extract(filename)
content, err := ioutil.ReadFile(dataDir + filename)
if err != nil {
return err
}
itr, err := roaring.NewRoaringIterator(content)
if err != nil {
return err
}
db, err := cache.getDB(backupPath, index, shard)
if err != nil {
return err
}
tx, err := db.Begin(true)
if err != nil {
return err
}
key := string(txkey.Prefix(index, field, view, shard))
_, _, err = tx.ImportRoaringBits(key, itr, clear, log, rowSize)
if err != nil {
tx.Rollback()
return err
}
err = tx.Commit()
if err != nil {
return err
}
}
cache.Close()
keys := FetchIndexKeys(dataDir)
for _, filename := range keys {
fmt.Println("index keys", filename)
content, err := ioutil.ReadFile(filepath.Join(dataDir, filename))
if err != nil {
return err
}
parts := strings.Split(filename, "/")
//destFile := fmt.Sprintf("%v/indexes/%v/translate/%v", backupPath, parts[1], parts[3])
destFile := filepath.Join(backupPath, "indexes", parts[1], "translaste", parts[3])
err = writeIfBigger(destFile, content)
if err != nil {
return err
}
}
//deal with index field(row)keys
keys = FetchRowkeys(dataDir)
for _, filename := range keys {
fmt.Println("field", filename)
content, err := ioutil.ReadFile(dataDir + filename)
if err != nil {
return err
}
parts := strings.Split(filename, "/")
//destFile := fmt.Sprintf("%v/indexes/%v/fields/%v/translate", backupPath, parts[1], parts[2])
destFile := filepath.Join(backupPath, "indexes", parts[1], "fields", parts[2], "translate")
err = writeIfBigger(destFile, content)
if err != nil {
return err
}
}
return nil
}
func writeIfBigger(dst string, content []byte) error {
if stats, err := os.Stat(dst); os.IsNotExist(err) {
err = os.MkdirAll(filepath.Dir(dst), 0777)
if err != nil {
return err
}
return ioutil.WriteFile(dst, content, 0644)
} else {
if stats.Size() < int64(len(content)) {
return ioutil.WriteFile(dst, content, 0644)
}
}
return nil //simply skip it
}
func ignore(path string, items ...string) bool {
f := filepath.Base(path)
for i := range items {
if items[i] == f {
return true
}
}
return false
}
func FetchIndexKeys(base string) []string {
var directory []string
ff := func(pathX string, infoX os.FileInfo, errX error) error {
// first thing to do, check error. and decide what to do about it
if errX != nil {
fmt.Printf("error 「%v」 at a path 「%q」\n", errX, pathX)
return errX
}
pathX = pathX[len(base):]
if infoX.IsDir() {
//filepath.Walk(pathX, ff)
} else {
if strings.Contains(pathX, "_keys") && !ignore(pathX, ".data", "keys") {
directory = append(directory, pathX)
}
}
return nil
}
err := filepath.Walk(base, ff)
if err != nil {
fmt.Printf("error walking the path %q: %v\n", base, err)
}
return directory
}
func FetchRowkeys(base string) []string {
var directory []string
ff := func(pathX string, infoX os.FileInfo, errX error) error {
// first thing to do, check error. and decide what to do about it
if errX != nil {
fmt.Printf("error 「%v」 at a path 「%q」\n", errX, pathX)
return errX
}
pathX = pathX[len(base):]
if infoX.IsDir() {
//filepath.Walk(pathX, ff)
} else {
fp := filepath.Base(pathX)
if fp == "keys" {
p := strings.Split(pathX, "/")
if len(p) != 4 {
return nil //skip all but field/key files
}
directory = append(directory, pathX)
}
}
return nil
}
err := filepath.Walk(base, ff)
if err != nil {
fmt.Printf("error walking the path %q: %v\n", base, err)
}
return directory
}