Checkmk git commits March 2016

checkmk-commits@lists.checkmk.com

7 participants
275 discussions

by Lars Michelsen

Module: check_mk Branch: master Commit: a409354c7ac09b7bdb3ecbb4b1eaf9f9037d11c6 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=a409354c7ac09b… Author: Lars Michelsen <lm(a)mathias-kettner.de> Date: Fri Mar 18 09:26:55 2016 +0100 Fixed broken WATO host search --- web/htdocs/watolib.py | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/web/htdocs/watolib.py b/web/htdocs/watolib.py index e0e8ccc..691bff9 100644 --- a/web/htdocs/watolib.py +++ b/web/htdocs/watolib.py @@ -623,11 +623,18 @@ class Folder(BaseFolder): # Find folder that is specified by the current URL. This is either by a folder # path in the variable "folder" or by a host name in the variable "host". In the # latter case we need to load all hosts in all folders and actively search the host. + # Another case is the host search which has the "host_search" variable set. To handle + # the later case we call .current() of SearchFolder() to let it decide whether or not + # this is a host search. This method has to return a folder in all cases. @staticmethod def current(): if html.is_cached("wato_current_folder"): return html.get_cached("wato_current_folder") + folder = SearchFolder.current_search_folder() + if folder: + return folder + if html.has_var("folder"): folder = Folder.folder(html.var("folder")) else: @@ -1706,20 +1713,15 @@ class SearchFolder(BaseFolder): return crit + # This method is allowed to return None when no search is currently performed. @staticmethod - def current(): - if html.is_cached("wato_current_folder"): - return html.get_cached("wato_current_folder") - + def current_search_folder(): if html.has_var("host_search"): base_folder = Folder.folder(html.var("folder", "")) search_criteria = SearchFolder.criteria_from_html_vars() folder = SearchFolder(base_folder, search_criteria) Folder.set_current(folder) - return - else: - # FIXME: Is this needed for SearchFolder()? - return Folder.current() + return folder # .--------------------------------------------------------------------.

8 years, 6 months

3292 livestatus_status: Added Microcore relevant metrics like helper usage and latency

