def server_drilldown(url_key, server_id=None, compound_service_id=None, timezone=None):
timezone = request.args.get('timezone', None)
if timezone:
tz = pytz.timezone(str(timezone))
inline = server_id or compound_service_id
server_id = server_id or request.args.get('server_id', None)
compound_service_id = compound_service_id or request.args.get('compound_service_id', None)
item = None
try:
pr = masterdb.PublicReportV2.selectBy(url_key=url_key)[0]
customer = pr.customer
if customer.status == 'disabled': raise Exception
contact = customer.getMainUser().contact
if pr.require_login and "tg-visit" not in request.cookies:
url = customer.brand.controlpanel_url if customer.brand else config.CP_BASE_URL
return redirect("%s/login?forward_url=%s" % (url.strip("/"), request.url))
# Extract the item
if server_id:
item = masterdb.Server.get(int(server_id))
elif compound_service_id:
item = masterdb.CompoundService.get(int(compound_service_id))
else:
return redirect(url_for('server_group_overview', url_key=url_key))
if item not in pr.all_items: raise Exception
# Today stats
local_end_time = contact.toLocalTime(datetime.now()).replace(hour=23, minute=59, second=59)
local_start_time = local_end_time.replace(hour=0, minute=0, second=0)
end_time = contact.fromLocalTime(local_end_time).replace(tzinfo=None)
start_time = contact.fromLocalTime(local_start_time).replace(tzinfo=None)
end_time = datetime.now()
today_alerts = item.get_events(
exclude_anomalies=not pr.include_resources,
custom_filter=AND(masterdb.Event.q.start_time <= end_time,
masterdb.Event.q.status == 'confirmed')
)
recent_alerts = item.get_events(
exclude_anomalies=not pr.include_resources,
custom_filter=IN(masterdb.Event.q.status, ['confirmed', 'resolved']),
count=1
)
current_alerts = len(item.get_events(
exclude_anomalies=not pr.include_resources,
custom_filter=AND(masterdb.Event.q.start_time <= end_time + relativedelta(seconds=-5),
masterdb.Event.q.status == 'confirmed')
))
if recent_alerts:
if recent_alerts[0].end_time:
latest_outage = datetime.now() - recent_alerts[0].end_time
latest_outage = masterdb.getApproximateDuration(latest_outage, sig_digits=2, short_labels=True)
else:
latest_outage = 'Currently Down'
today_alerts_count = item.get_events_count(
exclude_anomalies=not pr.include_resources,
custom_filter=OR(AND(masterdb.Event.q.start_time <= start_time,
masterdb.Event.q.status == 'confirmed'),
AND(masterdb.Event.q.start_time >= start_time,
masterdb.Event.q.start_time <= end_time,
IN(masterdb.Event.q.status, ['confirmed', 'resolved'])))
)
current_month_start_range = end_time.replace(day=1, hour=0, second=0, minute=0)
this_month_alerts_count = item.get_events_count(
exclude_anomalies=not pr.include_resources,
custom_filter=OR(AND(masterdb.Event.q.start_time <= current_month_start_range,
masterdb.Event.q.status == 'confirmed'),
AND(masterdb.Event.q.start_time >= current_month_start_range,
masterdb.Event.q.start_time <= end_time,
IN(masterdb.Event.q.status, ['confirmed', 'resolved'])))
)
current_month_heatmap_data = masterdb.HeatmapCache.get_results_in_range(item, start_time, end_time, pr.include_resources)
total_down_time = 0
for key in current_month_heatmap_data:
total_down_time += current_month_heatmap_data[key].total_downtime
if total_down_time > 0:
delta_str = masterdb.getApproximateDuration(timedelta(seconds=total_down_time), sig_digits=None, short_labels=True)
today_availability = "%.0f%%" % (max(item.getAvailability(start_time, end_time, include_anomalies=pr.include_resources, with_excluded_outages=False)[0], 0.0) * 100)
