[Compaction] Do not move the non-durable cursor position when trimming ledgers while topic with compaction - #12602
Merged
merlimat merged 4 commits intoNov 4, 2021
Conversation
…g ledgers while topic with compaction. For the non-durable cursor, the ledgers trimming task will cause skip the removed ledgers to avoid readers introduced backlogs and make sure the data can be removed if over the retention, more details to see apache#6787. But for a topic which enabled compaction, this will lead to the reader skips the compacted data. The new added test can illustrate this problem well. For reading compacted data, reading a message ID that earlier that the first message ID of the original data is a normal behavior, so we should not move forward the cursor which will read the compacted data.
|
@codelipenghui:Thanks for providing doc info! |
codelipenghui
added
release/2.8.2
release/2.9.1
and removed
doc-not-needed
codelipenghui
requested review from
315157973,
congbobo184,
gaoran10,
hangc0276,
merlimat and
sijie
November 3, 2021 10:35
|
@codelipenghui:Thanks for your contribution. For this PR, do we need to update docs? |
1 similar comment
|
@codelipenghui:Thanks for your contribution. For this PR, do we need to update docs? |
Contributor
Author
|
/pulsarbot run-failure-checks |
Anonymitaet
added
doc-not-needed
merlimat
approved these changes
Nov 4, 2021
codelipenghui
deleted the
penghui/fix-read-compacted-lost-data-trimming-ledger
branch
November 5, 2021 00:33
codelipenghui
added a commit
that referenced
this pull request
Nov 5, 2021
…g ledgers while topic with compaction (#12602) * [Compaction] Do not move the non-durable cursor position when trimming ledgers while topic with compaction. For the non-durable cursor, the ledgers trimming task will cause skip the removed ledgers to avoid readers introduced backlogs and make sure the data can be removed if over the retention, more details to see #6787. But for a topic which enabled compaction, this will lead to the reader skips the compacted data. The new added test can illustrate this problem well. For reading compacted data, reading a message ID that earlier that the first message ID of the original data is a normal behavior, so we should not move forward the cursor which will read the compacted data. * Fix checkstyle. * Fix tests. * Fix tests. (cherry picked from commit a6b1b34)
eolivelli
pushed a commit
to eolivelli/pulsar
that referenced
this pull request
Nov 29, 2021
…g ledgers while topic with compaction (apache#12602) * [Compaction] Do not move the non-durable cursor position when trimming ledgers while topic with compaction. For the non-durable cursor, the ledgers trimming task will cause skip the removed ledgers to avoid readers introduced backlogs and make sure the data can be removed if over the retention, more details to see apache#6787. But for a topic which enabled compaction, this will lead to the reader skips the compacted data. The new added test can illustrate this problem well. For reading compacted data, reading a message ID that earlier that the first message ID of the original data is a normal behavior, so we should not move forward the cursor which will read the compacted data. * Fix checkstyle. * Fix tests. * Fix tests.
codelipenghui
added a commit
that referenced
this pull request
Dec 20, 2021
…g ledgers while topic with compaction (#12602) * [Compaction] Do not move the non-durable cursor position when trimming ledgers while topic with compaction. For the non-durable cursor, the ledgers trimming task will cause skip the removed ledgers to avoid readers introduced backlogs and make sure the data can be removed if over the retention, more details to see #6787. But for a topic which enabled compaction, this will lead to the reader skips the compacted data. The new added test can illustrate this problem well. For reading compacted data, reading a message ID that earlier that the first message ID of the original data is a normal behavior, so we should not move forward the cursor which will read the compacted data. * Fix checkstyle. * Fix tests. * Fix tests. (cherry picked from commit a6b1b34)
codelipenghui
restored the
penghui/fix-read-compacted-lost-data-trimming-ledger
branch
May 17, 2022 01:23
codelipenghui
deleted the
penghui/fix-read-compacted-lost-data-trimming-ledger
branch
May 17, 2022 01:29
This was referenced Jul 5, 2022
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 participants
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.
Motivation
For the non-durable cursor, the ledgers trimming task will cause skip the removed ledgers
to avoid readers introduced backlogs and make sure the data can be removed if over the retention,
more details to see #6787.
But for a topic that enabled compaction, this will lead to the reader skipping the compacted data.
The newly added test can illustrate this problem well. For reading compacted data, reading a message ID
that earlier that the first message ID of the original data is normal behavior, so we should not
move forward the cursor which will read the compacted data.
Documentation
Check the box below and label this PR (if you have committer privilege).
Need to update docs?
doc-required(If you need help on updating docs, create a doc issue)
no-need-doc(Please explain why)
doc(If this PR contains doc changes)