Bump version to 0.9.87

CHANGELOG.md

@@ -6,7 +6,6 @@
   * [Node Daemon] Adds network traffic rate calculation subsystem.
   * [All Daemons] Fixes a printing bug where newlines were not added atomically.
   * [CLI Client] Fixes a bug listing connections if no default is specified.
-  * [All Daemons] Simplifies debug logging conditionals by moving into the Logger instance itself.
 
 ###### [v0.9.86](https://github.com/parallelvirtualcluster/pvc/releases/tag/v0.9.86)
 

daemon-common/log.py

@@ -115,10 +115,6 @@ class Logger(object):
 
     # Output function
     def out(self, message, state=None, prefix=""):
-        # Only handle d-state (debug) messages if we're in debug mode
-        if state in ["d"] and not self.config["debug"]:
-            return
-
         # Get the date
         if self.config["log_dates"]:
             date = "{} ".format(datetime.now().strftime("%Y/%m/%d %H:%M:%S.%f"))

debian/changelog

@@ -4,9 +4,8 @@ pvc (0.9.87-0) unstable; urgency=high
   * [Node Daemon] Adds network traffic rate calculation subsystem.
   * [All Daemons] Fixes a printing bug where newlines were not added atomically.
   * [CLI Client] Fixes a bug listing connections if no default is specified.
-  * [All Daemons] Simplifies debug logging conditionals by moving into the Logger instance itself.
 
- -- Joshua M. Boniface <joshua@boniface.me>  Wed, 27 Dec 2023 13:03:14 -0500
+ -- Joshua M. Boniface <joshua@boniface.me>  Wed, 27 Dec 2023 12:42:52 -0500
 
 pvc (0.9.86-0) unstable; urgency=high
 

health-daemon/pvchealthd/objects/MonitoringInstance.py

@@ -157,6 +157,9 @@ class MonitoringPlugin(object):
             "w": warning
             "e": error
         """
+        if state == "d" and not self.config["debug"]:
+            return
+
         self.logger.out(message, state=state, prefix=self.plugin_name)
 
     #
@@ -520,6 +523,7 @@ class MonitoringInstance(object):
 
             entries = fault_data["entries"]()
 
+            if self.config["debug"]:
                 self.logger.out(
                     f"Entries for fault check {fault_type}: {dumps(entries)}",
                     state="d",

node-daemon/pvcnoded/objects/DNSAggregatorInstance.py

@@ -333,6 +333,7 @@ class AXFRDaemonInstance(object):
                     z = dns.zone.from_xfr(axfr)
                     records_raw = [z[n].to_text(n) for n in z.nodes.keys()]
                 except Exception as e:
+                    if self.config["debug"]:
                         self.logger.out(
                             "{} {} ({})".format(e, dnsmasq_ip, domain),
                             state="d",
@@ -369,6 +370,7 @@ class AXFRDaemonInstance(object):
                         "SELECT * FROM records WHERE domain_id=%s", (domain_id,)
                     )
                     results = list(sql_curs.fetchall())
+                    if self.config["debug"]:
                         self.logger.out(
                             "SQL query results: {}".format(results),
                             state="d",
@@ -386,6 +388,7 @@ class AXFRDaemonInstance(object):
                 records_old = list()
                 records_old_ids = list()
                 if not results:
+                    if self.config["debug"]:
                         self.logger.out(
                             "No results found, skipping.",
                             state="d",
@@ -401,6 +404,7 @@ class AXFRDaemonInstance(object):
                     r_data = record[4]
                     # Assemble a list element in the same format as the AXFR data
                     entry = "{} {} IN {} {}".format(r_name, r_ttl, r_type, r_data)
+                    if self.config["debug"]:
                         self.logger.out(
                             "Found record: {}".format(entry),
                             state="d",
@@ -409,6 +413,7 @@ class AXFRDaemonInstance(object):
 
                     # Skip non-A or AAAA records
                     if r_type != "A" and r_type != "AAAA":
+                        if self.config["debug"]:
                             self.logger.out(
                                 'Skipping record {}, not A or AAAA: "{}"'.format(
                                     entry, r_type
@@ -424,6 +429,7 @@ class AXFRDaemonInstance(object):
                 records_new.sort()
                 records_old.sort()
 
