-
Notifications
You must be signed in to change notification settings - Fork 371
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Java py4j gateway server for trino queries still be open even after query finishes and in later makes queris Hang #3223
Comments
@divincode Thanks a lot for the detailed report and suggested solution. I've added the Good first issue but will also assign it to one of our core devs according to our new github issue process https://gethue.com/blog/hue-community-2023/ |
We also met this error, it will cause py4j jdbc connections leakage, and hue's memory grows up quickly |
which engine are u using? |
We are also using trino 433 jdbc with hue, you mentioned that we can use this code to fix this:
Thanks a lot, we will try to modify the source code like this~ |
I think this wont work though completely .. Do try but i think we should use https://github.com/trinodb/trino-python-client , make the code changes to use this
use this this should remove py4j competely |
@divincode we find this problem |
If it works can u please contribute back to the community ? @hackeryang @zhuzhengjun01 |
I tried use shutdown interface in the close function , but it does't work well, py4j gateway did not shutdown correctly |
No i mean the trino python client, would solve the use case. Please try that. |
Oh understand, we extended hue jdbc_trino, but use trino java jdbc client, change to python client will be a big change, I will try again |
The main thing is that the jdbc connector does not support set sessions, and some of our trino syntax depends on this |
I have tried using python client to connect to trino, the rss memory is fine, no more memory leaks, I will contribute this connector to the community if needed |
Nice to hear ! do contribute it back or at least paste the diff , so that it can be used by others. |
|
Is there an existing issue for this?
Description
The default connector for trino is jdbc as there is no sqlalchemy support as per my knowledge from sqlalchemy community.
When running hue with multiple clients, the jdbc class calls the py4j server, which still exist it does not close after we get the results, this leads to memory usage and hence the queries get slower as time passes and after certain time it still hangs . eg
For executing the trino query - code flow goes like this -
I have added various debugging points , to check where the bottleneck is
the bottle neck is this point -
data = curs.fetchmany(n) this line is the bottleneck usually.
The issue should be solved if add the following lines in the close fxn of this file in jdbc class -
desktop/libs/librdbms/src/librdbms/jdbc.py
After adding these lines the child process of py4j gets killed. Verified by the ps auxxx and pstree command.
This can be a good first issue to solve to raise pr . Nowadays work on Hadoop so not that band width to raise here . Just trying to contribute to oss. This flow is also same for presto.
@Harshg999 @bjornalm
Regards
Vinay Devadiga
Steps To Reproduce
As stated in descritption use trino with hue , create multiple hue clients and fire huge trino queries . In some time, the py4j servers will take the memory , hence queries get hangs.
Logs
Attached above.
Hue version
Open Source 4.10
The text was updated successfully, but these errors were encountered: