diff --git a/pageserver/src/tenant/timeline.rs b/pageserver/src/tenant/timeline.rs
index 9838640097..65aad07257 100644
--- a/pageserver/src/tenant/timeline.rs
+++ b/pageserver/src/tenant/timeline.rs
@@ -3389,6 +3389,22 @@ impl Timeline {
                         // NOTE: this is racy, if there can be any other task that concurrently
                         // creates L1 layers. Currently, there can be only one compaction task
                         // running at any time, so this is fine.
+						//
+						// Also we hold `layer_removal_cs` guard which should prevent race condition
+						// even if there are two or more concurrent compaction tasks.
+						//
+						// But there is an opposite issue: we check presence of duplicates under
+						// `layers` shared lock, but then it is released. So there is a gap between
+						// this check and adding new layer to layer map. In principle in this gap some
+						// some other task (i.e. GC) can drop this layer and we already abandon insertion
+						// of duplicate layer. As a result there will be no such layer at all.
+						// In other words: we have some state S1 of pageserver where layer L1 can be removed by GC.
+						// Then we run compaction and it switch pageserver to the state S2 which writes duplicate of
+						// layer L1 and where it can not be removed. With this patch it is possible that
+						// we switch pageserver to state S2 but... with L1 lost.
+						// It is just hypothetical situation and there is no such concrete scenario which
+						// reproduces this problem. So let's take this risk.
+						//
                         if self.layers.read().unwrap().contains(
                             &(w.key_start()..end_key),
                             &w.lsn_range(),