by Lars Michelsen

Module: check_mk Branch: master Commit: cf692e0b75b77f31d60d23a54901f49a6f269500 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=cf692e0b75b77f… Author: Lars Michelsen <lm(a)mathias-kettner.de> Date: Fri Mar 18 08:58:08 2016 +0100 3292 livestatus_status: Added Microcore relevant metrics like helper usage and latency The livestatus_status check was missing some performance indicators like the helper usage and check latency which are only available in the Microcore. These metrics have now been added to this check. --- .werks/3292 | 12 +++++++++ ChangeLog | 1 + checks/livestatus_status | 36 ++++++++++++++++++++------- web/plugins/metrics/check_mk.py | 52 +++++++++++++++++++++++++++++++++++++++ 4 files changed, 92 insertions(+), 9 deletions(-) diff --git a/.werks/3292 b/.werks/3292 new file mode 100644 index 0000000..1c58a9a --- /dev/null +++ b/.werks/3292 @@ -0,0 +1,12 @@ +Title: livestatus_status: Added Microcore relevant metrics like helper usage and latency +Level: 1 +Component: checks +Class: feature +Compatible: compat +State: unknown +Version: 1.2.9i1 +Date: 1458287805 + +The livestatus_status check was missing some performance indicators like the helper usage +and check latency which are only available in the Microcore. These metrics have now been +added to this check. diff --git a/ChangeLog b/ChangeLog index c69076f..2b9b397 100644 --- a/ChangeLog +++ b/ChangeLog @@ -33,6 +33,7 @@ * 3303 cisco_fru_module_status: new check which monitors the operational status of FRU modules of Cisco devices which support the CISCO-ENTITY-FRU-CONTROL-MIB * 3012 jolokia_metrics.threads: Improved check output, now showing the actual problem and levels * 3272 oracle_tablespaces: severity of check for autoextension is now configurable... + * 3292 livestatus_status: Added Microcore relevant metrics like helper usage and latency... * 3073 FIX: windows agent: relative paths to mrpe scripts are now treated as relative to the agent installation directory... * 3061 FIX: mk_jolokia: Fixed debugging of the agent plugin * 3074 FIX: windows agent: fixed incorrect values for 32-bit performance counters diff --git a/checks/livestatus_status b/checks/livestatus_status index 0ba088e..8d50d4c 100644 --- a/checks/livestatus_status +++ b/checks/livestatus_status @@ -84,17 +84,35 @@ def check_livestatus_status(item, params, parsed): # Check Performance counters this_time = time.time() status_counters = [ - ( "host_checks", "Host Checks", ), - ( "service_checks", "Service Checks", ), - ( "forks", "Process Creations", ), - ( "connections", "Livestatus Connects", ), - ( "requests", "Livestatus Requests", ), - ( "log_messages", "Log Messages", ), + # conv_func, factor, calc_rate, column, format, title + (int, 1, True, "%.1f/s", "host_checks", "Host Checks", ), + (int, 1, True, "%.1f/s", "service_checks", "Service Checks", ), + (int, 1, True, "%.1f/s", "forks", "Process Creations", ), + (int, 1, True, "%.1f/s", "connections", "Livestatus Connects", ), + (int, 1, True, "%.1f/s", "requests", "Livestatus Requests", ), + (int, 1, True, "%.1f/s", "log_messages", "Log Messages", ), ] - for counter, title in status_counters: - rate = get_rate("livestatus_status.%s.%s" % (item, counter), this_time, int(status[counter])) - yield 0, ("%.1f %s/s" % (rate, title)), [(counter, rate)] + if status["program_version"].startswith("Check_MK"): + # We have a CMC here. + + status_counters += [ + # conv_func, factor, calc_rate, format, column, title + (float, 1, False, "%.3fs", "average_latency_generic", "Average check latency"), + (float, 1, False, "%.3fs", "average_latency_cmk", "Average Check_MK latency"), + (float, 100, False, "%.1f%%", "helper_usage_generic", "Check helper usage"), + (float, 100, False, "%.1f%%", "helper_usage_cmk", "Check_MK helper usage"), + (float, 100, False, "%.1f%%", "livestatus_usage", "Livestatus usage"), + (float, 1, False, "%.1f/s", "livestatus_overflows_rate", "Livestatus overflow rate"), + ] + + for conv_func, factor, calc_rate, fmt, counter, title in status_counters: + value = factor * conv_func(status[counter]) + + if calc_rate: + value = get_rate("livestatus_status.%s.%s" % (item, counter), this_time, value) + + yield 0, ("%s: %s" % (title, fmt % value)), [(counter, value)] yield 0, "%d Hosts" % int(status["num_hosts"]), [("monitored_hosts", int(status["num_hosts"]))] yield 0, "%d Services" % int(status["num_services"]), [("monitored_services", int(status["num_services"]))] diff --git a/web/plugins/metrics/check_mk.py b/web/plugins/metrics/check_mk.py index 8056665..45f0b02 100644 --- a/web/plugins/metrics/check_mk.py +++ b/web/plugins/metrics/check_mk.py @@ -2086,6 +2086,42 @@ metric_info["livestatus_request_rate"] = { "color" : "#bbccdd", } +metric_info["helper_usage_cmk"] = { + "title" : _("Check_MK helper usage"), + "unit" : "%", + "color" : "15/a", +} + +metric_info["helper_usage_generic"] = { + "title" : _("Generic helper usage"), + "unit" : "%", + "color" : "41/a", +} + +metric_info["average_latency_cmk"] = { + "title" : _("Check_MK check latency"), + "unit" : "s", + "color" : "15/a", +} + +metric_info["average_latency_generic"] = { + "title" : _("Check latency"), + "unit" : "s", + "color" : "41/a", +} + +metric_info["livestatus_usage"] = { + "title" : _("Livestatus usage"), + "unit" : "%", + "color" : "12/a", +} + +metric_info["livestatus_overflows_rate"] = { + "title" : _("Livestatus overflows"), + "unit" : "1/s", + "color" : "16/a", +} + metric_info["log_message_rate"] = { "title" : _("Log messages"), "unit" : "1/s", @@ -5825,6 +5861,22 @@ graph_info.append({ }) graph_info.append({ + "title" : _("Check helper usage"), + "metrics" : [ + ( "helper_usage_cmk", "area" ), + ( "helper_usage_generic", "area" ), + ], +}) + +graph_info.append({ + "title" : _("Average check latency"), + "metrics" : [ + ( "average_latency_cmk", "area" ), + ( "average_latency_generic", "area" ), + ], +}) + +graph_info.append({ "title" : _("Pending updates"), "metrics" : [ ( "normal_updates", "stack" ),

