jitendrakr88 opened a new pull request, #15271:
URL: https://github.com/apache/pinot/pull/15271
## Description
When OOM is enabled, the grpc streming queries always return 0 rows. This
issue has been observered with spark connector as well as trino connector
(reported by another user).
```py
from pyspark.sql.functions import col
cutoff_seconds_since_spoch = 1741169418
df = spark \
.read \
.format("pinot") \
.option("controller", "<host>:<port>") \
.option("table", "<table>") \
.option("useGrpcServer", "true") \
.option("tableType", "REALTIME") \
.option("segmentsPerSplit", 1) \
.load() \
.filter(col("secondsSinceEpoch") >= cutoff_seconds_since_spoch)
print("Number of rows: " + str(df.count()))
```
<img width="358" alt="image"
src="https://github.com/user-attachments/assets/ba264900-0b35-4fd3-9f82-c8d99d1402db";
/>
### Observation:
- Number of rows return is always zero if OOM is enabled on server tenant
- After disabling OOM, the grpc streaming query works fine.
- After disabling `useGrpcServer` option, the query works fine with http
transport.
## Solution
This PR fixes the issue by initialising the missing tracing context for OOM
protection.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
