[HUDI-7919][DNM] Migrate integration tests to run on Spark 3.5 #11994
+134
−89
Conversation
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
yihua
changed the title
yihua
force-pushed
the
HUDI-7919-spark35-it
branch
3 times, most recently
from
079bc1c
to
9cddf3e
Compare
github-actions
bot
added
size:M
yihua
force-pushed
the
HUDI-7919-spark35-it
branch
from
9cddf3e
to
c2f4791
Compare
yihua
force-pushed
the
HUDI-7919-spark35-it
branch
from
c2f4791
to
381e3a0
Compare
yihua
commented
Sep 28, 2024
| @@ -37,11 +36,12 @@ services: | |||
| retries: 3 | |||
|
|
|||
| datanode1: | |||
| image: apachehudi/hudi-hadoop_2.8.4-datanode:latest | |||
| image: apachehudi/hudi-hadoop_2.8.4-datanode:bullseye | |||
There was a problem hiding this comment.
The image tag will be reverted from bullseye to latest once the PR is close to merging.
| @@ -17,7 +17,7 @@ | |||
|
|
|||
| ARG HADOOP_VERSION=2.8.4 | |||
| ARG HADOOP_DN_PORT=50075 | |||
| FROM apachehudi/hudi-hadoop_${HADOOP_VERSION}-base:latest | |||
| FROM apachehudi/hudi-hadoop_${HADOOP_VERSION}-base:bullseye | |||
There was a problem hiding this comment.
Similar here in Dockerfiles: the image tag will be reverted from bullseye to latest once the PR is close to merging.
| @@ -268,11 +269,12 @@ | |||
| </run> | |||
| </image> | |||
| <image> | |||
| <name>motoserver/moto:${moto.version}</name> | |||
| <name>apachehudi/moto:${moto.version}</name> | |||
There was a problem hiding this comment.
We have to do this to pull the image of the correct architecture as the plugin cannot do this properly.
Comment on lines
+63
to
+65
| // This port number must be the same as {@code moto.port} defined in pom.xml | ||
| private static final int MOTO_PORT = 5002; | ||
| private static final String MOTO_ENDPOINT = "http://localhost:" + MOTO_PORT; |
There was a problem hiding this comment.
To make this integration test run locally.
| @@ -170,7 +170,7 @@ private boolean checkHealth(String fromContainerName, String hostname, int port) | |||
| TestExecStartResultCallback resultCallback = | |||
| executeCommandStringInDocker(fromContainerName, command, false, true); | |||
| String stderrString = resultCallback.getStderr().toString().trim(); | |||
| if (!stderrString.contains("open")) { | |||
| if (!stderrString.contains("succeeded")) { | |||
There was a problem hiding this comment.
Output message change.
| assertStdOutContains(stdOutErrPair, | ||
| "|default |stock_ticks_cow |false |\n" | ||
| + "|default |stock_ticks_cow_bs |false |\n" | ||
| + "|default |stock_ticks_mor |false |\n" |
There was a problem hiding this comment.
Now the original table name is also synced to the metastore based on the recent behavior change (#10685).
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Change Logs
This PR migrates the integration tests from running on Spark 2.4 to running on Spark 3.5. Changes include:
hudi-hive-sync-bundleto include necessary jackson and parquet classes in the bundle to avoid missing or conflicted classes in Hive sync;hudi-sync/hudi-hive-sync/run_sync_tool.shto avoid jackson dependency conflict;integration-teststask in.github/workflows/bot.ymlto run on Spark 3.5;linux/amd64architecture only in this PR);Dockerfiles are changed. Debian stretch release is no longer supported. Debian bullseye release is used instead as the base.hudi-aws/pom.xmlso that we can pull themotoimage based on the correct architecture.apachehudi/moto imageis uploaded. Moto port is changed so it can run locally on the Macbook to avoid port collision;ITTestBase#checkHealthbased on the new output;ITTestHoodieDem#testParquetDemoto verify that Spark job and queries work fine in the new docker demo setup;ITTestHoodieSanityandITTestHoodieSyncCommandwhich are already covered by other tests (HUDI-8274 to revisit).New docker images are uploaded to the Docker Hub:
apachehudi/motoapachehudi/hudi-hadoop_2.8.4-baseapachehudi/hudi-hadoop_2.8.4-datanodeapachehudi/hudi-hadoop_2.8.4-historyapachehudi/hudi-hadoop_2.8.4-hive_2.3.3apachehudi/hudi-hadoop_2.8.4-hive_2.3.3-sparkbase_3.5.3apachehudi/hudi-hadoop_2.8.4-hive_2.3.3-sparkadhoc_3.5.3apachehudi/hudi-hadoop_2.8.4-hive_2.3.3-sparkmaster_3.5.3apachehudi/hudi-hadoop_2.8.4-hive_2.3.3-sparkworker_3.5.3Old images under the same names are pushed to a different tag,
stretch, in case we'd like to use that with Spark 2.4.Impact
Makes integration test run on Spark 3.5, to unblock deprecation of Spark 2 integration in Hudi.
Risk level
low
Documentation Update
N/A
Contributor's checklist