8 years, 6 months

3272 oracle_tablespaces: severity of check for autoextension is now configurable

by Mathias Kettner

Module: check_mk Branch: master Commit: 1514b256f1b1563145be068a399624ef0b2ace60 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=1514b256f1b156… Author: Mathias Kettner <mk(a)mathias-kettner.de> Date: Thu Mar 17 16:25:01 2016 +0100 3272 oracle_tablespaces: severity of check for autoextension is now configurable There is now a new check parameter that defines the check state in case that the current autoextension setting does not match the expected setting. The expected setting is either determined during discovery (default) or by an explicit value in the rule. By setting the state in such a case to OK the checking for autoextension is effectlivey turned off. Since the expected autoextension setting and the severity are two different parameters you can easily create independent rules for the expected setting and the severity of a mismatch. --- .werks/3272 | 19 +++++++++ ChangeLog | 1 + checkman/oracle_tablespaces | 72 ++++------------------------------ checks/oracle_tablespaces | 22 ++++++++--- web/plugins/wato/check_parameters.py | 16 ++++++-- 5 files changed, 56 insertions(+), 74 deletions(-) diff --git a/.werks/3272 b/.werks/3272 new file mode 100644 index 0000000..5959e13 --- /dev/null +++ b/.werks/3272 @@ -0,0 +1,19 @@ +Title: oracle_tablespaces: severity of check for autoextension is now configurable +Level: 1 +Component: checks +Compatible: compat +Version: 1.2.9i1 +Date: 1458228140 +Class: feature + +There is now a new check parameter that defines the check state in case that +the current autoextension setting does not match the expected setting. The +expected setting is either determined during discovery (default) or by an +explicit value in the rule. + +By setting the state in such a case to OK the checking for autoextension +is effectlivey turned off. + +Since the expected autoextension setting and the severity are two different +parameters you can easily create independent rules for the expected setting +and the severity of a mismatch. diff --git a/ChangeLog b/ChangeLog index 290a20a..c69076f 100644 --- a/ChangeLog +++ b/ChangeLog @@ -32,6 +32,7 @@ * 3103 brocade_optical: new check to monitor signal strength of optical network ports on Brocade switches * 3303 cisco_fru_module_status: new check which monitors the operational status of FRU modules of Cisco devices which support the CISCO-ENTITY-FRU-CONTROL-MIB * 3012 jolokia_metrics.threads: Improved check output, now showing the actual problem and levels + * 3272 oracle_tablespaces: severity of check for autoextension is now configurable... * 3073 FIX: windows agent: relative paths to mrpe scripts are now treated as relative to the agent installation directory... * 3061 FIX: mk_jolokia: Fixed debugging of the agent plugin * 3074 FIX: windows agent: fixed incorrect values for 32-bit performance counters diff --git a/checkman/oracle_tablespaces b/checkman/oracle_tablespaces index c5b2304..06eb929 100644 --- a/checkman/oracle_tablespaces +++ b/checkman/oracle_tablespaces @@ -9,17 +9,13 @@ description: and the autoextension and optionally the setting of the increment size. - {Autoextension}: During inventory the autoextension - setting is determined and frozen into the check parameter. - That means that if the auto extension setting changes - later, the check will get critical. A table space may - consist of several data files with different settings. - If at least one data file has auto extension turned on, - the table space is considered to have auto extension - turned on. You can turn off auto extension checking - by setting the first check parameter to {None} rather - then to {True} or {False}. Inventory will do this for - you if you set {oracle_tablespaces_check_autoext} to {False}. + {Autoextension}: During inventory the autoextension setting is determined + and remembered for later. That means that if the auto extension setting + changes later, the check will get critical. A table space may consist of + several data files with different settings. If at least one data file + has auto extension turned on, the table space is considered to have auto + extension turned on. The behaviour of autoextension setting can be configured + with a check parameter. {Increment size}: If the configuration parameter {oracle_tablespaces_check_default_increment} is set to @@ -50,16 +46,6 @@ item: A string composited of the SID and the tablespace name, separated by a dot, for example {"pengt.TEMP"}. -examples: - # Set levels for free space to 20% (WARN) and 10% (CRIT) - # of maximum tablespace size and deactivate check of autoextend - - - oracle_tablespaces_defaults = { - "levels" : (20.0, 10.0), - "autoextend" : None - } - perfdata: Three values in bytes: the current size, the used size of the tablespace and the maximum size of the table space. @@ -68,47 +54,3 @@ inventory: One separate check will be created for each tablespace which is {ONLINE}. -[parameters] -parameters (dict): - - {"autoextend"}: Either {True} or {False} - depending on whether auto extension is - expected to be on or off - - {"levels"}: By default a tuple containing the warn and crit levels for free space. - Supported types (int / float). If this is a float, it is considered as - percentage of the maximum size of the tablespace. If it is an integer, - it is considered to be MB of free space. If the space drops to this level or below, - the check will get into warning state. Default is {(10.0, 5.0)}. - This check also supports some kind of dynamic level definition. Using this you can - apply thresholds to tablespace checks depending on their size, just like using the - magic factor, but more simple do configure. To configure thi, you have to set this - variable to a list of tuples, where the first element is an integer representing - the minimum tablespace size in bytes to apply these thresholds to and a tuple - containing the thresholds as described above. - - {"magic"}: The magic number, which scales the levels according to the tablespace size. - Default is a factor of 1.0, which does no adaptation. A factor less than 1.0 lowers the - percentual levels for the free space in large tablespaces and raises them for small ones. - - {"magic_normsize"}: Norm size of a tablespace for the magic computation - The default for this is {1000}. It is measured in {MB}. - - {"magic_maxlevels"}: Maximum levels if using magic factor. The warn / crit levels will go beyond - these numbers. The preset is set to (60.0, 50.0) - -[configuration] -oracle_tablespaces_check_autoext(bool): This boolean setting determines, whether - inventory should activate monitoring of the auto extension setting. Per default - this is set to {True}, so the current state of the auto extension will be - put into the check parameters as either {True} or {False}. If you set the - configuration variable to {False}, then inventory will enter {None} as - check parameter und thus disable the checking of the auto extensible flag - for newly inventorized table spaces. - -oracle_tablespaces_check_default_increment(bool): This setting does not - control inventory but is used live during the checking und thus affects - all checks, not only newly inventorized ones. Per default it is set to - {True}. That way the check will warn if the table space has its increment - size set to {1} (the ORACLE default value). - -oracle_tablespaces_defaults(dict): A dictionary with the optional keys mentioned in parameters diff --git a/checks/oracle_tablespaces b/checks/oracle_tablespaces index 6d07628..c939aa5 100644 --- a/checks/oracle_tablespaces +++ b/checks/oracle_tablespaces @@ -45,7 +45,9 @@ factory_settings["oracle_tablespaces_defaults"] = { "defaultincrement": True, } -# Whether to check auto extend settings +# Whether to check auto extend settings. Note: this setting is not ment to +# be changed anymore. It cannot be edited via WATO either. There now exists +# a check parameter, where the behaviour can be configured on a per-service-base. oracle_tablespaces_check_autoext = True # this parameter is deprecated and needed for old configurations with @@ -267,11 +269,21 @@ def check_oracle_tablespaces(item, params, parsed): # Check autoextend status if parameter not set to None if autoext != None: if autoext and num_extensible == 0: - infotext += ", NO AUTOEXTEND(!!)" - status = 2 + autoext_info = "NO AUTOEXTEND" elif not autoext and num_extensible > 0: - infotext += ", AUTOTEXTEND(!!)" - status = 2 + autoext_info = "AUTOTEXTEND" + else: + autoext_info = None + + if autoext_info: + autoextend_severity = params.get("autoextend_severity", 2) + if autoextend_severity == 2 or status == 2: + status = 2 + else: + status = max(autoextend_severity, status) + infotext += ", " + autoext_info + state_markers[autoextend_severity] + + elif num_extensible > 0: infotext += ", autoextend" else: diff --git a/web/plugins/wato/check_parameters.py b/web/plugins/wato/check_parameters.py index 399dd78..bbf30a6 100644 --- a/web/plugins/wato/check_parameters.py +++ b/web/plugins/wato/check_parameters.py @@ -5694,10 +5694,18 @@ register_check_parameters( "This rule allows you to define checks on the size of tablespaces."), elements = db_levels_common + [ ( "autoextend", - Checkbox( - title = _("Autoextend"), - label = _("Autoextension is expected"), - help = "")), + DropdownChoice( + title = _("Expected autoextend setting"), + choices = [ + ( True, _("Autoextend is expected to be ON")), + ( False, _("Autoextend is expected to be OFF")), + ] + )), + ( "autoextend_severity", + MonitoringState( + title = _("Severity of invalid autoextend setting"), + default_value = 2, + )), ( "defaultincrement", Checkbox( title = _("Default Increment"),

