fix(sink): set query timeout for jdbc sink to avoid stuck

[StrikeW]

I hereby agree to the terms of the RisingWave Labs, Inc. Contributor License Agreement.

What's changed and what's your intention?

• add jdbc.query.timeout to allow user set the timeout for query, default to 10mins if not provided.

related: #18372

Checklist

• I have written necessary rustdoc comments
• I have added necessary unit tests and integration tests
• I have added test labels as necessary. See details.
• I have added fuzzing tests or opened an issue to track them. (Optional, recommended for new SQL features Sqlsmith: Sql feature generation #7934).
• My PR contains breaking changes. (If it deprecates some features, please create a tracking issue to remove them in the future).
• All checks passed in ./risedev check (or alias, ./risedev c)
• My PR changes performance-critical code. (Please run macro/micro-benchmarks and show the results.)

• My PR contains critical fixes that are necessary to be merged into the latest release. (Please check out the details)

Documentation

• My PR needs documentation updates. (Please use the Release note section below to summarize the impact on users)

Release note

If this PR includes changes that directly affect users or other significant modifications relevant to the community, kindly draft a release note to provide a concise summary of these changes. Please prioritize highlighting the impact these changes will have on users.

[fuyufjh]

Better to make it a configurable WITH arguments of create sink...

[hzxa21]

The PR description mentions that this PR can fix #18372. Do we confirm that this can actually resolve the issue? If it is caused by a deadlock mentioned in pgJDBC, I suspect that setting the timeout may not work.

It is okay to merge this PR though because the statement timeout may help in other cases.

[StrikeW]

The PR description mentions that this PR can fix #18372. Do we confirm that this can actually resolve the issue? If it is caused by a deadlock mentioned in pgJDBC, I suspect that setting the timeout may not work.

It is okay to merge this PR though because the statement timeout may help in other cases.

Since the stuck problem is not always reproducible, I didn't find a way to confirm it.
This pr is a best effort try to fix the stuck issue indeed.

[hzxa21]

LGTM.

If we cannot confirm whether this PR can fix the issue, I suggest we keep the original issue opened until we rollout the fix and confirm it works.

How about cherry-picking it to 2.0 as well?

[StrikeW]

If we cannot confirm whether this PR can fix the issue, I suggest we keep the original issue opened until we rollout the fix and confirm it works.

Sounds ok to me.

How about cherry-picking it to 2.0 as well?

Yeah, I have labeled it.

[wenym1]

Just take a look at the code of pgjdbc. It seems that when this timeout is set, in every executeBatch, it will set a timer task with this timeout value. When timeout is reached, it will send separate cancel query from a separate PgStream. So even if the query is stuck in lower level connection, we are still able to cancel the query. So this setting can probably work.

Btw, can users specify the statement timeout in the jdbc connection string, so that for previous version, they can also set the timeout in the jdbc connection string?

[StrikeW]

Just take a look at the code of pgjdbc. It seems that when this timeout is set, in every executeBatch, it will set a timer task with this timeout value. When timeout is reached, it will send separate cancel query from a separate PgStream. So even if the query is stuck in lower level connection, we are still able to cancel the query. So this setting can probably work.

Btw, can users specify the statement timeout in the jdbc connection string, so that for previous version, they can also set the timeout in the jdbc connection string?

From the doc, I think users of previous versions could add the ?options=-c%20statement_timeout=5min to the jdbc url to set the timeout for a statement.