From cfe6b3d6182f2e865567f0f89c4e47e94303386f Mon Sep 17 00:00:00 2001
From: Sebastian Baunsgaard <baunsgaard@apache.org>
Date: Tue, 23 Jun 2026 23:10:37 +0000
Subject: [PATCH 1/2] Raise federated request timeout for multitenant Spark
 tests

Set sysds.federated.timeout in the multi-tenant test config from 16s to
60s. The 16s bound was too aggressive for the Spark-backed (SP) variants
of the federated multitenant reuse tests: Spark context creation alone
takes ~14s, so under shared CI load a single federated request (both the
rightIndex/rblk instruction execution and the end-of-run stats
collection) routinely exceeded 16s and threw TimeoutException.

These spurious timeouts caused FederatedReuseReadTest.testModifiedValLineageSP
and FederatedSerializationReuseTest.testRowSumsSP to fail intermittently;
surefire then reran each failing test, and the accumulated rerun time
repeatedly pushed the monitoring/multitenant CI job past its 30-minute
cap, cancelling the whole shard.

60s keeps a hard bound on a genuinely runaway request (so the suite still
cannot hang silently) while giving the Spark variants enough headroom to
complete on the first attempt, which also removes the expensive reruns.
---
 src/test/config/SystemDS-MultiTenant-config.xml | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/src/test/config/SystemDS-MultiTenant-config.xml b/src/test/config/SystemDS-MultiTenant-config.xml
index 321fcc0b282..871aa861b53 100644
--- a/src/test/config/SystemDS-MultiTenant-config.xml
+++ b/src/test/config/SystemDS-MultiTenant-config.xml
@@ -20,7 +20,10 @@
 <root>
    <!-- The timeout of the federated tests to initialize the federated matrixes -->
    <sysds.federated.initialization.timeout>30</sysds.federated.initialization.timeout>
-   <!-- The timeout of each instruction sent to federated workers -->
-   <sysds.federated.timeout>16</sysds.federated.timeout>
+   <!-- The timeout of each instruction sent to federated workers. This bounds a single
+        runaway request so the suite cannot hang silently, while leaving enough headroom
+        for the Spark-backed variants (Spark context creation alone is ~14s) to complete
+        under shared CI load without spuriously timing out. -->
+   <sysds.federated.timeout>60</sysds.federated.timeout>
    <sysds.local.spark>true</sysds.local.spark>
 </root>

From f02420a53176cf5b388bfa7b29821ac0b8e5878b Mon Sep 17 00:00:00 2001
From: Sebastian Baunsgaard <baunsgaard@apache.org>
Date: Tue, 23 Jun 2026 23:34:46 +0000
Subject: [PATCH 2/2] Shorten federated timeout comment in multitenant test
 config

Replace the verbose explanatory comment with the original one-line
description; the rationale lives in the commit history.
---
 src/test/config/SystemDS-MultiTenant-config.xml | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/src/test/config/SystemDS-MultiTenant-config.xml b/src/test/config/SystemDS-MultiTenant-config.xml
index 871aa861b53..ae915177617 100644
--- a/src/test/config/SystemDS-MultiTenant-config.xml
+++ b/src/test/config/SystemDS-MultiTenant-config.xml
@@ -20,10 +20,7 @@
 <root>
    <!-- The timeout of the federated tests to initialize the federated matrixes -->
    <sysds.federated.initialization.timeout>30</sysds.federated.initialization.timeout>
-   <!-- The timeout of each instruction sent to federated workers. This bounds a single
-        runaway request so the suite cannot hang silently, while leaving enough headroom
-        for the Spark-backed variants (Spark context creation alone is ~14s) to complete
-        under shared CI load without spuriously timing out. -->
+   <!-- The timeout of each instruction sent to federated workers -->
    <sysds.federated.timeout>60</sysds.federated.timeout>
    <sysds.local.spark>true</sysds.local.spark>
 </root>