8 years, 6 months

Fixed confusing naming of host graph view

by Lars Michelsen

Module: check_mk Branch: master Commit: a5750465df945959020604c242a4f6f8657fa3e3 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=a5750465df9459… Author: Lars Michelsen <lm(a)mathias-kettner.de> Date: Thu Mar 17 15:08:40 2016 +0100 Fixed confusing naming of host graph view --- web/plugins/views/builtin.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/web/plugins/views/builtin.py b/web/plugins/views/builtin.py index 37be945..f43f295 100644 --- a/web/plugins/views/builtin.py +++ b/web/plugins/views/builtin.py @@ -1576,8 +1576,8 @@ multisite_builtin_views.update({ 'sorters': [('site', False), ('site_host', False), ('svcdescr', False)], - 'linktitle': _('Host graphs'), - 'title': _('Graphs for host'), + 'linktitle': _('Service graphs'), + 'title': _('Service graphs of host'), }, 'recentsvc': {'browser_reload': 30, 'column_headers': 'pergroup',

8 years, 6 months

Fixed link from host graphs to host graph detail page

by Lars Michelsen

Module: check_mk Branch: master Commit: 954d6f1081cb16d803d6da222aae25137ded82c7 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=954d6f1081cb16… Author: Lars Michelsen <lm(a)mathias-kettner.de> Date: Thu Mar 17 15:04:04 2016 +0100 Fixed link from host graphs to host graph detail page --- web/plugins/icons/builtin.py | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/web/plugins/icons/builtin.py b/web/plugins/icons/builtin.py index 3e3db24..5486ff1 100644 --- a/web/plugins/icons/builtin.py +++ b/web/plugins/icons/builtin.py @@ -329,13 +329,17 @@ def new_graphing_url(row, what): site_id = row["site"] urivars = [ - ("view_name", "service_graphs"), ("siteopt", site_id), ("host", row["host_name"]), ] if what == "service": - urivars.append(("service", row["service_description"])) + urivars += [ + ("service", row["service_description"]), + ("view_name", "service_graphs"), + ] + else: + urivars.append(("view_name", "host_graphs")) return html.makeuri_contextless(urivars, filename="view.py")

8 years, 6 months

3238 FIX if.include: Fixed handling of old-skool traffic limits

by Sven Panne

Module: check_mk Branch: master Commit: 5915487a0f47ae991bdee90490e22aa50e5ae310 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=5915487a0f47ae… Author: Sven Panne <sp(a)mathias-kettner.de> Date: Thu Mar 17 14:53:57 2016 +0100 3238 FIX if.include: Fixed handling of old-skool traffic limits --- .werks/3238 | 10 ++++++++++ ChangeLog | 1 + checks/if.include | 4 +++- 3 files changed, 14 insertions(+), 1 deletion(-) diff --git a/.werks/3238 b/.werks/3238 new file mode 100644 index 0000000..546f4c0 --- /dev/null +++ b/.werks/3238 @@ -0,0 +1,10 @@ +Title: if.include: Fixed handling of old-skool traffic limits +Level: 1 +Component: checks +Class: fix +Compatible: compat +State: unknown +Version: 1.2.9i1 +Date: 1458222777 + + diff --git a/ChangeLog b/ChangeLog index 191c8a7..290a20a 100644 --- a/ChangeLog +++ b/ChangeLog @@ -87,6 +87,7 @@ * 3009 FIX: ESX agent: fixed exception with utf-8 encoded strings... * 3010 FIX: Fixed vanished snmp subchecks for snmp hosts... * 3237 FIX: brocade_vdx_status: Fixed broken discovery + * 3238 FIX: if.include: Fixed handling of old-skool traffic limits Multisite: * 3187 notification view: new filter for log command via regex diff --git a/checks/if.include b/checks/if.include index fe3b795..edb20f3 100644 --- a/checks/if.include +++ b/checks/if.include @@ -136,7 +136,9 @@ def if_get_traffic_levels(params): new_traffic = [] if 'traffic' in params and type(params['traffic']) != list: warn, crit = params['traffic'] - if type(warn) == int: + if warn == None: + new_traffic.append(('both', ('upper', (None, (None, None))))) + elif type(warn) == int: new_traffic.append(('both', ('upper', ('abs', (warn, crit))))) elif type(warn) == float: new_traffic.append(('both', ('upper', ('perc', (warn, crit)))))