+                if self.config["debug"]:
                     self.logger.out(
                         "New: {}".format(records_new),
                         state="d",
@@ -444,6 +450,7 @@ class AXFRDaemonInstance(object):
                     in_new_not_in_old = in_new - in_old
                     in_old_not_in_new = in_old - in_new
 
+                    if self.config["debug"]:
                         self.logger.out(
                             "New but not old: {}".format(in_new_not_in_old),
                             state="d",
@@ -480,6 +487,7 @@ class AXFRDaemonInstance(object):
                     if len(remove_records) > 0:
                         # Remove the invalid old records
                         for record_id in remove_records:
+                            if self.config["debug"]:
                                 self.logger.out(
                                     "Removing record: {}".format(record_id),
                                     state="d",
@@ -499,6 +507,7 @@ class AXFRDaemonInstance(object):
                             r_ttl = record[1]
                             r_type = record[3]
                             r_data = record[4]
+                            if self.config["debug"]:
                                 self.logger.out(
                                     "Add record: {}".format(name),
                                     state="d",
@@ -511,6 +520,7 @@ class AXFRDaemonInstance(object):
                                 )
                                 changed = True
                             except psycopg2.IntegrityError as e:
+                                if self.config["debug"]:
                                     self.logger.out(
                                         "Failed to add record due to {}: {}".format(
                                             e, name
@@ -519,6 +529,7 @@ class AXFRDaemonInstance(object):
                                         prefix="dns-aggregator",
                                     )
                             except psycopg2.errors.InFailedSqlTransaction as e:
+                                if self.config["debug"]:
                                     self.logger.out(
                                         "Failed to add record due to {}: {}".format(
                                             e, name
@@ -537,6 +548,7 @@ class AXFRDaemonInstance(object):
                         current_serial = int(soa_record[2])
                         new_serial = current_serial + 1
                         soa_record[2] = str(new_serial)
+                        if self.config["debug"]:
                             self.logger.out(
                                 "Records changed; bumping SOA: {}".format(new_serial),
                                 state="d",
@@ -548,6 +560,7 @@ class AXFRDaemonInstance(object):
                         )
 
                         # Commit all the previous changes
+                        if self.config["debug"]:
                             self.logger.out(
                                 "Committing database changes and reloading PDNS",
                                 state="d",

node-daemon/pvcnoded/util/keepalive.py

@@ -80,6 +80,8 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
     pool_list = zkhandler.children("base.pool")
     osd_list = zkhandler.children("base.osd")
 
+    debug = config["debug"]
+    if debug:
         logger.out("Thread starting", state="d", prefix="ceph-thread")
 
     # Connect to the Ceph cluster
@@ -88,6 +90,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             conffile=config["ceph_config_file"],
             conf=dict(keyring=config["ceph_admin_keyring"]),
         )
+        if debug:
             logger.out("Connecting to cluster", state="d", prefix="ceph-thread")
         ceph_conn.connect(timeout=1)
     except Exception as e:
@@ -97,6 +100,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
     # Primary-only functions
     if this_node.coordinator_state == "primary":
         # Get Ceph status information (pretty)
+        if debug:
             logger.out(
                 "Set Ceph status information in zookeeper (primary only)",
                 state="d",
@@ -113,6 +117,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             logger.out("Failed to set Ceph status data: {}".format(e), state="e")
 
         # Get Ceph health information (JSON)
+        if debug:
             logger.out(
                 "Set Ceph health information in zookeeper (primary only)",
                 state="d",
@@ -129,6 +134,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             logger.out("Failed to set Ceph health data: {}".format(e), state="e")
 
         # Get Ceph df information (pretty)
+        if debug:
             logger.out(
                 "Set Ceph rados df information in zookeeper (primary only)",
                 state="d",
@@ -145,6 +151,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
         except Exception as e:
             logger.out("Failed to set Ceph utilization data: {}".format(e), state="e")
 
