Prisma1: deleteManyX mutations error when deleting a large number of records

[goulderb]

Describe the bug
When running a "deleteManyX" mutation against a table with a large number of records, an error is returned with a message of

Whoops. Looks like an internal server error. Search your server logs for request ID: local:api:cjkcx8x7rvcxy0871suxurkfr

Looking at the prisma logs, I found the following:

[Bugsnag - local / testing] Error report: com.bugsnag.Report@54935f02
org.postgresql.util.PSQLException: An I/O error occurred while sending to the backend.
    at org.postgresql.core.v3.QueryExecutorImpl.execute(QueryExecutorImpl.java:333)
    at org.postgresql.jdbc.PgStatement.executeInternal(PgStatement.java:441)
    at org.postgresql.jdbc.PgStatement.execute(PgStatement.java:365)
    at org.postgresql.jdbc.PgPreparedStatement.executeWithFlags(PgPreparedStatement.java:155)
    at org.postgresql.jdbc.PgPreparedStatement.execute(PgPreparedStatement.java:144)
    at com.zaxxer.hikari.pool.ProxyPreparedStatement.execute(ProxyPreparedStatement.java:44)
    at com.zaxxer.hikari.pool.HikariProxyPreparedStatement.execute(HikariProxyPreparedStatement.java)
    at com.prisma.api.connector.jdbc.database.BuilderBase.$anonfun$deleteToDBIO$1(BuilderBase.scala:70)
    at com.prisma.api.connector.jdbc.database.BuilderBase.$anonfun$deleteToDBIO$1$adapted(BuilderBase.scala:70)
    at com.prisma.api.connector.jdbc.database.BuilderBase.$anonfun$jooqToDBIO$1(BuilderBase.scala:87)
    at slick.jdbc.SimpleJdbcAction.run(StreamingInvokerAction.scala:70)
    at slick.jdbc.SimpleJdbcAction.run(StreamingInvokerAction.scala:69)
    at slick.dbio.DBIOAction$$anon$4.$anonfun$run$3(DBIOAction.scala:239)
    at scala.collection.Iterator.foreach(Iterator.scala:929)
    at scala.collection.Iterator.foreach$(Iterator.scala:929)
    at scala.collection.AbstractIterator.foreach(Iterator.scala:1417)
    at scala.collection.IterableLike.foreach(IterableLike.scala:71)
    at scala.collection.IterableLike.foreach$(IterableLike.scala:70)
    at scala.collection.AbstractIterable.foreach(Iterable.scala:54)
    at slick.dbio.DBIOAction$$anon$4.run(DBIOAction.scala:239)
    at slick.dbio.DBIOAction$$anon$4.run(DBIOAction.scala:237)
    at slick.basic.BasicBackend$DatabaseDef$$anon$2.liftedTree1$1(BasicBackend.scala:275)
    at slick.basic.BasicBackend$DatabaseDef$$anon$2.run(BasicBackend.scala:275)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)
 Caused by: java.io.IOException: Tried to send an out-of-range integer as a 2-byte value: 44446
    at org.postgresql.core.PGStream.sendInteger2(PGStream.java:224)
    at org.postgresql.core.v3.QueryExecutorImpl.sendParse(QueryExecutorImpl.java:1440)
    at org.postgresql.core.v3.QueryExecutorImpl.sendOneQuery(QueryExecutorImpl.java:1762)
    at org.postgresql.core.v3.QueryExecutorImpl.sendQuery(QueryExecutorImpl.java:1326)
    at org.postgresql.core.v3.QueryExecutorImpl.execute(QueryExecutorImpl.java:298)
    ... 25 more

To Reproduce

1. Initialize & deploy a new prisma project
2. Create at least 32768 users
3. Issue a query of the form:
   mutation { deleteManyUsers { count } }
4. This will produce this error and not delete the records.

Expected behavior
I expect this to not produce errors and delete all the records.

Versions (please complete the following information):

• OS: OS X High Sierra
• prisma CLI: prisma/1.13.4 (darwin-x64) node-v10.8.0
• Prisma Server: 1.13.4

[marktani]

I can confirm this bug using the Postgres connector.

Additional reproduction setup:

type User {
  id: ID! @unique
  name: String!
}

prisma import -d nodes.zip

nodes.zip

---

I just confirmed that this is specific to the Postgres connector. I cannot reproduce this using the MySQL connector.

[do4gr]

This might be related to the maximum number of parameters the respective DBs accept in a query.

[goulderb]

To be clear, this worked on the previous version that my team was using (1.11.1). If there is anything I can do to help track this down further, please let me know.

[do4gr]

Thanks for the offer, but I'm pretty sure it's related to this. https://github.com/jOOQ/jOOQ/issues/5701 We started using jOOQ in 1.12. I just need to find out how to fix it since from the issue it sounds like it should already be handled by jOOQ.

[goulderb]

Gotcha, thanks for the information! The support from everyone on the prisma team on bug reports like this has really impressed me.

[do4gr]

We just released a fix for this in 1.13.6

[goulderb]

@do4gr Thanks for your quick action on this. You guys have really impressed me and my team with your response to bug reports.