8 years, 6 months

Preparation for Check_MK Graphs in notifications

by Lars Michelsen

Module: check_mk Branch: master Commit: 4b95823775e91c845549a2b4ab8dbef24cf2bfbb URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=4b95823775e91c… Author: Lars Michelsen <lm(a)mathias-kettner.de> Date: Thu Mar 17 14:42:52 2016 +0100 Preparation for Check_MK Graphs in notifications --- notifications/mail | 149 ++++++++++++++++++++++++++++++----------- web/htdocs/html_mod_python.py | 4 ++ 2 files changed, 113 insertions(+), 40 deletions(-) diff --git a/notifications/mail b/notifications/mail index c2a2c81..75039c4 100755 --- a/notifications/mail +++ b/notifications/mail @@ -36,6 +36,8 @@ import os, re, sys, subprocess +import urllib2 +import base64 try: from email.mime.multipart import MIMEMultipart @@ -51,6 +53,11 @@ except ImportError: from email import Encoders MIMEApplication = None +try: + from simplejson import json +except ImportError: + import json + tmpl_head_html = ''' <html> <head> @@ -464,6 +471,103 @@ def extract_graph_error(output): return output +# Fetch graphs for this object. It first tries to detect how many sources +# are available for this object. Then it loops through all sources and +# fetches retrieves the images. If a problem occurs, it is printed to +# stderr (-> notify.log) and the graph is not added to the mail. +def render_pnp_graphs(context): + try: + num_sources = fetch_num_sources(context) + except GraphException, e: + graph_error = extract_graph_error(str(e)) + if '.xml" not found.' not in graph_error: + sys.stderr.write('Unable to fetch number of graphs: %s\n' % graph_error) + num_sources = 0 + + graph_list = [] + for source in range(0, num_sources): + try: + content = fetch_graph(context, source) + except GraphException, e: + sys.stderr.write('Unable to fetch graph: %s\n' % e) + continue + + graph_list.append(content) + + return graph_list + + +def render_cmk_graphs(context): + if context["WHAT"] == "HOST": + svc_desc = "_HOST_" + else: + svc_desc = context["SERVICEDESC"] + + url = "http://localhost:%d/%s/check_mk/ajax_graph_images.py?host=%s&service=%s" % \ + (get_apache_port(), os.environ["OMD_SITE"], context["HOSTNAME"], svc_desc) + + try: + json_data = urllib2.urlopen(url).read() + except Exception, e: + if opt_debug: + raise + sys.stderr.write("ERROR: %s\n" % e) + return [] + + return map(base64.b64decode, json.loads(json_data)) + + +def get_omd_config(key): + for l in file(os.environ["OMD_ROOT"] + "/etc/omd/site.conf"): + if l.startswith(key + "="): + return l.split("=")[-1].strip("'\n") + return None + + +def get_apache_port(): + port = get_omd_config("CONFIG_APACHE_TCP_PORT") + if port == None: + return 80 + else: + return int(port) + + +def use_cmk_graphs(): + return get_omd_config("CONFIG_CORE") == "cmc" + + +def render_performance_graphs(context): + if use_cmk_graphs(): + graphs = render_cmk_graphs(context) + else: + graphs = render_pnp_graphs(context) + + attachments, graph_code = [], '' + for source, graph_png in enumerate(graphs): + if context['WHAT'] == 'HOST': + svc_desc = '_HOST_' + else: + svc_desc = context['SERVICEDESC'].replace(' ', '_') + # replace forbidden windows characters < > ? " : | \ / * + for token in ["<", ">", "?", "\"", ":", "|", "\\", "/", "*"] : + svc_desc = svc_desc.replace(token, "x%s" % ord(token)) + + filename = '%s-%s-%d.png' % (context['HOSTNAME'], svc_desc, source) + + attachments.append(('img', filename, graph_png, 'inline')) + + cls = '' + if context.get('PARAMETER_NO_FLOATING_GRAPHS'): + cls = ' class="nofloat"' + graph_code += '<img src="cid:%s"%s />' % (filename, cls) + + if graph_code: + graph_code = '<tr><th colspan=2>Graphs</th></tr>' \ + '<tr class="even0"><td colspan=2 class=graphs>%s</td></tr>' % graph_code + + return attachments, graph_code + + def construct_content(context): # A list of optional information is configurable via the parameter "elements" # (new configuration style) @@ -575,48 +679,13 @@ def construct_content(context): # Prepare the mail contents content_txt, content_html = render_elements(context, elements) - # Add PNP Graph if "graph" in elements: - # Fetch graphs for this object. It first tries to detect how many sources - # are available for this object. Then it loops through all sources and - # fetches retrieves the images. If a problem occurs, it is printed to - # stderr (-> notify.log) and the graph is not added to the mail. + # Add PNP or Check_MK graph try: - num_sources = fetch_num_sources(context) - except GraphException, e: - graph_error = extract_graph_error(str(e)) - if '.xml" not found.' not in graph_error: - sys.stderr.write('Unable to fetch number of graphs: %s\n' % graph_error) - num_sources = 0 - - graph_code = '' - for source in range(0, num_sources): - try: - content = fetch_graph(context, source) - except GraphException, e: - sys.stderr.write('Unable to fetch graph: %s\n' % e) - continue - - if context['WHAT'] == 'HOST': - svc_desc = '_HOST_' - else: - svc_desc = context['SERVICEDESC'].replace(' ', '_') - # replace forbidden windows characters < > ? " : | \ / * - for token in ["<", ">", "?", "\"", ":", "|", "\\", "/", "*"] : - svc_desc = svc_desc.replace(token, "x%s" % ord(token)) - name = '%s-%s-%d.png' % (context['HOSTNAME'], svc_desc, source) - - attachments.append(('img', name, content, 'inline')) - cls = '' - if context.get('PARAMETER_NO_FLOATING_GRAPHS'): - cls = ' class="nofloat"' - graph_code += '<img src="cid:%s"%s />' % (name, cls) - - if graph_code: - content_html += ( - '<tr><th colspan=2>Graphs</th></tr>' - '<tr class="even0"><td colspan=2 class=graphs>%s</td></tr>' % graph_code - ) + attachments, graph_code = render_performance_graphs(context) + content_html += graph_code + except Exception, e: + sys.stderr.write("Failed to add graphs to mail. Continue without them. (%s)\n" % e) content_html = substitute_context(tmpl_head_html, context) + \ content_html + \ diff --git a/web/htdocs/html_mod_python.py b/web/htdocs/html_mod_python.py index 8d56604..4c49a25 100644 --- a/web/htdocs/html_mod_python.py +++ b/web/htdocs/html_mod_python.py @@ -77,6 +77,10 @@ class html_mod_python(htmllib.html): return self.req.method + def remote_ip(self): + return self.req.connection.remote_ip + + def is_ssl_request(self): return self.get_request_header('X-Forwarded-Proto') == 'https'

