Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Avoid crashes in rowGroupUncompressedSize due to column index being o… #9223

Closed

Conversation

makagonov
Copy link

fixes #9222

Turns out that in the getParquetColumnInfo function, the column was incremented twice for array types, which would result in out-of-bounds vector access down the road. We could confirm it in Uber by deploying the VELOX_CHECK_LT statements in the code, where we started seeing tons of logs like:

VeloxRuntimeError: type.column() < fileMetaData_->row_groups[rowGroupIndex].columns.size() (228 vs. 228) type column: 228, columns.size(): 228 Split [Hive: gs://....parquet 671088640 - 134217728] Task 20240308_025912_03027_rt9ch.1.0.32.0

The crash was not happening frequently because the [] operator on vector does not do out-of-bounds check and would result in an undefined behavior in case if the code tries to access an element at wrong index. Undefined behavior does not necessarily mean you'll get an error: you might, but you might instead get some result that doesn't make much sense.

@facebook-github-bot
Copy link
Contributor

Hi @makagonov!

Thank you for your pull request.

We require contributors to sign our Contributor License Agreement, and yours needs attention.

You currently have a record in our system, but the CLA is no longer valid, and will need to be resubmitted.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at cla@meta.com. Thanks!

Copy link

netlify bot commented Mar 22, 2024

Deploy Preview for meta-velox canceled.

Name Link
🔨 Latest commit 5df577e
🔍 Latest deploy log https://app.netlify.com/sites/meta-velox/deploys/65fddeba94dcc3000823bd96

@makagonov makagonov marked this pull request as draft March 22, 2024 19:46
qqibrow pushed a commit to qqibrow/velox that referenced this pull request Mar 28, 2024
qqibrow pushed a commit to qqibrow/velox that referenced this pull request Mar 28, 2024
@@ -418,7 +418,7 @@ std::shared_ptr<const ParquetTypeWithId> ReaderBase::getParquetColumnInfo(
std::move(children),
curSchemaIdx,
maxSchemaElementIdx,
columnIdx++,
columnIdx - 1, // was already incremented for leafTypePtr
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

for my understanding, why here it's not ParquetTypeWithId::kNonLeaf but columnIdx - 1?

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the else statement is for the leaf type columns. If the column is a repeated type (like array), it would have the same column index as the column itself.

@yingsu00
Copy link
Collaborator

closing with #9187

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Native work crashes in ParquetRowReader::Impl::estimatedRowSize
4 participants