Clean up the unused sync from DynamoDB to benchmark.oss_ci_benchmark_v2

[huydhn]

This route is not used anymore, and always fails. I will need to clean up the write into dynamoDB in another PR and the dynamoDB table torchci-oss-ci-benchmark itself later.

UPSERTING 3 INTO benchmark.oss_ci_benchmark_v2
--
/var/task/lambda_function.py:206: UserWarning: Failed to upsert 3 into benchmark.oss_ci_benchmark_v2: HTTPDriver for https://hyt81izu0c.us-east-1.aws.clickhouse.cloud:8443 received ClickHouse error code 60
Code: 60. DB::Exception: Table benchmark.oss_ci_benchmark_v2 does not exist. Maybe you meant benchmark.oss_ci_benchmark_v3?. (UNKNOWN_TABLE) (version 25.4.1.37615 (official build))
warn(f"Failed to upsert {len(documents)} into {table}: {error}")
[ERROR] DatabaseError: HTTPDriver for https://hyt81izu0c.us-east-1.aws.clickhouse.cloud:8443 received ClickHouse error code 60 Code: 60. DB::Exception: Table benchmark.oss_ci_benchmark_v2 does not exist. Maybe you meant benchmark.oss_ci_benchmark_v3?. (UNKNOWN_TABLE) (version 25.4.1.37615 (official build))Traceback (most recent call last):  File "/var/task/lambda_function.py", line 58, in lambda_handler    handle_event(event, False)  File "/var/task/lambda_function.py", line 83, in handle_event    upsert_documents(table, ids_and_docs, dry_run)  File "/var/task/lambda_function.py", line 219, in upsert_documents    raise error  File "/var/task/lambda_function.py", line 203, in upsert_documents    res = get_clickhouse_client().query(query)  File "/var/task/clickhouse_connect/driver/client.py", line 228, in query    return self._query_with_context(query_context)  File "/var/task/clickhouse_connect/driver/httpclient.py", line 239, in _query_with_context    response = self._raw_request(body,  File "/var/task/clickhouse_connect/driver/httpclient.py", line 472, in _raw_request    self._error_handler(response)  File "/var/task/clickhouse_connect/driver/httpclient.py", line 395, in _error_handler    raise OperationalError(err_str) if retried else DatabaseError(err_str) from None

The error dropped to 0 after I removed this line:

Also remove the print json.dumps(body) part because it's writing too much information into the log. We needed it at the start to debug issue, but it's too verbose now. We can remove the whole print statement now if no one really looks into it. @clee2000 Your thoughts on this? I'll do this in a separate PR to improve the logging mechanism here instead of using poor man print

[vercel]

@huydhn is attempting to deploy a commit to the Meta Open Source Team on Vercel.

A member of the Team first needs to authorize it.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Preview	Updated (UTC)
torchci	Ignored	Preview	Aug 18, 2025 7:34pm

[zxiiro]

Could we do like a log.debug() and then not print the debug statement unless we need it?

That way we can keep the line but have it off by default.

[huydhn]

This makes sense, let me do that. The logging of this important lambda could be improved

[huydhn]

Now, I think it's better to break this into 2 PR. This will just remove the unused sync from DynamoDB to benchmark.oss_ci_benchmark_v2. I will do the logging part separately as it's more involved.