8 years, 6 months

Fixed possible wrong tag matching caching due to wrong cache id computation

by Lars Michelsen

Module: check_mk Branch: master Commit: 8f0a5d91bd0e74a65290f27e9f7f3762d99e950a URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=8f0a5d91bd0e74… Author: Lars Michelsen <lm(a)mathias-kettner.de> Date: Thu Mar 17 13:28:59 2016 +0100 Fixed possible wrong tag matching caching due to wrong cache id computation --- modules/check_mk.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/modules/check_mk.py b/modules/check_mk.py index 2274b0b..280ea23 100755 --- a/modules/check_mk.py +++ b/modules/check_mk.py @@ -1269,7 +1269,7 @@ g_global_caches.append('g_hosttag_taglist_cache') # New in 1.1.13: a trailing + means a prefix match def hosttags_match_taglist(hosttags, required_tags): try: - cache_id = tuple(hosttags) + tuple(required_tags) + cache_id = tuple(hosttags), tuple(required_tags) return g_hosttag_taglist_cache[cache_id] except KeyError: pass

8 years, 6 months

3271 New option in aggregation for tuning aggregation of downtimes

by Mathias Kettner

Module: check_mk Branch: master Commit: a97c2cb3822122059c362b4216658da828cf4a3d URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=a97c2cb3822122… Author: Mathias Kettner <mk(a)mathias-kettner.de> Date: Thu Mar 17 12:09:35 2016 +0100 3271 New option in aggregation for tuning aggregation of downtimes When computing the state 'in scheduled downtime' for an aggregate first all leaf nodes that are within downtime are assumed CRIT and all others OK. Then each aggregated node is assumed to be in downtime if the state is CRIT under this assumption. You can change this to WARN now by checking the new option {{Escalate downtimes based on aggregated WARN state}} in the properties of an aggregation now. The influence of this setting is especially relevant if you use aggregation functions of type count and want the downtime information also escalated in case such a node would go into WARN state. --- .werks/3271 | 17 +++++++++++++++++ ChangeLog | 1 + web/htdocs/bi.py | 37 ++++++++++++++++++++++++++----------- web/htdocs/config.py | 1 + web/plugins/wato/bi.py | 32 +++++++++++++++++++++++++++----- 5 files changed, 72 insertions(+), 16 deletions(-) diff --git a/.werks/3271 b/.werks/3271 new file mode 100644 index 0000000..671c87b --- /dev/null +++ b/.werks/3271 @@ -0,0 +1,17 @@ +Title: New option in aggregation for tuning aggregation of downtimes +Level: 1 +Component: bi +Compatible: compat +Version: 1.2.9i1 +Date: 1458212879 +Class: feature + +When computing the state 'in scheduled downtime' for an aggregate +first all leaf nodes that are within downtime are assumed CRIT and all others +OK. Then each aggregated node is assumed to be in downtime if the state +is CRIT under this assumption. You can change this to WARN now by checking +the new option {{Escalate downtimes based on aggregated WARN state}} in +the properties of an aggregation now. The influence of +this setting is especially relevant if you use aggregation functions of type count +and want the downtime information also escalated in case such a node would go into +WARN state. diff --git a/ChangeLog b/ChangeLog index 8f745cd..191c8a7 100644 --- a/ChangeLog +++ b/ChangeLog @@ -147,6 +147,7 @@ * 3253 FIX: sms: notification script sms now handles single quotes in the message in the right way BI: + * 3271 New option in aggregation for tuning aggregation of downtimes... * 3247 FIX: Do not allow multiple BI aggregates with the same title... * 3232 FIX: BI: Only querying site of host for specific host status info... * 3291 FIX: Improved error handling when searching for BI aggregations with invalid regex diff --git a/web/htdocs/bi.py b/web/htdocs/bi.py index b4c5200..be7e25f 100644 --- a/web/htdocs/bi.py +++ b/web/htdocs/bi.py @@ -316,6 +316,12 @@ def compile_forest(user, only_hosts = None, only_groups = None): if entry[0] == config.DISABLED: continue + if entry[0] == config.DT_AGGR_WARN: + downtime_aggr_warn = True + entry = entry[1:] + else: + downtime_aggr_warn = False + if entry[0] == config.HARD_STATES: use_hard_states = True entry = entry[1:] @@ -345,6 +351,7 @@ def compile_forest(user, only_hosts = None, only_groups = None): for this_entry in new_entries: remove_empty_nodes(this_entry) this_entry["use_hard_states"] = use_hard_states + this_entry["downtime_aggr_warn"] = downtime_aggr_warn new_entries = [ e for e in new_entries if len(e["nodes"]) > 0 ] @@ -1018,20 +1025,25 @@ service_nomatch_cache = set([]) # Execution of the trees. Returns a tree object reflecting # the states of all nodes def execute_tree(tree, status_info = None): - use_hard_states = tree["use_hard_states"] + aggregation_options = { + "use_hard_states" : tree["use_hard_states"], + "downtime_aggr_warn" : tree["downtime_aggr_warn"], + } + if status_info == None: required_hosts = tree["reqhosts"] status_info = get_status_info(required_hosts) - return execute_node(tree, status_info, use_hard_states) + return execute_node(tree, status_info, aggregation_options) + -def execute_node(node, status_info, use_hard_states): +def execute_node(node, status_info, aggregation_options): if