[7.x][ML] Include peak memory usage in model snapshots (#1572) (#1575)

Ensure that the static counter tracking the peak memory usage for a job is included in the limited set to be persisted/restored from model state snapshots. Relates to elastic/elasticsearch#64154 Backports #1572
elastic · Nov 12, 2020 · 6102daf · 6102daf
1 parent ba4e091
commit 6102daf
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 0 deletions.
diff --git a/bin/autodetect/Main.cc b/bin/autodetect/Main.cc
@@ -63,6 +63,7 @@ int main(int argc, char** argv) {
         ml::counter_t::E_TSADNumberNewPeopleRecycled,
         ml::counter_t::E_TSADNumberApiRecordsHandled,
         ml::counter_t::E_TSADMemoryUsage,
+        ml::counter_t::E_TSADPeakMemoryUsage,
         ml::counter_t::E_TSADNumberMemoryUsageChecks,
         ml::counter_t::E_TSADNumberMemoryUsageEstimates,
         ml::counter_t::E_TSADNumberRecordsNoTimeField,

diff --git a/docs/CHANGELOG.asciidoc b/docs/CHANGELOG.asciidoc
@@ -43,6 +43,13 @@
 * Estimate upper bound of potential gains before splitting a decision tree node to avoid 
   unnecessary computation. (See {ml-pull}1537[#1537].)
 
+== {es} version 7.10.1
+
+=== Bug Fixes
+
+* Fix a bug where the peak_model_bytes value of the model_size_stats object  was not
+  restored from the anomaly detector job snapshots.
+
 == {es} version 7.10.0
 
 === Enhancements