[checkmk-commits] 3471 FIX Repair ignoring discovery errors when bogus %s is contained in service description

Module: check_mk Branch: master Commit: 0d33acc9bbdcf86357a748074b9569e0a2121083 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=0d33acc9bbdcf8… Author: Mathias Kettner <mk(a)mathias-kettner.de> Date: Wed May 11 15:34:08 2016 +0200 3471 FIX Repair ignoring discovery errors when bogus %s is contained in service description This problem often appeard after an upgrade from 1.2.6 to 1.2.8 on monitored ESX servers. --- .werks/3471 | 10 ++++++++++ ChangeLog | 1 + modules/discovery.py | 39 +++++++++++++++++++++++++++++++++------ 3 files changed, 44 insertions(+), 6 deletions(-) diff --git a/.werks/3471 b/.werks/3471 new file mode 100644 index 0000000..31bddb1 --- /dev/null +++ b/.werks/3471 @@ -0,0 +1,10 @@ +Title: Repair ignoring discovery errors when bogus %s is contained in service description +Level: 2 +Component: core +Class: fix +Compatible: compat +State: unknown +Version: 1.2.9i1 +Date: 1462973569 + +This problem often appeard after an upgrade from 1.2.6 to 1.2.8 on monitored ESX servers. diff --git a/ChangeLog b/ChangeLog index 2d6af3e..a9ed054 100644 --- a/ChangeLog +++ b/ChangeLog @@ -17,6 +17,7 @@ * 3385 FIX: Fixed not updated discovery service after grouped auto discovery... * 3416 FIX: Avoid warning in cmc.log when rescheduling non-existing discovery check... * 3437 FIX: Improved performance of "activate changes", especially in distributed environments + * 3471 FIX: Repair ignoring discovery errors when bogus %s is contained in service description... Checks & Agents: * 3183 aruba_wlc_aps: new check which monitors the provisioned accesspoints of an Aruba Network WLAN Controller diff --git a/modules/discovery.py b/modules/discovery.py index bbfd17c..ee84837 100644 --- a/modules/discovery.py +++ b/modules/discovery.py @@ -940,19 +940,28 @@ def get_node_services(hostname, ipaddress, use_caches, do_snmp_scan, on_error): # Identify clustered services for (check_type, item), (check_source, paramstring) in services.items(): - descr = service_description(hostname, check_type, item) + try: + descr = service_description(hostname, check_type, item) + except Exception, e: + if on_error == "raise": + raise + elif on_error == "warn": + sys.stderr.write("Invalid service description: %s\n" % e) + else: + continue # ignore + if hostname != host_of_clustered_service(hostname, descr): if check_source == "vanished": del services[(check_type, item)] # do not show vanished clustered services here else: services[(check_type, item)] = ("clustered_" + check_source, paramstring) - merge_manual_services(services, hostname) + merge_manual_services(services, hostname, on_error) return services # To a list of discovered services add/replace manual and active # checks and handle ignoration -def merge_manual_services(services, hostname): +def merge_manual_services(services, hostname, on_error): # Find manual checks. These can override discovered checks -> "manual" manual_items = get_check_table(hostname, skip_autochecks=True) for (check_type, item), (params, descr, deps) in manual_items.items(): @@ -978,7 +987,16 @@ def merge_manual_services(services, hostname): # Handle disabled services -> "obsolete" and "ignored" for (check_type, item), (check_source, paramstring) in services.items(): - descr = service_description(hostname, check_type, item) + try: + descr = service_description(hostname, check_type, item) + except Exception, e: + if on_error == "raise": + raise + elif on_error == "warn": + sys.stderr.write("Invalid service description: %s\n" % e) + else: + continue # ignore + if service_ignored(hostname, check_type, descr): if check_source == "vanished": new_source = "obsolete" @@ -1015,7 +1033,7 @@ def get_cluster_services(hostname, use_caches, with_snmp_scan, on_error): # In all other cases either both must be "new" or "vanished" -> let it be # Now add manual and active serivce and handle ignored services - merge_manual_services(cluster_items, hostname) + merge_manual_services(cluster_items, hostname, on_error) return cluster_items @@ -1041,7 +1059,16 @@ def get_check_preview(hostname, use_caches, do_snmp_scan, on_error): except: raise MKGeneralException("Invalid check parameter string '%s'" % paramstring) - descr = service_description(hostname, check_type, item) + try: + descr = service_description(hostname, check_type, item) + except Exception, e: + if on_error == "raise": + raise + elif on_error == "warn": + sys.stderr.write("Invalid service description: %s\n" % e) + else: + continue # ignore + global g_service_description g_service_description = descr infotype = check_type.split('.')[0]