[
https://issues.apache.org/jira/browse/BEAM-3714?focusedWorklogId=92880&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-92880
]
ASF GitHub Bot logged work on BEAM-3714:
----------------------------------------
Author: ASF GitHub Bot
Created on: 19/Apr/18 22:21
Start Date: 19/Apr/18 22:21
Worklog Time Spent: 10m
Work Description: jkff commented on issue #5109: [BEAM-3714]modified
result set to be forward only and read only
URL: https://github.com/apache/beam/pull/5109#issuecomment-382898723
The Go precommit is a flake, but in Java there seems to be a couple of final
simple checkstyle errors (line too long) remaining.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 92880)
Time Spent: 3h 20m (was: 3h 10m)
> JdbcIO.read() should create a forward-only, read-only result set
> ----------------------------------------------------------------
>
> Key: BEAM-3714
> URL: https://issues.apache.org/jira/browse/BEAM-3714
> Project: Beam
> Issue Type: Bug
> Components: io-java-jdbc
> Reporter: Eugene Kirpichov
> Assignee: Innocent
> Priority: Major
> Time Spent: 3h 20m
> Remaining Estimate: 0h
>
> [https://stackoverflow.com/questions/48784889/streaming-data-from-cloudsql-into-dataflow/48819934#48819934]
> - a user is trying to load a large table from MySQL, and the MySQL JDBC
> driver requires special measures when loading large result sets.
> JdbcIO currently calls simply "connection.prepareStatement(query)"
> https://github.com/apache/beam/blob/bb8c12c4956cbe3c6f2e57113e7c0ce2a5c05009/sdks/java/io/jdbc/src/main/java/org/apache/beam/sdk/io/jdbc/JdbcIO.java#L508
> - it should specify type TYPE_FORWARD_ONLY and concurrency CONCUR_READ_ONLY
> - these values should always be used.
> Seems that different databases have different requirements for streaming
> result sets.
> E.g. MySQL requires setting fetch size; PostgreSQL says "The Connection must
> not be in autocommit mode."
> https://jdbc.postgresql.org/documentation/head/query.html#query-with-cursor .
> Oracle, I think, doesn't have any special requirements but I don't know.
> Fetch size should probably still be set to a reasonably large value.
> Seems that the common denominator of these requirements is: set fetch size to
> a reasonably large but not maximum value; disable autocommit (there's nothing
> to commit in read() anyway).
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
