Add ZFS replication metrics

README.rst

@@ -52,6 +52,7 @@ Usage
                         [--collector.cluster | --no-collector.cluster]
                         [--collector.resources | --no-collector.resources]
                         [--collector.config | --no-collector.config]
+                        [--collector.replication | --no-collector.replication]
                         [--config.file CONFIG_FILE]
                         [--web.listen-address WEB_LISTEN_ADDRESS]
                         [--server.keyfile SERVER_KEYFILE]
@@ -90,6 +91,8 @@ Usage
 
       --collector.config, --no-collector.config
                             Exposes PVE onboot status
+      --collector.replication, --no-collector.replication
+                            Exposes PVE replication info
 
 
 Use `[::]` in the `--web.listen-address` flag in order to bind to both IPv6 and
@@ -191,6 +194,24 @@ Here's an example of the metrics exported.
     # HELP pve_version_info Proxmox VE version info
     # TYPE pve_version_info gauge
     pve_version_info{release="7.1",repoid="6fe299a0",version="7.1-5"} 1.0
+    # HELP pve_replication_duration_seconds Proxmox vm replication duration
+    # TYPE pve_replication_duration_seconds gauge
+    pve_replication_duration_seconds{id="1-0"} 7.73584
+    # HELP pve_replication_last_sync_timestamp_seconds Proxmox vm replication last_sync
+    # TYPE pve_replication_last_sync_timestamp_seconds gauge
+    pve_replication_last_sync_timestamp_seconds{id="1-0"} 1.713382503e+09
+    # HELP pve_replication_last_try_timestamp_seconds Proxmox vm replication last_try
+    # TYPE pve_replication_last_try_timestamp_seconds gauge
+    pve_replication_last_try_timestamp_seconds{id="1-0"} 1.713382503e+09
+    # HELP pve_replication_next_sync_timestamp_seconds Proxmox vm replication next_sync
+    # TYPE pve_replication_next_sync_timestamp_seconds gauge
+    pve_replication_next_sync_timestamp_seconds{id="1-0"} 1.7134689e+09
+    # HELP pve_replication_failed_syncs Proxmox vm replication fail_count
+    # TYPE pve_replication_failed_syncs gauge
+    pve_replication_failed_syncs{id="1-0"} 0.0
+    # HELP pve_replication_info Proxmox vm replication info
+    # TYPE pve_replication_info gauge
+    pve_replication_info{guest="qemu/1",id="1-0",source="node/proxmox1",target="node/proxmox2",type="local"} 1.0
 
 Authentication
 --------------

src/pve_exporter/cli.py

@@ -46,6 +46,10 @@ def main():
                            action=BooleanOptionalAction, default=True,
                            help='Exposes PVE onboot status')
 
+    nodeflags.add_argument('--collector.replication', dest='collector_replication',
+                           action=BooleanOptionalAction, default=True,
+                           help='Exposes PVE replication info')
+
     parser.add_argument('--config.file', type=pathlib.Path,
                         dest="config_file", default='/etc/prometheus/pve.yml',
                         help='Path to config file (/etc/prometheus/pve.yml)')
@@ -69,7 +73,8 @@ def main():
         node=params.collector_node,
         cluster=params.collector_cluster,
         resources=params.collector_resources,
-        config=params.collector_config
+        config=params.collector_config,
+        replication=params.collector_replication
     )
 
     # Load configuration.

src/pve_exporter/collector/__init__.py

@@ -14,7 +14,10 @@
     VersionCollector,
     ClusterInfoCollector
 )
-from pve_exporter.collector.node import NodeConfigCollector
+from pve_exporter.collector.node import (
+    NodeConfigCollector,
+    NodeReplicationCollector
+)
 
 CollectorsOptions = collections.namedtuple('CollectorsOptions', [
     'status',
@@ -23,6 +26,7 @@
     'cluster',
     'resources',
     'config',
+    'replication'
 ])
 
 
@@ -44,5 +48,7 @@ def collect_pve(config, host, cluster, node, options: CollectorsOptions):
         registry.register(VersionCollector(pve))
     if node and options.config:
         registry.register(NodeConfigCollector(pve))
+    if node and options.replication:
+        registry.register(NodeReplicationCollector(pve))
 
     return generate_latest(registry)

src/pve_exporter/collector/node.py

@@ -4,6 +4,7 @@
 # pylint: disable=too-few-public-methods
 
 import logging
+import itertools
 
 from prometheus_client.core import GaugeMetricFamily
 
@@ -57,3 +58,71 @@ def collect(self):  # pylint: disable=missing-docstring
                     metrics[key].add_metric(label_values, metric_value)
 
         return metrics.values()
+
+class NodeReplicationCollector:
+    """
+    Collects Proxmox VE Replication information directly from status, i.e. replication duration,
+    last_sync, last_try, next_sync, fail_count.
+    For manual test: "pvesh get /nodes/<node>/replication/<id>/status"
+    """
+
+    def __init__(self, pve):
+        self._pve = pve
+
+    def collect(self): # pylint: disable=missing-docstring
+
+        info_metrics = {
+            'info': GaugeMetricFamily(
+            'pve_replication_info',
+            'Proxmox vm replication info',
+            labels=['id', 'type', 'source', 'target', 'guest'])
+        }
+
+        metrics = {
+            'duration': GaugeMetricFamily(
+                'pve_replication_duration_seconds',
+                'Proxmox vm replication duration',
+                labels=['id']),
+            'last_sync': GaugeMetricFamily(
+                'pve_replication_last_sync_timestamp_seconds',
+                'Proxmox vm replication last_sync',
+                labels=['id']),
+            'last_try': GaugeMetricFamily(
+                'pve_replication_last_try_timestamp_seconds',
+                'Proxmox vm replication last_try',
+                labels=['id']),
+            'next_sync': GaugeMetricFamily(
+                'pve_replication_next_sync_timestamp_seconds',
+                'Proxmox vm replication next_sync',
+                labels=['id']),
+            'fail_count': GaugeMetricFamily(
+                'pve_replication_failed_syncs',
+                'Proxmox vm replication fail_count',
+                labels=['id']),
+        }
+
+        node = None
+        for entry in self._pve.cluster.status.get():
+            if entry['type'] == 'node' and entry['local']:
+                node = entry['name']
+                break
+
+        for jobdata in self._pve.nodes(node).replication.get():
+            # Add info metric
+            label_values = [
+                str(jobdata['id']),
+                str(jobdata['type']),
+                f"node/{jobdata['source']}",
+                f"node/{jobdata['target']}",
+                f"{jobdata['vmtype']}/{jobdata['guest']}",
+            ]
+            info_metrics['info'].add_metric(label_values, 1)
+
+            # Add metrics
+            label_values = [str(jobdata['id'])]
+            status = self._pve.nodes(node).replication(jobdata['id']).status.get()
+            for key, metric_value in status.items():
+                if key in metrics:
+                    metrics[key].add_metric(label_values, metric_value)
+
+        return itertools.chain(metrics.values(), info_metrics.values())