diff --git a/casadm/cas_main.c b/casadm/cas_main.c
index 6d8f8b1..3fae855 100644
--- a/casadm/cas_main.c
+++ b/casadm/cas_main.c
@@ -740,7 +740,7 @@ static struct cas_param cas_cache_params[] = {
 #define CLEANING_POLICY_TYPE_DESC "Cleaning policy type. " \
 	"Available policy types: {nop|alru|acp}"
 
-#define CLEANING_ALRU_WAKE_UP_DESC "Period of time between awakenings of flushing thread <%d-%d>[s] (default: %d s)"
+#define CLEANING_ALRU_WAKE_UP_DESC "Cleaning thread sleep time after an idle wake up <%d-%d>[s] (default: %d s)"
 #define CLEANING_ALRU_STALENESS_TIME_DESC "Time that has to pass from the last write operation before a dirty cache" \
 	 " block can be scheduled to be flushed <%d-%d>[s] (default: %d s)"
 #define CLEANING_ALRU_FLUSH_MAX_BUFFERS_DESC "Number of dirty cache blocks to be flushed in one cleaning cycle" \
diff --git a/casadm/casadm.8 b/casadm/casadm.8
index 4edffd2..8575906 100644
--- a/casadm/casadm.8
+++ b/casadm/casadm.8
@@ -286,7 +286,9 @@ Identifier of cache instance <1-16384>.
 
 .TP
 .B -w, --wake-up <NUMBER>
-Period of time between awakenings of flushing thread [s] (default: 20 s).
+Cleaning thread sleep time after an idle wake up [s] (default: 20 s).
+Idle wake up happens when there is no dirty data or the cleaning thread does not start cleaning
+due to staleness time and/or activity threshold constraints.
 
 .TP
 .B -s, --staleness-time <NUMBER>
diff --git a/test/functional/tests/lazy_writes/cleaning_policy/test_cleaning_params.py b/test/functional/tests/lazy_writes/cleaning_policy/test_cleaning_params.py
index 8ae4f32..9a4e91e 100644
--- a/test/functional/tests/lazy_writes/cleaning_policy/test_cleaning_params.py
+++ b/test/functional/tests/lazy_writes/cleaning_policy/test_cleaning_params.py
@@ -141,10 +141,13 @@ def test_cleaning_policy_config():
                 f"Dirty data before pause: {core_dirty_before}\n"
                 f"Dirty data after pause: {core_dirty_after}"
             )
-        elif core_dirty_before != core_dirty_after + data_to_flush:
+        # Only check whether a minimum amount of data is flushed, as ALRU does not have
+        # a configurable sleep time between active flushing iterations which would allow
+        # to precisely estimate expected amount of data.
+        elif core_dirty_before < core_dirty_after + data_to_flush:
             TestRun.LOGGER.error(
                 f"Number of dirty blocks flushed differs from configured in policy.\n"
-                f"Expected dirty data flushed: {data_to_flush}\n"
+                f"Expected minimum dirty data flushed: {data_to_flush}\n"
                 f"Actual dirty data flushed: "
                 f"{core_dirty_before - core_dirty_after}"
             )