# Maintenance schedules
current_maintenance = []
for mp in masterdb.MaintenancePeriod.select(AND(masterdb.MaintenancePeriod.q.customerID == customer.id,
masterdb.MaintenancePeriod.q.status == 'active'),
orderBy=masterdb.MaintenancePeriod.q.start_time):
ms = mp.maintenance_schedule
if ms.show_in_public_report and item in ms.servers + ms.compound_services:
current_maintenance.append(mp)
# Outages info
outages_info = ""
for alert in today_alerts:
description = alert.getNewDescription(exclude_agent_heartbeats=True, join_str="<br/><span style='margin-left: 20px'></span>", show_full_fqdn=not pr.hide_fqdn) if isinstance(alert, masterdb.ServerOutage) else alert.description
outages_info += '<p style="color:black;!important"><i class="glyphicon glyphicon-fire"></i> %s' % (description or alert.alert_label)
logs = [log for log in alert.outage_logs if log.public]
if logs:
outages_info += ":<ul>"
for log in logs:
local_time = contact.formatLocalTime(log.timestamp)
if timezone:
local_time = log.timestamp.replace(tzinfo=pytz.utc).astimezone(tz)
outages_info += '<li style="color:black;!important">[%s] %s</li>' % (local_time, log.message)
outages_info += "</ul>"
outages_info += "</p>"
# Network service graph
mps = [mp.id for mp in item.active_monitor_points if not mp.is_panopta_agent_heartbeat]
series, labels, start, end, title, width, height, label_incr_count, grid_spacing, bgcolor, zones = \
graph_utils.ServerGraphHighCharts(mps, 'hour', None, item.id, 'performance', timezone=timezone)
for s in series: del s['monitor_point_obj']
graph = dict(
title = title,
yAxisTitle = title,
xAxisLabels = labels,
series = series,
majorTickStep = label_incr_count,
minorTickStep = grid_spacing,
zones = zones
)
network_service_json_data = json.dumps(graph)
# Server resource graph
if item.isServer():
srs = [sr.id for sr in item.active_server_resources]
series, labels, start, end, title, width, height, label_incr_count, grid_spacing, bgcolor = \
graph_utils.ServerResourceGraphHighCharts(srs, 'hour', timezone=timezone)
for s, d in series.items():
for val in d: del val['server_resource']
graph = dict(
title = title,
yAxisTitle = title,
xAxisLabels = labels,
series = series,
majorTickStep = label_incr_count,
minorTickStep = grid_spacing
)
server_resource_json_data = json.dumps(graph)
# Stats table
stats_list = [
["Last 24 Hours", 1],
["Last 7 Days", 7],
["Last 30 Days", 30]
]
stats = []
end_time = datetime.now()
for n, d in stats_list:
start_time = end_time + timedelta(days=-d, seconds=1)
alerts_count = item.get_events_count(
exclude_anomalies=not pr.include_resources,
custom_filter=OR(AND(masterdb.Event.q.start_time <= start_time,
masterdb.Event.q.status == 'confirmed'),
AND(masterdb.Event.q.start_time >= start_time,
masterdb.Event.q.start_time <= end_time,
IN(masterdb.Event.q.status, ['confirmed', 'resolved'])))
)
availability = calculate_availability(end_time, item, pr, start_time)
stats.append([n, availability, alerts_count])
historical_stats_list = []
for i in range(0, 12):
start_range = end_time.replace(day=1, hour=0, second=0, minute=0) + relativedelta(months=-i)
if start_range.year > item.created.year or (start_range.year == item.created.year and start_range.month >= item.created.month):
end_range = start_range + relativedelta(months=+1, seconds=-1)
label = None
if i == 0:
label = 'Current Month'
end_range = end_time
elif i == 1:
label = 'Last Month'
else:
label = (end_time + relativedelta(months=-i)).strftime("%B")
availability = calculate_availability(end_range, item, pr, start_range)
historical_stats_list.append([label,availability])
# Heatmap
local_end_time = contact.toLocalTime(datetime.now()).replace(hour=23, minute=59, second=59)
local_start_time = (local_end_time + timedelta(weeks=-24, days=1)).replace(hour=0, minute=0, second=0)
end_time = contact.fromLocalTime(local_end_time).replace(tzinfo=None)
start_time = contact.fromLocalTime(local_start_time).replace(tzinfo=None)
yAxisLabels = []
for weekday in range(7):
next_day = local_end_time + relativedelta(days=-weekday)
yAxisLabels.append(next_day.strftime("%A"))
#heatmap_data = masterdb.HeatmapCache.create_cache_in_range(item, start_time, end_time, pr.include_resources)
#heatmap_data = dict([(contact.toLocalTime(hc.timestamp).strftime("%Y-%m-%d"), hc) for hc in heatmap_data])
heatmap_data = masterdb.HeatmapCache.get_results_in_range(item, start_time, end_time, pr.include_resources)
series = []
xAxisLabels = []
start_date = start_time
for col in range(24):
label = ''
for row in range(7):
end_date = start_date + relativedelta(days=1, seconds=-1)
local_date = contact.toLocalTime(end_date)
if local_date.day == 1: label = local_date.strftime("%b")
tooltip = '<b>%s</b><br>' % local_date.strftime("%A %B %d, %Y")
item_id = item.id if item.isServer() else -item.id
url = "%s?date=%s" % (url_for('server_drilldown_data', url_key=url_key, item_id=item_id), str(start_date))
start_date = start_date + timedelta(days=1)
hc = heatmap_data.get(local_date.strftime("%Y-%m-%d"), None)
if hc and hc.outages_count:
tooltip += '%s outages (%s)' % (hc.outages_count, hc.downtime_duration)
elif hc:
tooltip += 'No outages'
else:
tooltip += 'No data'
series.append({
'x': col,
'y': row,
'color': hc.status_color if hc else '#ffffff',
'tooltip': tooltip,
'url': url
})
xAxisLabels.append(label)
graph = dict(
xAxisLabels = xAxisLabels,
yAxisLabels = yAxisLabels[::-1],
series = series,
today_url = url,
)
heatmap_json_data = json.dumps(graph)
# Render template
report_body = render_template('server_drilldown.html', **locals())
if not inline:
template = pr.template.decode("utf-8")
response = template.replace('REPORT_BODY', unicode(report_body))
else:
response = unicode(report_body)
return response
except:
app.log("Uncought exception in PublicReport:\n%s" % traceback.format_exc(), "CRITICAL")
return redirect("http://www.panopta.com")
def server_drilldown_data(url_key, item_id):
timezone = request.args.get("timezone", '')
if timezone:
tz = pytz.timezone(str(timezone))
try:
pr = masterdb.PublicReportV2.selectBy(url_key=url_key)[0]
customer = pr.customer
if customer.status == 'disabled': raise Exception
contact = customer.getMainUser().contact
if pr.require_login and "tg-visit" not in request.cookies:
url = customer.brand.controlpanel_url if customer.brand else config.CP_BASE_URL
return redirect("%s/login?forward_url=%s" % (url.strip("/"), request.url))
# Extract the item
item_id = int(item_id)
if item_id > 0:
item = masterdb.Server.get(item_id)
else:
item = masterdb.CompoundService.get(abs(item_id))
if item not in pr.all_items: raise Exception
date = parse(request.args.get('date', str(datetime.now())))
local_start_time = contact.toLocalTime(date).replace(hour=0, minute=0, second=0)
local_end_time = local_start_time.replace(hour=23, minute=59, second=59)
start_time = contact.fromLocalTime(local_start_time).replace(tzinfo=None)
end_time = contact.fromLocalTime(local_end_time).replace(tzinfo=None)
end_time = datetime.now()
alerts = item.get_events(
exclude_anomalies=not pr.include_resources,
custom_filter=OR(AND(masterdb.Event.q.start_time <= start_time,
masterdb.Event.q.status == 'confirmed'),
AND(masterdb.Event.q.start_time >= start_time,
masterdb.Event.q.start_time <= end_time,
IN(masterdb.Event.q.status, ['confirmed', 'resolved'])))
)
availability = "%.0f%%" % (max(item.getAvailability(start_time, end_time, include_anomalies=pr.include_resources, with_excluded_outages=False)[0], 0.0) * 100)
alert_list = []
for alert in alerts:
localized_start = contact.toLocalTime(alert.start_time)
localized_end = contact.toLocalTime(alert.end_time)
if timezone:
localized_start = alert.start_time.replace(tzinfo=pytz.utc)
localized_start = localized_start.astimezone(tz)
localized_end = alert.end_time.replace(tzinfo=pytz.utc)
localized_end = localized_end.astimezone(tz)
duration = masterdb.getApproximateDuration((alert.end_time or min(datetime.now(), end_time)) - alert.start_time, sig_digits=None, short_labels=True)
description = alert.getNewDescription(exclude_agent_heartbeats=True, show_full_fqdn=not pr.hide_fqdn) if isinstance(alert, masterdb.ServerOutage) else alert.description
alert_list.append({
'start_time': localized_start.strftime("%Y-%m-%d %H:%M:%S"),
'end_time': localized_end.strftime("%Y-%m-%d %H:%M:%S") if alert.end_time else None,
'duration': duration,
'description': description or alert.alert_label,
})
return jsonify(
availability = availability,
alerts_count = len(alerts),
title_date = local_start_time.strftime("%B %d, %Y"),
alerts = alert_list
)
except:
app.log("Uncought exception in PublicReport:\n%s" % traceback.format_exc(), "CRITICAL")
return jsonify(
date = '',
availability = '100%',
alerts_count = 0,
title_date = '',
alerts = []
)
def server_group_overview(url_key=None, pr=None):
try:
if url_key:
pr = masterdb.PublicReportV2.selectBy(url_key=url_key)[0]
elif request.host:
pr = masterdb.PublicReportV2.selectBy(hostname=request.host.split(":")[0])[0]
url_key = pr.url_key
customer = pr.customer
if customer.status == 'disabled': raise Exception
contact = customer.getMainUser().contact
if pr.require_login and "tg-visit" not in request.cookies:
url = customer.brand.controlpanel_url if customer.brand else config.CP_BASE_URL
return redirect("%s/login?forward_url=%s" % (url.strip("/"), request.url))
local_end_time = contact.toLocalTime(datetime.now()).replace(hour=23, minute=59, second=59)
local_start_time = (local_end_time + timedelta(days=-6)).replace(hour=0, minute=0, second=0)
end_time = contact.fromLocalTime(local_end_time).replace(tzinfo=None)
start_time = contact.fromLocalTime(local_start_time).replace(tzinfo=None)
# Get dates for the past week
days = []
day = local_end_time
for i in range(7):
if i == 0:
days.append("Today")
elif i == 1:
days.append("Yesterday")
else:
days.append(day.strftime("%a - %m/%d"))
day += timedelta(days=-1)
# Get server group data
server_groups = pr.groups.items()
server_groups.sort(key=lambda (sg, items): sg.name.lower())
for sg, items in server_groups:
sg.detail_link = (url_for('server_group_drilldown', url_key=url_key, server_group_id=sg.id))
# Sorting
if pr.sort_order == 'avail_asc':
items.sort(key=lambda item: [max(item.getAvailability(start_time, end_time, include_anomalies=pr.include_resources)[0], 0.0), item.formatted_name.lower()])
elif pr.sort_order == 'avail_desc':
items.sort(key=lambda item: [-max(item.getAvailability(start_time, end_time, include_anomalies=pr.include_resources)[0], 0.0), item.formatted_name.lower()])
else:
items.sort(key=lambda item: item.formatted_name.lower())
for item in items:
if item.status != 'active' or (item.isServer() and item.isTemplate()): continue
