You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by "micheal-o (via GitHub)" <gi...@apache.org> on 2026/03/09 20:12:19 UTC
[PR] [SPARK-55892][SS] Fix unable to load state store because reused SST file was deleted by maintenance [spark]
micheal-o opened a new pull request, #54697:
URL: https://github.com/apache/spark/pull/54697
<!--
Thanks for sending a pull request! Here are some tips for you:
1. If this is your first time, please read our contributor guidelines: https://spark.apache.org/contributing.html
2. Ensure you have added or run the appropriate tests for your PR: https://spark.apache.org/developer-tools.html
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g., '[WIP][SPARK-XXXX] Your PR title ...'.
4. Be sure to keep the PR description updated to reflect all changes.
5. Please write your PR title to summarize what this PR proposes.
6. If possible, provide a concise example to reproduce the issue for a faster review.
7. If you want to add a new configuration, please read the guideline first for naming configurations in
'core/src/main/scala/org/apache/spark/internal/config/ConfigEntry.scala'.
8. If you want to add or modify an error type or message, please read the guideline first in
'common/utils/src/main/resources/error/README.md'.
-->
### What changes were proposed in this pull request?
<!--
Please clarify what changes you are proposing. The purpose of this section is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster reviews in your PR. See the examples below.
1. If you refactor some codes with changing classes, showing the class hierarchy will help reviewers.
2. If you fix some SQL features, you can provide some references of other DBMSes.
3. If there is design documentation, please add the link.
4. If there is a discussion in the mailing list, please add the link.
-->
When there is a large delay between when a snapshot was created and when it was uploaded by maintenance, it is possible for the files that were reused from old versions, which are now eligible for deletion by another executor's maintenance to be deleted.
Hence, when trying to load the state store using the uploaded snapshot, it will fail with FileNotFoundException for the reused SST file.
### Why are the changes needed?
<!--
Please clarify why the changes are needed. For instance,
1. If you propose a new API, clarify the use case for a new API.
2. If you fix a bug, you can clarify why it is a bug.
-->
To prevent state store corruption and streaming query failure
### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such as new features, bug fixes, or other behavior changes. Documentation-only updates are not considered user-facing changes.
If yes, please clarify the previous behavior and the change this PR proposes - provide the console output, description and/or an example to show the behavior difference if possible.
If possible, please also clarify if this is a user-facing change compared to the released Spark versions or within the unreleased branches such as master.
If no, write 'No'.
-->
No
### How was this patch tested?
<!--
If tests were added, say they were added here. Please make sure to add some test cases that check the changes thoroughly including negative and positive cases if possible.
If it was tested in a way different from regular unit tests, please clarify how you tested step by step, ideally copy and paste-able, so that other reviewers can test and check, and descendants can verify in the future.
If tests were not added, please describe why they were not added and/or why it was difficult to add.
If benchmark tests were added, please run the benchmarks in GitHub Actions for the consistent environment, and the instructions could accord to: https://spark.apache.org/developer-tools.html#github-workflow-benchmarks.
-->
New test case
### Was this patch authored or co-authored using generative AI tooling?
<!--
If generative AI tooling has been used in the process of authoring this patch, please include the
phrase: 'Generated-by: ' followed by the name of the tool and its version.
If no, write 'No'.
Please refer to the [ASF Generative Tooling Guidance](https://www.apache.org/legal/generative-tooling.html) for details.
-->
Claude-4.5-opus
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
Re: [PR] [SPARK-55892][SS] Fix unable to load state store because reused SST file was deleted by maintenance [spark]
Posted by "anishshri-db (via GitHub)" <gi...@apache.org>.
anishshri-db commented on code in PR #54697:
URL: https://github.com/apache/spark/pull/54697#discussion_r2907960015
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/state/RocksDB.scala:
##########
@@ -1869,6 +1853,39 @@ class RocksDB(
logInfo(log"Cleaned old data, time taken: ${MDC(LogKeys.TIME_UNITS, cleanupTime)} ms")
}
+ /**
+ * This replaces stale reused files in the snapshot with new ones to be uploaded.
+ * Stale means they are potential candidates for deletion by another
+ * maintenance that could have happened on another executor before this snapshot is uploaded.
+ * Checking if the file exists in DFS is not sufficient, since that could lead to race
+ * with an ongoing maintenance on another executor.
Review Comment:
nit: can you also add comments for what arg is passed and what is returned
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
Re: [PR] [SPARK-55892][SS] Fix unable to load state store because reused SST file was deleted by maintenance [spark]
Posted by "micheal-o (via GitHub)" <gi...@apache.org>.
micheal-o commented on code in PR #54697:
URL: https://github.com/apache/spark/pull/54697#discussion_r2908675116
##########
sql/core/src/test/scala/org/apache/spark/sql/execution/streaming/state/RocksDBSuite.scala:
##########
@@ -3982,6 +3982,70 @@ class RocksDBSuite extends AlsoTestWithRocksDBFeatures with SharedSparkSession
}
}
+ testWithChangelogCheckpointingEnabled(
+ "SPARK-55892: Stale reused files in snapshot are replaced during maintenance upload") {
+ // This test verifies that when a snapshot is uploaded after a delay, files that were
+ // reused from old versions (which are now eligible for deletion by another executor's
+ // maintenance) are replaced with new files to be uploaded.
+ //
+ // Scenario:
+ // db1 loads v0 and commits v1. Generates 1.changelog and 1.zip (but not uploaded to DFS)
+ // db2 loads v1 but since no snapshot, it uses 1.changelog and generates snapshot 2.zip
+ // db 1 maintenance uploads snapshot 1.zip to DFS
+ // db 3 loads v2 using 1.zip + 2.changelog and generates 3.zip but not uploaded to DFS
+ // db2 maintenance uploads 2.zip, and deletes files < 2.zip, because minVersionsToRetain = 1
+ // db 3 uploads 3.zip to DFS <-- (with new fix, shouldn't include reused files from 1.zip)
+ // db 4 loads v3 using 3.zip, should be successful (previously cause FileNotFoundException)
+
+ val conf = dbConf.copy(
+ minVersionsToRetain = 1,
+ minDeltasForSnapshot = 0, // Force snapshot on every commit
Review Comment:
@anishshri-db That's not needed, the result is the same. I'm testing this way to have a more predictable test behavior and avoid flakiness.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
Re: [PR] [SPARK-55892][SS] Fix unable to load state store because reused SST file was deleted by maintenance [spark]
Posted by "anishshri-db (via GitHub)" <gi...@apache.org>.
anishshri-db commented on code in PR #54697:
URL: https://github.com/apache/spark/pull/54697#discussion_r2907964489
##########
sql/core/src/test/scala/org/apache/spark/sql/execution/streaming/state/RocksDBSuite.scala:
##########
@@ -3982,6 +3982,70 @@ class RocksDBSuite extends AlsoTestWithRocksDBFeatures with SharedSparkSession
}
}
+ testWithChangelogCheckpointingEnabled(
+ "SPARK-55892: Stale reused files in snapshot are replaced during maintenance upload") {
+ // This test verifies that when a snapshot is uploaded after a delay, files that were
+ // reused from old versions (which are now eligible for deletion by another executor's
+ // maintenance) are replaced with new files to be uploaded.
+ //
+ // Scenario:
+ // db1 loads v0 and commits v1. Generates 1.changelog and 1.zip (but not uploaded to DFS)
+ // db2 loads v1 but since no snapshot, it uses 1.changelog and generates snapshot 2.zip
+ // db 1 maintenance uploads snapshot 1.zip to DFS
+ // db 3 loads v2 using 1.zip + 2.changelog and generates 3.zip but not uploaded to DFS
+ // db2 maintenance uploads 2.zip, and deletes files < 2.zip, because minVersionsToRetain = 1
+ // db 3 uploads 3.zip to DFS <-- (with new fix, shouldn't include reused files from 1.zip)
+ // db 4 loads v3 using 3.zip, should be successful (previously cause FileNotFoundException)
+
+ val conf = dbConf.copy(
+ minVersionsToRetain = 1,
+ minDeltasForSnapshot = 0, // Force snapshot on every commit
Review Comment:
Should we also have a test where snapshots are intermittent/infrequent ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
Re: [PR] [SPARK-55892][SS] Fix unable to load state store because reused SST file was deleted by maintenance [spark]
Posted by "anishshri-db (via GitHub)" <gi...@apache.org>.
anishshri-db closed pull request #54697: [SPARK-55892][SS] Fix unable to load state store because reused SST file was deleted by maintenance
URL: https://github.com/apache/spark/pull/54697
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org
Re: [PR] [SPARK-55892][SS] Fix unable to load state store because reused SST file was deleted by maintenance [spark]
Posted by "anishshri-db (via GitHub)" <gi...@apache.org>.
anishshri-db commented on code in PR #54697:
URL: https://github.com/apache/spark/pull/54697#discussion_r2907958354
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/state/RocksDB.scala:
##########
@@ -1857,7 +1830,18 @@ class RocksDB(
}
if (mostRecentSnapshot.isDefined) {
- uploadSnapshot(mostRecentSnapshot.get)
+ // SPARK-55892: We are doing this because the time between when a snapshot was created,
+ // and the time when maintenance upload is happening could be very long, such that the
+ // reused files could have been deleted by maintenance on another executor due to
Review Comment:
nit: `or could be deleted by the same executor with the instance not being reloaded` ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org