pitr: add ingest recorder to repair indexes #41670
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,24 @@ | ||
| load("@io_bazel_rules_go//go:def.bzl", "go_library", "go_test") | ||
|
|
||
| go_library( | ||
| name = "ingestrec", | ||
| srcs = ["ingest_recorder.go"], | ||
| importpath = "github.com/pingcap/tidb/br/pkg/restore/ingestrec", | ||
| visibility = ["//visibility:public"], | ||
| deps = [ | ||
| "//parser/model", | ||
| "//types", | ||
| "@com_github_pingcap_errors//:errors", | ||
| ], | ||
| ) | ||
|
|
||
| go_test( | ||
| name = "ingestrec_test", | ||
| srcs = ["ingest_recorder_test.go"], | ||
| deps = [ | ||
| ":ingestrec", | ||
| "//parser/model", | ||
| "@com_github_pkg_errors//:errors", | ||
| "@com_github_stretchr_testify//require", | ||
| ], | ||
| ) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,182 @@ | ||
| // Copyright 2023 PingCAP, Inc. | ||
| // | ||
| // 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 ingestrec | ||
|
|
||
| import ( | ||
| "fmt" | ||
| "strings" | ||
|
|
||
| "github.com/pingcap/errors" | ||
| "github.com/pingcap/tidb/parser/model" | ||
| "github.com/pingcap/tidb/types" | ||
| ) | ||
|
|
||
| // IngestIndexInfo records the information used to generate index drop/re-add SQL. | ||
| type IngestIndexInfo struct { | ||
| SchemaName string | ||
| TableName string | ||
| ColumnList string | ||
| IsPrimary bool | ||
| IndexInfo *model.IndexInfo | ||
| Updated bool | ||
| } | ||
|
|
||
| // IngestRecorder records the indexes information that use ingest mode to construct kvs. | ||
| // Currently log backup cannot backed up these ingest kvs. So need to re-construct them. | ||
| type IngestRecorder struct { | ||
| // Table ID -> Index ID -> Index info | ||
| items map[int64]map[int64]*IngestIndexInfo | ||
| } | ||
|
|
||
| // Return an empty IngestRecorder | ||
| func New() *IngestRecorder { | ||
| return &IngestRecorder{ | ||
| items: make(map[int64]map[int64]*IngestIndexInfo), | ||
| } | ||
| } | ||
|
|
||
| func notIngestJob(job *model.Job) bool { | ||
| return job.ReorgMeta == nil || | ||
| job.ReorgMeta.ReorgTp != model.ReorgTypeLitMerge | ||
| } | ||
|
|
||
| func notAddIndexJob(job *model.Job) bool { | ||
| /* support new index using accelerated indexing in future: | ||
| * // 1. skip if the new index didn't generate new kvs | ||
| * // 2. shall the ReorgTp of ModifyColumnJob be ReorgTypeLitMerge if use accelerated indexing? | ||
| * if job.RowCount > 0 && notIngestJob(job) { | ||
| * // ASSERT: select new indexes, which have the highest IDs in this job's BinlogInfo | ||
| * newIndexesInfo := getIndexesWithTheHighestIDs(len(indexIDs)) | ||
| * for _, newIndexInfo := range newIndexesInfo { | ||
| * tableindexes[newIndexInfo.ID] = ... | ||
| * } | ||
| * } | ||
| */ | ||
| return job.Type != model.ActionAddIndex && | ||
| job.Type != model.ActionAddPrimaryKey | ||
| } | ||
|
|
||
| func notSynced(job *model.Job) bool { | ||
| return job.State != model.JobStateSynced | ||
| } | ||
|
|
||
| // AddJob firstly filters the ingest index add operation job, and records it into IngestRecorder. | ||
| func (i *IngestRecorder) AddJob(job *model.Job) error { | ||
| if job == nil || notIngestJob(job) || notAddIndexJob(job) || notSynced(job) { | ||
| return nil | ||
| } | ||
|
|
||
| // If the add-index operation affects no row, the index doesn't need to be repair. | ||
| if job.RowCount == 0 { | ||
| return nil | ||
| } | ||
|
|
||
| var indexID int64 = 0 | ||
