From db4b62b4622b44f4bedc757bdf401dff6aa0f880 Mon Sep 17 00:00:00 2001
From: Nitzan Mordechai <nmordech@redhat.com>
Date: Wed, 7 May 2025 13:37:34 +0000
Subject: [PATCH] test/librados/aio_cxx: skip EIO boundary assertion when no
 in-flight I/Os
MIME-Version: 1.0
Content-Type: text/plain; charset=utf8
Content-Transfer-Encoding: 8bit

In the PoolEIOFlag test we now track whether any async writes were still
in flight at the moment we flipped the pool’s `eio=true` flag using an
atomic `missed_eio` flag set inside the helper thread.  If we never
actually overlapped any in-flight AIOs (we completed all aio successfully)
with the flag flip, we `GTEST_SKIP()` the final `max_success+1 == min_failed`
boundary assertion, avoiding spurious failures when all writes complete
before EIO takes effect.

Fixes: https://tracker.ceph.com/issues/70852
Signed-off-by: Nitzan Mordechai <nmordech@redhat.com>
(cherry picked from commit 07319d9a37ef70b6422e26eba8a251f711db2c6b)
---
 src/test/librados/aio_cxx.cc | 41 +++++++++++++++++++++++-------------
 1 file changed, 26 insertions(+), 15 deletions(-)

diff --git a/src/test/librados/aio_cxx.cc b/src/test/librados/aio_cxx.cc
index 5e35869b5c2f3..b0b7b6c4e7d50 100644
--- a/src/test/librados/aio_cxx.cc
+++ b/src/test/librados/aio_cxx.cc
@@ -1,6 +1,8 @@
+#include <common/dout.h>
 #include <errno.h>
 #include <fcntl.h>
 #include <deque>
+#include <iostream>
 #include <sstream>
 #include <string>
 #include <utility>
@@ -2347,7 +2349,7 @@ void pool_io_callback(completion_t cb, void *arg /* Actually AioCompletion* */)
     ASSERT_EQ(0, info->c->wait_for_complete());
   }
   int r = info->c->get_return_value();
-  //cout << "finish " << i << " r = " << r << std::endl;
+  cout << "finish " << i << " r = " << r << std::endl;
 
   std::scoped_lock l(my_lock);
   inflight.erase(i);
@@ -2369,8 +2371,9 @@ TEST(LibRadosAio, PoolEIOFlag) {
   bufferlist bl;
   bl.append("some data");
   std::thread *t = nullptr;
+  std::atomic<bool> missed_eio{false};
   
-  unsigned max = 100;
+  unsigned max = 1000;
   unsigned timeout = max * 10;
   unsigned long i = 1;
   my_lock.lock();
@@ -2385,19 +2388,23 @@ TEST(LibRadosAio, PoolEIOFlag) {
     //cout << "start " << i << " r = " << r << std::endl;
 
     if (i == max / 2) {
-      cout << "setting pool EIO" << std::endl;
-      t = new std::thread(
-	[&] {
-	  bufferlist empty;
-	  ASSERT_EQ(0, test_data.m_cluster.mon_command(
-	    fmt::format(R"({{
-	                "prefix": "osd pool set",
-	                "pool": "{}",
-	                "var": "eio",
-	                "val": "true"
-	                }})", test_data.m_pool_name),
-	    empty, nullptr, nullptr));
-	});
+      t = new std::thread([&] {
+        bufferlist empty;
+        cout << "sending pool EIO time: " << ceph_clock_now() << std::endl;
+        ASSERT_EQ(0, test_data.m_cluster.mon_command(
+          fmt::format(R"({{
+            "prefix": "osd pool set",
+            "pool": "{}",
+            "var": "eio",
+            "val": "true"
+            }})", test_data.m_pool_name),
+       empty, nullptr, nullptr));
+
+        {
+          std::scoped_lock lk(my_lock);
+          missed_eio = (!min_failed && max_success == max);
+        }
+      });
     }
 
     std::this_thread::sleep_for(10ms);
@@ -2418,6 +2425,10 @@ TEST(LibRadosAio, PoolEIOFlag) {
     my_lock.lock();
   }
 
+  if (!missed_eio) {
+    GTEST_SKIP() << "eio flag missed all ios that already completed";
+    my_lock.unlock();
+  }
   cout << "max_success " << max_success << ", min_failed " << min_failed << std::endl;
   ASSERT_TRUE(max_success + 1 == min_failed);
   my_lock.unlock();
-- 
2.39.5