ci: Fix integration test timeouts with bounded nextest limit and job timeout

The integration CI jobs were hitting GitHub's 6-hour job limit, which
caused silent cancellation rather than a real test failure. Two issues
combined to cause this:

1. nextest's integration profile had terminate-after = 60, meaning a
   single hung test could run for 1200 × 60 = 72 000 s (20 h) before
   nextest force-killed it, far exceeding the 6-hour GHA limit.

2. The integration job had no explicit timeout-minutes, so a stalled
   job wasn't surfaced as a clear timeout failure.

Fix both: drop terminate-after to 1 (each test gets exactly 20 min before
nextest terminates it, which is generous for VM boot + test execution), and
add timeout-minutes: 90 to the integration job so any build or runner hang
fails cleanly rather than silently burning 6 hours.

Assisted-by: OpenCode (Claude Sonnet 4.6)
Signed-off-by: Colin Walters <walters@verbum.org>

Author: cgwalters

.config/nextest.toml

@@ -22,8 +22,12 @@ slow-timeout = { period = "300s", terminate-after = 2 }
 # Profile for integration tests — run with limited parallelism due to QEMU/KVM resources
 [profile.integration]
 test-threads = 2
-# 20 minutes for VM tests on slow CI runners
-slow-timeout = { period = "1200s", terminate-after = 60 }
+# 20 minutes per test; kill after 1 period to surface hangs quickly.
+# terminate-after = 1 means a test gets exactly 20 min before nextest force-kills it,
+# which is generous for VM boot + test execution on the slowest GHA runners.
+# With terminate-after = 60 (the old value) a single hung test could consume
+# 1200 × 60 = 72000 s (20 h), burning through the 6 h GHA job limit silently.
+slow-timeout = { period = "1200s", terminate-after = 1 }
 fail-fast = false
 failure-output = "immediate"
 success-output = "never"
@@ -44,3 +48,10 @@ store-failure-output = true
 [[profile.integration.overrides]]
 filter = 'test(/^privileged_/)'
 threads-required = 2
+
+# privileged_fast_suite consolidates multiple sub-tests into a single VM boot,
+# so it needs a longer timeout than the default 20-minute single-test budget.
+# 45 minutes covers one VM boot + 10 sequential sub-tests on the slowest TCG runner.
+[[profile.integration.overrides]]
+filter = 'test(/^privileged_fast_suite/)'
+slow-timeout = { period = "2700s", terminate-after = 1 }

.github/workflows/ci.yml

@@ -93,6 +93,11 @@ jobs:
     name: Integration tests (${{ matrix.name }})
     needs: smoke
     runs-on: ubuntu-24.04
+    # Generous ceiling: nextest kills individual tests at 20 min
+    # (terminate-after = 1 × period = 1200s in .config/nextest.toml), so
+    # this job-level limit only fires if the build step or the test runner
+    # itself hangs.  90 min is well above any realistic run time.
+    timeout-minutes: 90
     strategy:
       fail-fast: false
       matrix: