Fix recover index bug when Flint data index is deleted accidentally

[dai-chen]

Description

To address the 2 issues related:

1. Quick fixed recover index API to clean up metadata log entry if Flint data index is gone. This prevents index stuck in refreshing state and infinite attempt on recover index API.
2. Added check to prevent FlintJob hang even though no streaming job launched by recover statement.

Documentation

Updated user manual: https://github.com/dai-chen/opensearch-spark/blob/fix-recover-index-for-index-data-deleted/docs/index.md#index-job-management

TODO

1. Integration test with FlintJob [Checking with @kaituo if possible to add IT]
2. [BUG] Gracefully terminate index refresh job when Flint index deleted accidentally #244

Testing

Manual test to double confirm. First of all, replicate the problematic scenario as outlined below:

CREATE SKIPPING INDEX ON stream.lineitem_tiny
(l_shipdate VALUE_SET)
WITH ( auto_refresh = true );

# Delete Flint data index
DELETE flint_myglue_stream_lineitem_tiny_skipping_index

# Check index state metadata log
GET .query_execution_request_myglue/_search
      {
        "_index": ".query_execution_request_myglue",
        "_id": "ZmxpbnRfbXlnbHVlX3N0cmVhbV9saW5laXRlbV90aW55X3NraXBwaW5nX2luZGV4",
        "_score": 1,
        "_source": {
          "version": "1.0",
          "latestId": "ZmxpbnRfbXlnbHVlX3N0cmVhbV9saW5laXRlbV90aW55X3NraXBwaW5nX2luZGV4",
          "type": "flintindexstate",
          "state": "refreshing",
          "applicationId": "unknown",
          "jobId": "unknown",
          "dataSourceName": "myglue",
          "jobStartTime": 1706916312007,
          "lastUpdateTime": 1706916452474,
          "error": ""
        }
      }

Now verify the enhanced recover index API:

spark-sql> RECOVER INDEX JOB flint_myglue_stream_lineitem_tiny_skipping_index;
24/02/02 23:48:31 WARN FlintSpark: Cleaning up metadata log as index data has been deleted

# The metadata log is gone
GET .query_execution_request_myglue/_search

Issues Resolved

• [BUG] Flint job hangs if no streaming job launched #227
• [BUG] Index state stuck in refreshing when Flint index deleted directly #237

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.

[vmmusings]

@dai-chen RECOVER INDEX JOB flint_myglue_stream_lineitem_tiny_skipping_index;
What is the output of this command? Will it be the same in all the cases.

[dai-chen]

@dai-chen RECOVER INDEX JOB flint_myglue_stream_lineitem_tiny_skipping_index; What is the output of this command? Will it be the same in all the cases.

We're following Spark DDL and return empty result if success in all Flint DDL statement.

[opensearch-trigger-bot]

The backport to 0.1 failed:

The process '/usr/bin/git' failed with exit code 128

To backport manually, run these commands in your terminal:

# Navigate to the root of your repository
cd $(git rev-parse --show-toplevel)
# Fetch latest updates from GitHub
git fetch
# Create a new working tree
git worktree add ../.worktrees/opensearch-spark/backport-0.1 0.1
# Navigate to the new working tree
pushd ../.worktrees/opensearch-spark/backport-0.1
# Create a new branch
git switch --create backport/backport-241-to-0.1
# Cherry-pick the merged commit of this pull request and resolve the conflicts
git cherry-pick -x --mainline 1 f4744abf38caeb4f22758f112a32c8881842efad
# Push it to GitHub
git push --set-upstream origin backport/backport-241-to-0.1
# Go back to the original working tree
popd
# Delete the working tree
git worktree remove ../.worktrees/opensearch-spark/backport-0.1

Then, create a pull request where the base branch is 0.1 and the compare/head branch is backport/backport-241-to-0.1.