diff --git a/docs/scheduler/observability.md b/docs/scheduler/observability.md
index 6172a469..d56a5e44 100644
--- a/docs/scheduler/observability.md
+++ b/docs/scheduler/observability.md
@@ -256,6 +256,24 @@ store as `Unspecified`. The **Log rows with no severity** panel on the
 had a priority and arrived without a severity anyway signal a broken mapping,
 while lines that never had one signal nothing of the sort.
 
+### Why Explore's level buttons need the datasource told
+
+The stored field is `severity_text`, and Grafana's log-level buttons filter on
+a field called `level` — a name no row here carries. That is not something the
+pipeline can fix: VictoriaLogs' OTLP ingester chooses the stored name itself,
+and its ingest parameters have no option to rename it.
+
+So the datasource is told instead. `nix/host-modules/swarm-grafana.nix`
+provisions the VictoriaLogs datasource with a `logLevelRules` entry per
+severity in the table above, each matching `severity_text` against one of
+those values — which is the plugin's way of saying "the level lives in this
+field". A level button then filters on `level` **or** the matching
+`severity_text`, and the store answers the second half.
+
+`Unspecified` is deliberately left out of those rules, so rows with no
+severity stay unfiltered by every level button and keep showing up in the
+panel that counts them.
+
 ## Host-emitted container-resource metrics (hive-c0re)
 
 <!-- vale write-good.Passive = NO -->
diff --git a/nix/host-modules/swarm-grafana.nix b/nix/host-modules/swarm-grafana.nix
index f6e9d209..aaa301bd 100644
--- a/nix/host-modules/swarm-grafana.nix
+++ b/nix/host-modules/swarm-grafana.nix
@@ -40,6 +40,65 @@ let
   # free to drift into a panel that renders empty rather than erroring.
   logsDatasourceUid = "swarm-victorialogs";
 
+  # Which field carries the log level, told to the READER because the writer
+  # cannot be told. Grafana's log-level buttons filter on a field called
+  # `level`, and no row in this store has one: VictoriaLogs' OTLP ingester
+  # names the stored field `severity_text` unconditionally (v1.52.0,
+  # `app/vlinsert/opentelemetry/pb.go`) and its ingest parameters have no
+  # `_level_field` sibling to rename it with. So the mapping is made here.
+  #
+  # `logLevelRules` is the datasource plugin's own jsonData key and the only
+  # level-related one it has — there is no "the level lives in field X"
+  # string to set. A rule names its own field, so saying it takes one rule
+  # per level.
+  #
+  # 🔑 Each enabled rule appends an `OR <field>:<op>"<value>"` term to the
+  # query a level button emits, beside the `level:…` term that matches
+  # nothing here. Clicking "info" goes from
+  #
+  #   level:contains_common_case("info","information","informational","notice")
+  #
+  # to that OR `severity_text:="INFO"`, which our rows do match.
+  #
+  # ⚠️ Three ways to get a rule wrong, all of them SILENT — Grafana accepts
+  # any jsonData it does not recognise, so a bad rule provisions cleanly and
+  # the buttons go on returning zero:
+  #
+  #  - `enabled` must be literally `true`, not merely not-false. The query
+  #    builder keeps rules on `rule.enabled` being truthy while the row
+  #    colouring path keeps them on `!== false`, so an omitted flag colours
+  #    rows correctly and leaves the buttons broken — working in the half
+  #    nobody is looking at.
+  #  - `level` must be a CANONICAL Grafana `LogLevel` value. The builder
+  #    groups rules by it and only ever looks up `critical`, `error`,
+  #    `warning`, `info`, `debug`, `trace`; `warn` and `fatal` are enum
+  #    aliases that resolve to other spellings and match no group.
+  #  - `value` is compared with `===`, so it is the severity text exactly as
+  #    STORED: the uppercase OpenTelemetry short names `overwrite_text`
+  #    writes in ../journald-severity.nix, not that table's lowercase keys.
+  #
+  # `Unspecified` is deliberately unmapped. It is VictoriaLogs' own rendering
+  # of an absent severity, and it is what the logs dashboard's "no severity"
+  # panel counts — giving it a level would dress the missing data up as a
+  # colour and retire the instrument that measures it.
+  logLevelRules =
+    let
+      fromSeverityText = value: level: {
+        field = "severity_text";
+        operator = "equals";
+        enabled = true;
+        inherit value level;
+      };
+    in
+    [
+      (fromSeverityText "FATAL" "critical")
+      (fromSeverityText "ERROR" "error")
+      (fromSeverityText "WARN" "warning")
+      (fromSeverityText "INFO" "info")
+      (fromSeverityText "DEBUG" "debug")
+      (fromSeverityText "TRACE" "trace")
+    ];
+
   # The shipped dashboards carry `@datasourceUid@` / `@logsDatasourceUid@` where
   # a real deployment needs the uids above. They are substituted here rather
   # than committed with the literals so the single binding stays single.
@@ -944,6 +1003,11 @@ in
                   # two defaults is a coin toss over which one an untyped panel
                   # gets.
                   isDefault = false;
+                  # See `logLevelRules` above for why the level lives here and
+                  # not in the pipeline.
+                  jsonData = {
+                    inherit logLevelRules;
+                  };
                 }
               ];
             };
diff --git a/nix/module-eval/grafana.nix b/nix/module-eval/grafana.nix
index 9ade378d..c1008194 100644
--- a/nix/module-eval/grafana.nix
+++ b/nix/module-eval/grafana.nix
@@ -243,6 +243,43 @@ let
         lib.any (e: lib.hasInfix "PRIORITY:*" e) counts
         && lib.any (e: lib.hasInfix "PRIORITY:\"\"" e) counts;
     }
+    {
+      # The reader's half of the same mapping. The collector writes a severity
+      # TEXT; nothing downstream reads it as a level unless the datasource is
+      # told which field holds it, and that telling is a list of rules rather
+      # than a field name — so it goes stale one severity at a time. Asserted
+      # against ../journald-severity.nix rather than a literal list here,
+      # because the way this regresses is a severity added to the parser and
+      # not to Grafana: every line still arrives, the new one is just
+      # unfilterable, and no query errors to say so.
+      #
+      # `enabled` and the canonical `level` spelling are pinned alongside
+      # because both fail silently in the direction of "provisions fine,
+      # returns nothing" — see the comment on `logLevelRules` in
+      # ../host-modules/swarm-grafana.nix.
+      name = "the logs datasource maps every severity the collector can emit to a log level";
+      ok =
+        let
+          sources =
+            grafanaOldPath.containers.swarm-grafana.config.services.grafana.provision.datasources.settings.datasources;
+          logs = lib.head (lib.filter (d: d.uid == "swarm-victorialogs") sources);
+          rules = logs.jsonData.logLevelRules or [ ];
+          # `overwrite_text` makes the stored text the OpenTelemetry short name
+          # for each mapped severity, which is the parser's own key uppercased.
+          emitted = map lib.toUpper (lib.attrNames (lib.head (import ../journald-severity.nix)).mapping);
+          canonical = [
+            "critical"
+            "error"
+            "warning"
+            "info"
+            "debug"
+            "trace"
+          ];
+        in
+        rules != [ ]
+        && lib.all (r: r.field == "severity_text" && r.enabled == true && lib.elem r.level canonical) rules
+        && lib.all (text: lib.any (r: r.value == text) rules) emitted;
+    }
   ];
 in
 runGroup "grafana" cases