Fix: ml/engine/utils/FileUtils casts long file length to int incorrectly

[maxlepikhin]

Description

"(int) file.length()" makes length negative for file sizes greater than 2GB (but less than 4GB). This results in function returning empty list of chunks and model registration task being stuck in CREATED state.

The fix is to use longs when splitting model zip file. Tested locally that updated opensearch-ml-algorithms jar fixes the problem.

Bug: #3197

Related Issues

N/A

Check List

• New functionality includes testing.
• New functionality has been documented.
• API changes companion pull request created.
• Commits are signed per the DCO using --signoff.
• Public documentation issue/PR created.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[pyek-bot]

Should we create an issue for this and link it to the PR? @ylwu-amzn

Edit: I see that it is here already #3197
@maxlepikhin can you add it to the description?

[mingshl]

@maxlepikhin in every commit, you need to commit with your sign off using -s, for example. git commit -m"commit message" -s

Your last two commits are missing sign off, you can fix it by the following:

To add your Signed-off-by line to every commit in this branch:

Ensure you have a local copy of your branch by checking out the pull request locally via command line.
In your local branch, run: git rebase HEAD~2 --signoff
Force push your changes to overwrite the branch: git push --force-with-lease origin fix-3197

[b4sjoo]

Seems need to run spotlessApply

[brianf-aws]

Hey @maxlepikhin ! Just curious, how did you debug this?

[maxlepikhin]

Hey @maxlepikhin ! Just curious, how did you debug this?

By reading the code.

[maxlepikhin]

will approve after all tests passed.

@maxlepikhin can you identify the version when this bug is happening? trying to figure out the backport versions cc @ylwu-amzn

From 11/17/22 (bfb0748): all releases it seems. It'd be great if somebody from the maintainers can help the tests pass, are they flaky?

[mingshl]

It's a flaky test.

Created the issue to track it.

Approved.

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField STANDARD_ERROR
    REPRODUCE WITH: ./gradlew ':opensearch-ml-plugin:integTest' --tests "org.opensearch.ml.rest.RestMLInferenceSearchResponseProcessorIT.testMLInferenceProcessorRemoteModelStringField" -Dtests.seed=9E7BCE94AFC0318E -Dtests.security.manager=false -Dtests.locale=luy-KE -Dtests.timezone=Asia/Dubai -Druntime.java=21

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField FAILED
    org.opensearch.client.ResponseException: method [POST], host [http://127.0.0.1:33169/], URI [/_plugins/_ml/models/null/_deploy], status line [HTTP/1.1 404 Not Found]
    {"error":{"root_cause":[{"type":"status_exception","reason":"Failed to find model"}],"type":"status_exception","reason":"Failed to find model"},"status":404}

[maxlepikhin]

It's a flaky test.

Created the issue to track it.

Approved.

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField STANDARD_ERROR
    REPRODUCE WITH: ./gradlew ':opensearch-ml-plugin:integTest' --tests "org.opensearch.ml.rest.RestMLInferenceSearchResponseProcessorIT.testMLInferenceProcessorRemoteModelStringField" -Dtests.seed=9E7BCE94AFC0318E -Dtests.security.manager=false -Dtests.locale=luy-KE -Dtests.timezone=Asia/Dubai -Druntime.java=21

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField FAILED
    org.opensearch.client.ResponseException: method [POST], host [http://127.0.0.1:33169/], URI [/_plugins/_ml/models/null/_deploy], status line [HTTP/1.1 404 Not Found]
    {"error":{"root_cause":[{"type":"status_exception","reason":"Failed to find model"}],"type":"status_exception","reason":"Failed to find model"},"status":404}

Ok, thanks @mingshl . How to rerun it or override to submit?

[maxlepikhin]

It's a flaky test.
Created the issue to track it.
Approved.

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField STANDARD_ERROR
    REPRODUCE WITH: ./gradlew ':opensearch-ml-plugin:integTest' --tests "org.opensearch.ml.rest.RestMLInferenceSearchResponseProcessorIT.testMLInferenceProcessorRemoteModelStringField" -Dtests.seed=9E7BCE94AFC0318E -Dtests.security.manager=false -Dtests.locale=luy-KE -Dtests.timezone=Asia/Dubai -Druntime.java=21

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField FAILED
    org.opensearch.client.ResponseException: method [POST], host [http://127.0.0.1:33169/], URI [/_plugins/_ml/models/null/_deploy], status line [HTTP/1.1 404 Not Found]
    {"error":{"root_cause":[{"type":"status_exception","reason":"Failed to find model"}],"type":"status_exception","reason":"Failed to find model"},"status":404}

Ok, thanks @mingshl . How to rerun it or override to submit?

Can maintainers advise how to either re-run flaky tests or override the results to submit this PR?
@mingshl @pyek-bot @ylwu-amzn - can you help please?

[brianf-aws]

Hey Max @maxlepikhin will reach out to my team to rerun the workflow. Thank you for your patience, as of now only maintainers can rerun the workflow.

[dhrubo-os]

It's a flaky test.
Created the issue to track it.
Approved.

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField STANDARD_ERROR
    REPRODUCE WITH: ./gradlew ':opensearch-ml-plugin:integTest' --tests "org.opensearch.ml.rest.RestMLInferenceSearchResponseProcessorIT.testMLInferenceProcessorRemoteModelStringField" -Dtests.seed=9E7BCE94AFC0318E -Dtests.security.manager=false -Dtests.locale=luy-KE -Dtests.timezone=Asia/Dubai -Druntime.java=21

RestMLInferenceSearchResponseProcessorIT > testMLInferenceProcessorRemoteModelStringField FAILED
    org.opensearch.client.ResponseException: method [POST], host [http://127.0.0.1:33169/], URI [/_plugins/_ml/models/null/_deploy], status line [HTTP/1.1 404 Not Found]
    {"error":{"root_cause":[{"type":"status_exception","reason":"Failed to find model"}],"type":"status_exception","reason":"Failed to find model"},"status":404}

Ok, thanks @mingshl . How to rerun it or override to submit?

Can maintainers advise how to either re-run flaky tests or override the results to submit this PR? @mingshl @pyek-bot @ylwu-amzn - can you help please?

I tried to re-run the workflow, but github isn't letting me to do that. May be you can push another commit based on the provided suggestion which will run workflow again.