| if err := job.DecodeArgs(&indexID); err != nil { | ||
| return errors.Trace(err) | ||
| } | ||
|
|
||
| tableindexes, exists := i.items[job.TableID] | ||
| if !exists { | ||
| tableindexes = make(map[int64]*IngestIndexInfo) | ||
| i.items[job.TableID] = tableindexes | ||
| } | ||
|
|
||
| // the current information of table/index might be modified by other ddl jobs, | ||
| // therefore update the index information at last | ||
| tableindexes[indexID] = &IngestIndexInfo{ | ||
| IsPrimary: job.Type == model.ActionAddPrimaryKey, | ||
| Updated: false, | ||
| } | ||
tangenta marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
|
||
| return nil | ||
| } | ||
|
|
||
| // RerwiteIndexInfo rewrites the table id of the items in the IngestRecorder | ||
| func (i *IngestRecorder) RewriteTableID(rewriteFunc func(tableID int64) (int64, error)) error { | ||
| newItems := make(map[int64]map[int64]*IngestIndexInfo) | ||
| for tableID, item := range i.items { | ||
| newTableID, err := rewriteFunc(tableID) | ||
| if err != nil { | ||
| return errors.Annotatef(err, "failed to rewrite table id: %d", tableID) | ||
| } | ||
| newItems[newTableID] = item | ||
| } | ||
| i.items = newItems | ||
| return nil | ||
| } | ||
|
|
||
| // UpdateIndexInfo uses the newest schemas to update the ingest index's information | ||
| func (i *IngestRecorder) UpdateIndexInfo(dbInfos []*model.DBInfo) { | ||
| for _, dbInfo := range dbInfos { | ||
| for _, tblInfo := range dbInfo.Tables { | ||
| tableindexes, tblexists := i.items[tblInfo.ID] | ||
| if !tblexists { | ||
| continue | ||
| } | ||
| for _, indexInfo := range tblInfo.Indices { | ||
| index, idxexists := tableindexes[indexInfo.ID] | ||
| if !idxexists { | ||
| continue | ||
| } | ||
| var columnListBuilder strings.Builder | ||
| var isFirst bool = true | ||
| for _, column := range indexInfo.Columns { | ||
| if !isFirst { | ||
| columnListBuilder.WriteByte(',') | ||
| } | ||
| isFirst = false | ||
|
|
||
| // expression / column | ||
| col := tblInfo.Columns[column.Offset] | ||
| if col.Hidden { | ||
| // (expression) | ||
| columnListBuilder.WriteByte('(') | ||
| columnListBuilder.WriteString(col.GeneratedExprString) | ||
| columnListBuilder.WriteByte(')') | ||
| } else { | ||
| // `columnName` | ||
| columnListBuilder.WriteByte('`') | ||
| columnListBuilder.WriteString(column.Name.O) | ||
| columnListBuilder.WriteByte('`') | ||
| if column.Length != types.UnspecifiedLength { | ||
| columnListBuilder.WriteString(fmt.Sprintf("(%d)", column.Length)) | ||
| } | ||
| } | ||
| } | ||
| index.ColumnList = columnListBuilder.String() | ||
| index.IndexInfo = indexInfo | ||
| index.SchemaName = dbInfo.Name.O | ||
| index.TableName = tblInfo.Name.O | ||
| index.Updated = true | ||
| } | ||
| } | ||
| } | ||
| } | ||
|
|
||
| // Iterate iterates all the ingest index. | ||
| func (i *IngestRecorder) Iterate(f func(tableID int64, indexID int64, info *IngestIndexInfo) error) error { | ||
| for tableID, is := range i.items { | ||
| for indexID, info := range is { | ||
| if !info.Updated { | ||
| continue | ||
| } | ||
| if err := f(tableID, indexID, info); err != nil { | ||
| return errors.Trace(err) | ||
| } | ||
| } | ||
| } | ||
| return nil | ||
| } | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should we escape here? Perhaps some guys like to name their indices like(How absurd!):
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
😂
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done. test updated