[checkmk-commits] #2718 FIX Fixed agent timeouts not resulting in a crash report anympore

Module: check_mk Branch: master Commit: ef322440e607d7f9768ede2d06db9d4bccd3f897 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=ef322440e607d7… Author: Lars Michelsen <lm(a)mathias-kettner.de> Date: Mon Nov 2 08:36:19 2015 +0100 #2718 FIX Fixed agent timeouts not resulting in a crash report anympore When an agents takes too long to respond, this results in a timeout exception which was handled as "crash". This has been changed to output a proper error message. --- .werks/2718 | 12 ++++++++++++ ChangeLog | 1 + modules/discovery.py | 16 ++++++++++++---- 3 files changed, 25 insertions(+), 4 deletions(-) diff --git a/.werks/2718 b/.werks/2718 new file mode 100644 index 0000000..e1f3ca6 --- /dev/null +++ b/.werks/2718 @@ -0,0 +1,12 @@ +Title: Fixed agent timeouts not resulting in a crash report anympore +Level: 1 +Component: core +Class: fix +Compatible: compat +State: unknown +Version: 1.2.7i4 +Date: 1446449712 + +When an agents takes too long to respond, this results in a timeout +exception which was handled as "crash". This has been changed to output +a proper error message. diff --git a/ChangeLog b/ChangeLog index 6c0b8ac..5389ea8 100644 --- a/ChangeLog +++ b/ChangeLog @@ -2,6 +2,7 @@ Core & Setup: * 2682 FIX: Fixed exception Error creating configuration: global name 'extra_host_attributes' is not defined * 2443 FIX: Fixed quoting of ! in classical nagios checks configured via WATO... + * 2718 FIX: Fixed agent timeouts not resulting in a crash report anympore... Checks & Agents: * 2434 NetApp monitoring: Cluster-Mode is now supported, changes in existing 7Mode checks... diff --git a/modules/discovery.py b/modules/discovery.py index ee09c5d..d40bfab 100644 --- a/modules/discovery.py +++ b/modules/discovery.py @@ -299,14 +299,22 @@ def check_discovery(hostname, ipaddress=None): output += "\n" except (MKSNMPError, MKAgentError), e: - output = "Discovery failed: %s" % e + output = "Discovery failed: %s\n" % e # Honor rule settings for "Status of the Check_MK service". In case of # a problem we assume a connection error here. spec = exit_code_spec(hostname) status = spec.get("connection", 1) - except SystemExit, e: - raise e + except MKCheckTimeout: + if opt_keepalive: + raise + else: + output = "Discovery failed: Timed out\n" + spec = exit_code_spec(hostname) + status = spec.get("timeout", 2) + + except SystemExit: + raise except Exception, e: output = create_crash_dump(hostname, "discovery", None, None, "Check_MK Discovery", [])\ @@ -596,7 +604,7 @@ def discover_services(hostname, check_types, use_caches, do_snmp_scan, on_error, try: for item, paramstring in discover_check_type(hostname, ipaddress, check_type, use_caches, on_error): discovered_services.append((check_type, item, paramstring)) - except (KeyboardInterrupt, MKAgentError, MKSNMPError): + except (KeyboardInterrupt, MKAgentError, MKSNMPError, MKCheckTimeout): raise except Exception, e: if opt_debug: