Module: check_mk Branch: master Commit: 59d73fd74cb1622a1c73ea899f7839241d64cd7e URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=59d73fd74cb162… Author: Mathias Kettner <mk(a)mathias-kettner.de> Date: Sat Mar 2 17:05:03 2013 +0100 Prediction: select timegroup with dropdown choice --- modules/prediction.py | 1 + web/htdocs/js/prediction.js | 16 ++++++ web/htdocs/prediction.css | 2 +- web/htdocs/prediction.py | 114 ++++++++++++++++++++++++++++--------------- 4 files changed, 93 insertions(+), 40 deletions(-) diff --git a/modules/prediction.py b/modules/prediction.py index c8c3120..ea08712 100644 --- a/modules/prediction.py +++ b/modules/prediction.py @@ -270,6 +270,7 @@ def get_predictive_levels(dsname, params, cf): "range" : (from_time, until_time), "cf" : cf, "dsname" : dsname, + "slice" : period_info["slice"], } info.update(params) file(info_file, "w").write("%r\n" % info) diff --git a/web/htdocs/js/prediction.js b/web/htdocs/js/prediction.js index 09d4527..df41dbc 100644 --- a/web/htdocs/js/prediction.js +++ b/web/htdocs/js/prediction.js @@ -183,6 +183,22 @@ function line(c, t0, v0, t1, v1) c.stroke(); } +function render_point(t, v, color) +{ + var canvas = document.getElementById(canvas_id); + var c = canvas.getContext('2d'); + var p = point(t, v); + c.beginPath(); + c.lineWidth = 4; + c.strokeStyle = color; + c.moveTo(p[0]-6, p[1]-6); + c.lineTo(p[0]+6, p[1]+6); + c.moveTo(p[0]+6, p[1]-6); + c.lineTo(p[0]-6, p[1]+6); + c.stroke(); +} + + function render_curve(points, color) { var canvas = document.getElementById(canvas_id); diff --git a/web/htdocs/prediction.css b/web/htdocs/prediction.css index a0e3175..a07296c 100644 --- a/web/htdocs/prediction.css +++ b/web/htdocs/prediction.css @@ -42,7 +42,7 @@ table.prediction td { table.prediction td.legend { padding-left: 43px; color: #000000; - padding-bottom: 9px; + padding-bottom: 12px; padding-top: 2px; } diff --git a/web/htdocs/prediction.py b/web/htdocs/prediction.py index 331f581..e37e794 100644 --- a/web/htdocs/prediction.py +++ b/web/htdocs/prediction.py @@ -26,6 +26,8 @@ import defaults import os +import time + from lib import * graph_size = 2000, 700 @@ -39,60 +41,91 @@ def page_graph(): javascripts=["prediction"], stylesheets=["pages", "prediction"]) + # Get current value from perf_data via Livestatus + current_value = \ + get_current_perfdata(host, service, dsname) + dir = "%s/prediction/%s/%s/%s" % ( defaults.var_dir, host, pnp_cleanup(service), pnp_cleanup(dsname)) # Load all prediction information, sort by time of generation + tg_name = html.var("timegroup") + timegroup = None timegroups = [] for f in os.listdir(dir): if f.endswith(".info"): tg_info = eval(file(dir + "/" + f).read()) tg_info["name"] = f[:-5] timegroups.append(tg_info) + if tg_info["name"] == tg_name: + timegroup = tg_info timegroups.sort(cmp = lambda a,b: cmp(a["range"], b["range"])) - for tg_info in timegroups: - title = _("Prediction for %s") % tg_info["name"] - # time_range aus info-Datei holen - # v_range ermitteln - tg_data = eval(file(dir + "/" + tg_info["name"]).read()) - swapped = swap_and_compute_levels(tg_data, tg_info) - vertical_range = compute_vertical_range(swapped) - legend = [ - ( "#ff0000", _("Critical area") ), - ( "#ffff00", _("Warning area") ), - ( "#ffffff", _("OK area") ), - ( "#000000", _("Reference") ), - ] - create_graph(tg_info["name"], title, graph_size, tg_info["range"], vertical_range, legend) - - if "levels_upper" in tg_info: - render_dual_area(swapped["upper_warn"], swapped["upper_crit"], "#fff000", 0.4) - render_area_reverse(swapped["upper_crit"], "#ff0000", 0.1) - - if "levels_lower" in tg_info: - render_dual_area(swapped["lower_crit"], swapped["lower_warn"], "#fff000", 0.4) - render_area(swapped["lower_crit"], "#ff0000", 0.1) - - vert_scala = [ [x, "%.1f" % x] for x in range(int(vertical_range[0]), int(vertical_range[1] + 1)) ] - time_scala = [ [tg_info["range"][0] + i*3600, "%02d:00" % i] for i in range(0, 25, 2) ] - render_coordinates(vert_scala, time_scala); - - if "levels_lower" in tg_info: - render_dual_area(swapped["average"], swapped["lower_warn"], "#ffffff", 0.5) - render_curve(swapped["lower_warn"], "#e0e000") - render_curve(swapped["lower_crit"], "#j0b0a0") - - if "levels_upper" in tg_info: - render_dual_area(swapped["upper_warn"], swapped["average"], "#ffffff", 0.5) - render_curve(swapped["upper_warn"], "#e0e000") - render_curve(swapped["upper_crit"], "#f0b0b0") - render_curve(swapped["average"], "#000000") - render_curve(swapped["average"], "#000000") - # render_curve(stack(swapped["average"], swapped["stdev"], -1), "#008040") + choices = [ (tg_info["name"], tg_info["name"].title()) + for tg_info in timegroups ] + + html.begin_form("prediction") + html.write(_("Show prediction for ")) + html.select("timegroup", choices, choices[0], onchange="document.prediction.submit();") + html.hidden_fields() + html.end_form() + + title = _("Prediction for %s") % timegroup["name"] + # time_range aus info-Datei holen + # v_range ermitteln + tg_data = eval(file(dir + "/" + timegroup["name"]).read()) + swapped = swap_and_compute_levels(tg_data, timegroup) + vertical_range = compute_vertical_range(swapped) + legend = [ + ( "#000000", _("Reference") ), + ( "#ffffff", _("OK area") ), + ( "#ffff00", _("Warning area") ), + ( "#ff0000", _("Critical area") ), + ] + if current_value != None: + legend.append( ("#0000ff", _("Current value: %.2f") % current_value) ) + + create_graph(timegroup["name"], title, graph_size, timegroup["range"], vertical_range, legend) + + if "levels_upper" in timegroup: + render_dual_area(swapped["upper_warn"], swapped["upper_crit"], "#fff000", 0.4) + render_area_reverse(swapped["upper_crit"], "#ff0000", 0.1) + + if "levels_lower" in timegroup: + render_dual_area(swapped["lower_crit"], swapped["lower_warn"], "#fff000", 0.4) + render_area(swapped["lower_crit"], "#ff0000", 0.1) + + vert_scala = [ [x, "%.1f" % x] for x in range(int(vertical_range[0]), int(vertical_range[1] + 1)) ] + time_scala = [ [timegroup["range"][0] + i*3600, "%02d:00" % i] for i in range(0, 25, 2) ] + render_coordinates(vert_scala, time_scala); + + if "levels_lower" in timegroup: + render_dual_area(swapped["average"], swapped["lower_warn"], "#ffffff", 0.5) + render_curve(swapped["lower_warn"], "#e0e000") + render_curve(swapped["lower_crit"], "#j0b0a0") + + if "levels_upper" in timegroup: + render_dual_area(swapped["upper_warn"], swapped["average"], "#ffffff", 0.5) + render_curve(swapped["upper_warn"], "#e0e000") + render_curve(swapped["upper_crit"], "#f0b0b0") + render_curve(swapped["average"], "#000000") + render_curve(swapped["average"], "#000000") + # render_curve(stack(swapped["average"], swapped["stdev"], -1), "#008040") + + if current_value != None: + rel_time = time.time() % timegroup["slice"] + render_point(timegroup["range"][0] + rel_time, current_value, "#0000ff") html.footer() +def get_current_perfdata(host, service, dsname): + perf_data = html.live.query_value("GET services\nFilter: host_name = %s\nFilter: description = %s\nColumns: perf_data" % ( + host, service)) + for part in perf_data.split(): + name, rest = part.split("=") + if name == dsname: + return float(rest.split(";")[0]) + # Compute check levels from prediction data and check parameters def swap_and_compute_levels(tg_data, tg_info): columns = tg_data["columns"] @@ -165,6 +198,9 @@ def render_coordinates(v_scala, t_scala): def render_curve(points, color): html.javascript('render_curve(%r, %r);' % (points, color)) +def render_point(t, v, color): + html.javascript('render_point(%r, %r, %r);' % (t, v, color)) + def render_area(points, color, alpha=1.0): html.javascript('render_area(%r, %r, %f);' % (points, color, alpha))