node["type"] == NT_LEAF: - return execute_leaf_node(node, status_info, use_hard_states) + return execute_leaf_node(node, status_info, aggregation_options) else: - return execute_rule_node(node, status_info, use_hard_states) + return execute_rule_node(node, status_info, aggregation_options) -def execute_leaf_node(node, status_info, use_hard_states): +def execute_leaf_node(node, status_info, aggregation_options): site, host = node["host"] service = node.get("service") @@ -1065,7 +1077,7 @@ def execute_leaf_node(node, status_info, use_hard_states): if has_been_checked == 0: output = _("This service has not been checked yet") state = PENDING - if use_hard_states: + if aggregation_options["use_hard_states"]: st = hard_state else: st = state @@ -1098,7 +1110,7 @@ def execute_leaf_node(node, status_info, use_hard_states): }, None, node) else: - if use_hard_states: + if aggregation_options["use_hard_states"]: st = host_hard_state else: st = host_state @@ -1123,7 +1135,7 @@ def execute_leaf_node(node, status_info, use_hard_states): return (state, assumed_state, node) -def execute_rule_node(node, status_info, use_hard_states): +def execute_rule_node(node, status_info, aggregation_options): # get aggregation function funcspec = node["func"] parts = funcspec.split('!') @@ -1144,7 +1156,7 @@ def execute_rule_node(node, status_info, use_hard_states): ack_states = [] # Needed for computing the acknowledgement of non-OK nodes one_assumption = False for n in node["nodes"]: - result = execute_node(n, status_info, use_hard_states) # state, assumed_state, node [, subtrees] + result = execute_node(n, status_info, aggregation_options) # state, assumed_state, node [, subtrees] subtrees.append(result) # Assume items in downtime as CRIT when computing downtime state @@ -1174,7 +1186,10 @@ def execute_rule_node(node, status_info, use_hard_states): downtime_state = func(*([downtime_states] + funcargs)) host_downtime_state = func(*([host_downtime_states] + funcargs)) - state["in_downtime"] = downtime_state["state"] >= 2 + if aggregation_options["downtime_aggr_warn"]: + state["in_downtime"] = downtime_state["state"] >= 1 + else: + state["in_downtime"] = downtime_state["state"] >= 2 # Compute acknowledgedment state if state["state"] > 0: # Non-OK-State -> compute acknowledgedment diff --git a/web/htdocs/config.py b/web/htdocs/config.py index 5bc4557..19692ae 100644 --- a/web/htdocs/config.py +++ b/web/htdocs/config.py @@ -92,6 +92,7 @@ class FOREACH_SERVICE: pass class REMAINING: pass class DISABLED: pass class HARD_STATES: pass +class DT_AGGR_WARN: pass # Has to be declared here once since the functions can be assigned in # bi.py and also in multisite.mk. "Double" declarations are no problem diff --git a/web/plugins/wato/bi.py b/web/plugins/wato/bi.py index cd38c59..119126b 100644 --- a/web/plugins/wato/bi.py +++ b/web/plugins/wato/bi.py @@ -58,6 +58,7 @@ class ModeBI(WatoMode): 'REMAINING' : 'REMAINING-f41e728b-0bce-40dc-82ea-51091d034fc3', 'DISABLED' : 'DISABLED-f41e728b-0bce-40dc-82ea-51091d034fc3', 'HARD_STATES' : 'HARD_STATES-f41e728b-0bce-40dc-82ea-51091d034fc3', + 'DT_AGGR_WARN' : 'DT_AGGR_WARN-f41e728b-0bce-40dc-82ea-51091d034fc3', } self._load_config() @@ -206,6 +207,8 @@ class ModeBI(WatoMode): convaggr = conv + node if aggr["hard_states"]: convaggr = (self._bi_constants["HARD_STATES"],) + convaggr + if aggr["downtime_aggr_warn"]: + convaggr = (self._bi_constants["DT_AGGR_WARN"],) + convaggr if aggr["disabled"]: convaggr = (self._bi_constants["DISABLED"],) + convaggr return convaggr @@ -255,6 +258,12 @@ class ModeBI(WatoMode): else: disabled = False + if aggr[0] == self._bi_constants["DT_AGGR_WARN"]: + downtime_aggr_warn = True + aggr = aggr[1:] + else: + downtime_aggr_warn = False + if aggr[0] == self._bi_constants["HARD_STATES"]: hard_states = True aggr = aggr[1:] @@ -267,11 +276,12 @@ class ModeBI(WatoMode): groups = aggr[0] node = self._convert_node_from_bi(aggr[1:]) return { - "disabled" : disabled, - "hard_states" : hard_states, - "groups" : groups, - "node" : node, - "single_host" : single_host, + "disabled" : disabled, + "hard_states" : hard_states, + "downtime_aggr_warn" : downtime_aggr_warn, + "groups" : groups, + "node" : node, + "single_host" : single_host, } # Make some conversions so that the format of the @@ -630,6 +640,18 @@ class ModeBI(WatoMode): "just since one check then it's soft state is CRIT, but its hard state is still OK."), ) ), + ( "downtime_aggr_warn", + Checkbox( + title = _("Aggregation of Downtimes"), + label = _("Escalate downtimes based on aggregated WARN state"), + help = _("When computing the state 'in scheduled downtime' for an aggregate " + "first all leaf nodes that are within downtime are assumed CRIT and all others " + "OK. Then each aggregated node is assumed to be in downtime if the state " + "is CRIT under this assumption. You can change this to WARN. The influence of " + "this setting is especially relevant if you use aggregation functions of type count " + "and want the downtime information also escalated in case such a node would go into " + "WARN state."), + )), ( "single_host", Checkbox( title = _("Optimization"),

