Move debug condition handling to Logger

Avoids many dozens of conditionals sprinkled throughout the code by
centralizing this check into the main Logger instance.
This commit is contained in:
Joshua Boniface 2023-12-27 12:58:36 -05:00
parent 52bf5ad0ef
commit e654fbba08
4 changed files with 205 additions and 267 deletions

View File

@ -115,6 +115,10 @@ 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"))

View File

@ -157,9 +157,6 @@ 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)
#
@ -523,7 +520,6 @@ 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",

View File

@ -333,7 +333,6 @@ 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",
@ -370,7 +369,6 @@ 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",
@ -388,7 +386,6 @@ 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",
@ -404,7 +401,6 @@ 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",
@ -413,7 +409,6 @@ 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
@ -429,7 +424,6 @@ class AXFRDaemonInstance(object):
records_new.sort()
records_old.sort()
if self.config["debug"]:
self.logger.out(
"New: {}".format(records_new),
state="d",
@ -450,7 +444,6 @@ 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",
@ -487,7 +480,6 @@ 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",
@ -507,7 +499,6 @@ 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",
@ -520,7 +511,6 @@ 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
@ -529,7 +519,6 @@ 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
@ -548,7 +537,6 @@ 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",
@ -560,7 +548,6 @@ class AXFRDaemonInstance(object):
)
# Commit all the previous changes
if self.config["debug"]:
self.logger.out(
"Committing database changes and reloading PDNS",
state="d",

View File

@ -80,8 +80,6 @@ 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
@ -90,7 +88,6 @@ 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:
@ -100,7 +97,6 @@ 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",
@ -117,7 +113,6 @@ 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",
@ -134,7 +129,6 @@ 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",
@ -151,7 +145,6 @@ 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",
@ -179,7 +172,6 @@ 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",
@ -195,17 +187,13 @@ 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",
@ -248,7 +236,6 @@ 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
@ -264,7 +251,6 @@ 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(
@ -279,7 +265,6 @@ 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()
@ -293,7 +278,6 @@ 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(
@ -316,7 +300,6 @@ 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()
@ -330,7 +313,6 @@ 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"):
@ -400,7 +382,6 @@ 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()
@ -421,10 +402,7 @@ def collect_ceph_stats(logger, config, zkhandler, this_node, queue):
# Upload OSD data for the cluster (primary-only)
if this_node.coordinator_state == "primary":
if debug:
logger.out(
"Trigger updates for each OSD", state="d", prefix="ceph-thread"
)
logger.out("Trigger updates for each OSD", state="d", prefix="ceph-thread")
for osd in osd_list:
try:
@ -441,19 +419,15 @@ 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)
@ -467,7 +441,6 @@ 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",
@ -518,7 +491,6 @@ 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",
@ -537,12 +509,9 @@ 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",
)
@ -554,7 +523,6 @@ 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",
@ -578,7 +546,6 @@ 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),
@ -589,7 +556,6 @@ 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",
@ -619,12 +585,9 @@ 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",
)
@ -645,7 +608,6 @@ 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",
@ -657,7 +619,6 @@ 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",
@ -667,7 +628,6 @@ 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",
@ -679,14 +639,11 @@ 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":
@ -746,10 +703,7 @@ def node_keepalive(logger, config, zkhandler, this_node, netstats):
)
# Get past state and update if needed
if debug:
logger.out(
"Get past state and update if needed", state="d", prefix="main-thread"
)
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":
@ -759,7 +713,6 @@ 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"
)
@ -843,7 +796,6 @@ 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(
@ -932,7 +884,6 @@ 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"
)