+        if debug:
             logger.out(
                 "Set pool information in zookeeper (primary only)",
                 state="d",
@@ -172,6 +179,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             rados_pool_df_raw = []
 
         pool_count = len(ceph_pool_df_raw)
+        if debug:
             logger.out(
                 "Getting info for {} pools".format(pool_count),
                 state="d",
@@ -187,13 +195,17 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
 
                 # Ignore any pools that aren't in our pool list
                 if pool["name"] not in pool_list:
+                    if debug:
                         logger.out(
-                        "Pool {} not in pool list {}".format(pool["name"], pool_list),
+                            "Pool {} not in pool list {}".format(
+                                pool["name"], pool_list
+                            ),
                             state="d",
                             prefix="ceph-thread",
                         )
                     continue
                 else:
+                    if debug:
                         logger.out(
                             "Parsing data for pool {}".format(pool["name"]),
                             state="d",
@@ -236,6 +248,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
     osds_this_node = 0
     if len(osd_list) > 0:
         # Get data from Ceph OSDs
+        if debug:
             logger.out("Get data from Ceph OSDs", state="d", prefix="ceph-thread")
 
         # Parse the dump data
@@ -251,6 +264,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             logger.out("Failed to obtain OSD data: {}".format(e), state="w")
             osd_dump_raw = []
 
+        if debug:
             logger.out("Loop through OSD dump", state="d", prefix="ceph-thread")
         for osd in osd_dump_raw:
             osd_dump.update(
@@ -265,6 +279,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             )
 
         # Parse the df data
+        if debug:
             logger.out("Parse the OSD df data", state="d", prefix="ceph-thread")
 
         osd_df = dict()
@@ -278,6 +293,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             logger.out("Failed to obtain OSD data: {}".format(e), state="w")
             osd_df_raw = []
 
+        if debug:
             logger.out("Loop through OSD df", state="d", prefix="ceph-thread")
         for osd in osd_df_raw:
             osd_df.update(
@@ -300,6 +316,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             )
 
         # Parse the status data
+        if debug:
             logger.out("Parse the OSD status data", state="d", prefix="ceph-thread")
 
         osd_status = dict()
@@ -313,6 +330,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             logger.out("Failed to obtain OSD status data: {}".format(e), state="w")
             osd_status_raw = []
 
+        if debug:
             logger.out("Loop through OSD status data", state="d", prefix="ceph-thread")
 
         for line in osd_status_raw.split("\n"):
@@ -382,6 +400,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
             )
 
         # Merge them together into a single meaningful dict
+        if debug:
             logger.out("Merge OSD data together", state="d", prefix="ceph-thread")
 
         osd_stats = dict()
@@ -402,7 +421,10 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
 
         # Upload OSD data for the cluster (primary-only)
         if this_node.coordinator_state == "primary":
-            logger.out("Trigger updates for each OSD", state="d", prefix="ceph-thread")
+            if debug:
+                logger.out(
+                    "Trigger updates for each OSD", state="d", prefix="ceph-thread"
+                )
 
             for osd in osd_list:
                 try:
@@ -419,15 +441,19 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
 
     queue.put(osds_this_node)
 
+    if debug:
         logger.out("Thread finished", state="d", prefix="ceph-thread")
 
 
 # VM stats update function
 def collect_vm_stats(logger, config, zkhandler, this_node, queue):
+    debug = config["debug"]
+    if debug:
         logger.out("Thread starting", state="d", prefix="vm-thread")
 
     # Connect to libvirt
     libvirt_name = "qemu:///system"
+    if debug:
         logger.out("Connecting to libvirt", state="d", prefix="vm-thread")
     try:
         lv_conn = libvirt.open(libvirt_name)
@@ -441,6 +467,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
     memprov = 0
     vcpualloc = 0
     # Toggle state management of dead VMs to restart them
+    if debug:
         logger.out(
             "Toggle state management of dead VMs to restart them",
             state="d",
@@ -491,6 +518,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
             domain_name = domain.name()
 
