def main(rule_name=None):
RUN_METADATA = {
'RUN_ID': RUN_ID,
'RUN_TYPE': 'ALERT QUERY',
'START_TIME': datetime.datetime.utcnow(),
}
if rule_name:
metadata = [create_alerts(rule_name)]
else:
rules = list(db.load_rules(ALERT_QUERY_POSTFIX))
metadata = Pool(POOLSIZE).map(create_alerts, rules)
RUN_METADATA['ROW_COUNT'] = {
'INSERTED': sum(q['ROW_COUNT']['INSERTED'] for q in metadata),
'UPDATED': sum(q['ROW_COUNT']['UPDATED'] for q in metadata),
}
db.record_metadata(RUN_METADATA, table=RUN_METADATA_TABLE)
try:
if CLOUDWATCH_METRICS:
log.metric(
'Run',
'SnowAlert',
[{'Name': 'Component', 'Value': 'Alert Query Runner'}],
1,
)
except Exception as e:
log.error("Cloudwatch metric logging failed: ", e)
def main():
RUN_METADATA = {
'RUN_TYPE': 'VIOLATION SUPPRESSION',
'START_TIME': datetime.datetime.utcnow(),
'RUN_ID': RUN_ID,
}
for squelch_name in db.load_rules(VIOLATION_SQUELCH_POSTFIX):
run_suppression(squelch_name)
num_violations_passed = next(
db.fetch(SET_SUPPRESSED_FALSE))['number of rows updated']
RUN_METADATA['ROW_COUNT'] = {
'SUPPRESSED':
sum(rmr['ROW_COUNT']['SUPPRESSED'] for rmr in RULE_METADATA_RECORDS),
'PASSED':
num_violations_passed,
}
db.record_metadata(RUN_METADATA, table=RUN_METADATA_TABLE)
if CLOUDWATCH_METRICS:
log.metric(
'Run',
'SnowAlert',
[{
'Name': 'Component',
'Value': 'Violation Suppression Runner'
}],
1,
)
def main(squelch_name=None):
RUN_METADATA = {
'RUN_TYPE': 'ALERT SUPPRESSION',
'START_TIME': datetime.datetime.utcnow(),
'RUN_ID': RUN_ID,
}
rules = db.load_rules(ALERT_SQUELCH_POSTFIX) if squelch_name is None else [
squelch_name
]
for squelch_name in rules:
run_suppressions(squelch_name)
num_rows_updated = next(
db.fetch(SET_SUPPRESSED_FALSE))['number of rows updated']
log.info(
f'All suppressions done, {num_rows_updated} remaining alerts marked suppressed=FALSE.'
)
RUN_METADATA['ROW_COUNT'] = {
'PASSED': num_rows_updated,
'SUPPRESSED':
sum(m['ROW_COUNT']['SUPPRESSED'] for m in METADATA_HISTORY),
}
db.record_metadata(RUN_METADATA, table=RUN_METADATA_TABLE)
try:
if CLOUDWATCH_METRICS:
log.metric('Run', 'SnowAlert', [{
'Name': 'Component',
'Value': 'Alert Suppression Runner'
}], 1)
except Exception as e:
log.error("Cloudwatch metric logging failed: ", e)
def main():
RUN_METADATA = {}
RUN_METADATA['RUN_TYPE'] = 'ALERT SUPPRESSION'
RUN_METADATA['START_TIME'] = datetime.datetime.utcnow()
RUN_METADATA['RUN_ID'] = RUN_ID
ctx = db.connect()
for squelch_name in db.load_rules(ctx, ALERT_SQUELCH_POSTFIX):
run_suppressions(squelch_name)
flag_remaining_alerts(ctx)
log.metadata_record(ctx, RUN_METADATA, table=RUN_METADATA_TABLE)
if CLOUDWATCH_METRICS:
log.metric('Run', 'SnowAlert', [{'Name': 'Component', 'Value': 'Alert Suppression Runner'}], 1)
def main():
# Force warehouse resume so query runner doesn't have a bunch of queries waiting for warehouse resume
RUN_METADATA = {}
RUN_METADATA['RUN_TYPE'] = 'VIOLATION QUERY'
RUN_METADATA['START_TIME'] = datetime.datetime.utcnow()
RUN_METADATA['RUN_ID'] = RUN_ID
ctx = db.connect_and_execute("ALTER SESSION SET use_cached_result=FALSE;")
for query_name in db.load_rules(ctx, VIOLATION_QUERY_POSTFIX):
run_query(query_name)
log.metadata_record(ctx, RUN_METADATA, table=RUN_METADATA_TABLE)
if CLOUDWATCH_METRICS:
log.metric('Run', 'SnowAlert', [{
'Name': 'Component',
'Value': 'Violation Query Runner'
}], 1)
def main(rule_name=None):
# Force warehouse resume so query runner doesn't have a bunch of queries waiting for warehouse resume
RUN_METADATA = {}
RUN_METADATA['RUN_TYPE'] = 'ALERT QUERY'
RUN_METADATA['START_TIME'] = datetime.datetime.utcnow()
RUN_METADATA['RUN_ID'] = RUN_ID
ctx = db.connect_and_execute("ALTER SESSION SET USE_CACHED_RESULT=FALSE;")
if rule_name:
query_for_alerts(ctx, rule_name)
else:
for query_name in db.load_rules(ctx, ALERT_QUERY_POSTFIX):
query_for_alerts(ctx, query_name)
log.metadata_record(ctx, RUN_METADATA, table=RUN_METADATA_TABLE)
if CLOUDWATCH_METRICS:
log.metric('Run', 'SnowAlert', [{
'Name': 'Component',
'Value': 'Alert Query Runner'
}], 1)
def main():
ctx = db.connect()
alert_rows = list(get_new_alerts(ctx))
log.info(f'Found {len(alert_rows)} new alerts to handle.')
for alert_row in alert_rows:
alert = alert_row['ALERT']
results = []
handlers = alert.get('HANDLERS')
if handlers is None:
handlers = 'jira'
if isinstance(handlers, (str, dict)):
handlers = [handlers]
for handler in handlers:
if handler is None:
results.append(None)
else:
if type(handler) is str:
handler = {'type': handler}
if 'type' not in handler:
result = {
'success': False,
'error': 'missing type key',
'details': handler,
}
else:
handler_type = handler['type']
handler_kwargs = handler.copy()
handler_kwargs.update({
'alert':
alert,
'correlation_id':
alert_row.get('CORRELATION_ID'),
'alert_count':
alert_row['COUNTER'],
})
try:
handler_module = importlib.import_module(
f'runners.handlers.{handler_type}')
result = {
'success':
True,
'details':
apply_some(handler_module.handle,
**handler_kwargs),
}
except Exception as e:
result = {'success': False, 'details': e}
results.append(result)
record_status(results, alert['ALERT_ID'])
try:
if CLOUDWATCH_METRICS:
log.metric('Run', 'SnowAlert', [{
'Name': 'Component',
'Value': 'Alert Handler'
}], 1)
except Exception as e:
log.error("Cloudwatch metric logging failed", e)