item.active_alerts_count = item.get_events_count(status='confirmed', exclude_anomalies=not pr.include_resources)
# Getting detail and cp links
if item.isServer():
item.detail_link = "%s?server_id=%s" % (url_for('server_drilldown', url_key=url_key), item.id)
item.cp_link = "%s/report/ServerReport?server_id=%s" % (config.CP_BASE_URL, item.id)
else:
item.detail_link = "%s?compound_service_id=%s" % (url_for('server_drilldown', url_key=url_key), item.id)
item.cp_link = "%s/report/CompoundServiceReport?compound_service_id=%s" % (config.CP_BASE_URL, item.id)
# Getting stats for each item in each day
stats = []
end = end_time
for i in range(7):
start = end + timedelta(days=-1, seconds=1)
alerts = item.get_events(
exclude_anomalies=not pr.include_resources,
custom_filter=OR(AND(masterdb.Event.q.start_time <= start,
masterdb.Event.q.status == 'confirmed'),
AND(masterdb.Event.q.start_time >= start,
masterdb.Event.q.start_time <= end,
IN(masterdb.Event.q.status, ['confirmed', 'resolved'])))
)
# Getting status color
if alerts:
color = "red" if any([alert.severity == "critical" for alert in alerts]) else "orange"
else:
color = "green"
# Getting tooltip info
tooltip = '<div class="alert alert-%s"><a href="%s">%s</a></div>' % ("success" if not alerts else "danger", item.cp_link, item.name if pr.hide_fqdn else item.formatted_name)
if not alerts:
text = 'No outages'
log_text = ""
else:
log_text = ""
critical_outages_count = 0
critical_outages_duration = timedelta(seconds=0)
warning_outages_count = 0
warning_outages_duration = timedelta(seconds=0)
for alert in alerts:
et = alert.end_time or min(datetime.now(), end)
if alert.severity == "critical":
critical_outages_count += 1
if et > alert.start_time:
critical_outages_duration += (et - alert.start_time)
else:
warning_outages_count += 1
if et > alert.start_time:
warning_outages_duration += (et - alert.start_time)
description = alert.getNewDescription(exclude_agent_heartbeats=True, show_full_fqdn=not pr.hide_fqdn) if isinstance(alert, masterdb.ServerOutage) else alert.description
log_text += '<p><i class="glyphicon glyphicon-fire"></i> <a href="%s/outage/Outage?outage_id=%s">%s</a>' % (config.CP_BASE_URL, alert.alert_id, description or alert.alert_label)
logs = [log for log in alert.outage_logs if log.public]
if logs:
log_text += ":<ul>"
for log in logs:
log_text += "<li>[<span class='local-date'>%s</span>] %s</li>" % (contact.formatLocalTime(log.timestamp), log.message)
log_text += "</ul>"
log_text += "</p>"
log_text += "<span class='hidden contact-timezone'>%s</span>" % contact.getTimezoneObject()
critical_outages_duration = masterdb.getApproximateDuration(critical_outages_duration, sig_digits=None, short_labels=True)
warning_outages_duration = masterdb.getApproximateDuration(warning_outages_duration, sig_digits=None, short_labels=True)
text = "%s critical outage%s (%s)" % (critical_outages_count, 's' if critical_outages_count > 1 else '', critical_outages_duration)
text += "<br>%s warnings (%s)" % (warning_outages_count, warning_outages_duration)
tooltip += '<p class="text-center"><strong>%s</strong></p>' % text
if log_text:
tooltip += "<hr/>" + log_text
stats.append({"color": color, "tooltip": tooltip})
end += timedelta(days=-1)
item.stats = stats
# Render template
report_body = render_template('overview.html', **locals())
template = pr.template.decode("utf-8")
response = template.replace('REPORT_BODY', unicode(report_body))
return response
except:
app.log("Uncought exception in PublicReport:\n%s" % traceback.format_exc(), "CRITICAL")
return redirect("http://www.panopta.com")