             # Get all the raw information about the VM
+            if debug:
                 logger.out(
                     "Getting general statistics for VM {}".format(domain_name),
                     state="d",
@@ -509,9 +537,12 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
             domain_memory_stats = domain.memoryStats()
             domain_cpu_stats = domain.getCPUStats(True)[0]
         except Exception as e:
+            if debug:
                 try:
                     logger.out(
-                    "Failed getting VM information for {}: {}".format(domain.name(), e),
+                        "Failed getting VM information for {}: {}".format(
+                            domain.name(), e
+                        ),
                         state="d",
                         prefix="vm-thread",
                     )
@@ -523,6 +554,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
         if domain_uuid not in this_node.domain_list:
             this_node.domain_list.append(domain_uuid)
 
+        if debug:
             logger.out(
                 "Getting disk statistics for VM {}".format(domain_name),
                 state="d",
@@ -546,6 +578,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
                     }
                 )
         except Exception as e:
+            if debug:
                 try:
                     logger.out(
                         "Failed getting disk stats for {}: {}".format(domain.name(), e),
@@ -556,6 +589,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
                     pass
             continue
 
+        if debug:
             logger.out(
                 "Getting network statistics for VM {}".format(domain_name),
                 state="d",
@@ -585,9 +619,12 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
                     }
                 )
         except Exception as e:
+            if debug:
                 try:
                     logger.out(
-                    "Failed getting network stats for {}: {}".format(domain.name(), e),
+                        "Failed getting network stats for {}: {}".format(
+                            domain.name(), e
+                        ),
                         state="d",
                         prefix="vm-thread",
                     )
@@ -608,6 +645,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
             "net_stats": domain_network_stats,
         }
 
+        if debug:
             logger.out(
                 "Writing statistics for VM {} to Zookeeper".format(domain_name),
                 state="d",
@@ -619,6 +657,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
                 [(("domain.stats", domain_uuid), str(json.dumps(domain_stats)))]
             )
         except Exception as e:
+            if debug:
                 logger.out(
                     "Failed to write domain statistics: {}".format(e),
                     state="d",
@@ -628,6 +667,7 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
     # Close the Libvirt connection
     lv_conn.close()
 
+    if debug:
         logger.out(
             f"VM stats: doms: {len(running_domains)}; memalloc: {memalloc}; memprov: {memprov}; vcpualloc: {vcpualloc}",
             state="d",
@@ -639,11 +679,14 @@ def collect_vm_stats(logger, config, zkhandler, this_node, queue):
     queue.put(memprov)
     queue.put(vcpualloc)
 
+    if debug:
         logger.out("Thread finished", state="d", prefix="vm-thread")
 
 
 # Keepalive update function
 def node_keepalive(logger, config, zkhandler, this_node, netstats):
+    debug = config["debug"]
+
     # Display node information to the terminal
     if config["log_keepalives"]:
         if this_node.coordinator_state == "primary":
@@ -703,7 +746,10 @@ def node_keepalive(logger, config, zkhandler, this_node, netstats):
                 )
 
     # Get past state and update if needed
-    logger.out("Get past state and update if needed", state="d", prefix="main-thread")
+    if debug:
+        logger.out(
+            "Get past state and update if needed", state="d", prefix="main-thread"
+        )
 
     past_state = zkhandler.read(("node.state.daemon", this_node.name))
     if past_state != "run" and past_state != "shutdown":
@@ -713,6 +759,7 @@ def node_keepalive(logger, config, zkhandler, this_node, netstats):
         this_node.daemon_state = "run"
 
     # Ensure the primary key is properly set
+    if debug:
         logger.out(
             "Ensure the primary key is properly set", state="d", prefix="main-thread"
         )
@@ -796,6 +843,7 @@ def node_keepalive(logger, config, zkhandler, this_node, netstats):
 
     # Set our information in zookeeper
     keepalive_time = int(time.time())
+    if debug:
         logger.out("Set our information in zookeeper", state="d", prefix="main-thread")
     try:
         zkhandler.write(
@@ -884,6 +932,7 @@ def node_keepalive(logger, config, zkhandler, this_node, netstats):
 
     # Look for dead nodes and fence them
     if not this_node.maintenance:
+        if debug:
             logger.out(
                 "Look for dead nodes and fence them", state="d", prefix="main-thread"
             )