8 years, 6 months

3237 FIX brocade_vdx_status: Fixed broken discovery

by Sven Panne

Module: check_mk Branch: master Commit: 2a4df57a8ca5945f1c48ac61ca5a77cb4411a5da URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=2a4df57a8ca594… Author: Sven Panne <sp(a)mathias-kettner.de> Date: Thu Mar 17 10:49:56 2016 +0100 3237 FIX brocade_vdx_status: Fixed broken discovery --- .werks/3237 | 10 ++++++++++ ChangeLog | 1 + checks/brocade_vdx_status | 2 +- 3 files changed, 12 insertions(+), 1 deletion(-) diff --git a/.werks/3237 b/.werks/3237 new file mode 100644 index 0000000..29c0ab5 --- /dev/null +++ b/.werks/3237 @@ -0,0 +1,10 @@ +Title: brocade_vdx_status: Fixed broken discovery +Level: 1 +Component: checks +Class: fix +Compatible: compat +State: unknown +Version: 1.2.9i1 +Date: 1458208148 + + diff --git a/ChangeLog b/ChangeLog index fb13c64..8f745cd 100644 --- a/ChangeLog +++ b/ChangeLog @@ -86,6 +86,7 @@ * 3099 FIX: mk_mysql: worked around problems where running instances can't be discovered * 3009 FIX: ESX agent: fixed exception with utf-8 encoded strings... * 3010 FIX: Fixed vanished snmp subchecks for snmp hosts... + * 3237 FIX: brocade_vdx_status: Fixed broken discovery Multisite: * 3187 notification view: new filter for log command via regex diff --git a/checks/brocade_vdx_status b/checks/brocade_vdx_status index 22c7452..f550be2 100644 --- a/checks/brocade_vdx_status +++ b/checks/brocade_vdx_status @@ -58,7 +58,7 @@ check_info["brocade_vdx_status"] = { # another approach: check for existance of the first needed OID #not oid('.1.3.6.1.2.1.1.2.0').startswith( ".1.3.6.1.4.1.1588.2.1.1.1"), "snmp_scan_function" : lambda oid: oid('.1.3.6.1.2.1.1.2.0').startswith( ".1.3.6.1.4.1.1588") \ - and oid(".1.3.6.1.4.1.1588.2.1.1.1.1.6") != None, + and oid(".1.3.6.1.4.1.1588.2.1.1.1.1.6.0") != None, "snmp_info" : ( ".1.3.6.1.4.1.1588.2.1.1.1.1", [ 6, # Firmware 7 # Status

8 years, 6 months

← Newer
1
...
8
9
10
11
12
13
14
...
28
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Checkmk git commits March 2016