[checkmk-commits] Check_MK Git: check_mk: #1905 FIX: oracle_recovery_status:

Bugfix for checkpoints in the future Message-ID: &lt;552d17d2.5dAByGC3VNxJq6Zz%mk(a)mathias-kettner.de&gt; User-Agent: Heirloom mailx 12.5 6/20/10 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Module: check_mk Branch: master Commit: a2a035b3283099b51cb9fdea324b41d5d7950782 URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=a2a035b3283099… Author: Mathias Kettner &lt;mk(a)mathias-kettner.de&gt; Date: Tue Apr 14 15:36:13 2015 +0200 #1905 FIX: oracle_recovery_status: Bugfix for checkpoints in the future --- .werks/1905 | 13 +++++++++++++ ChangeLog | 1 + checkman/oracle_recovery_status | 5 ++++- checks/oracle_recovery_status | 16 +++++++++++++--- 4 files changed, 31 insertions(+), 4 deletions(-) diff --git a/.werks/1905 b/.werks/1905 new file mode 100644 index 0000000..51ac4ae --- /dev/null +++ b/.werks/1905 @@ -0,0 +1,13 @@ +Title: oracle_recovery_status: Bugfix for checkpoints in the future +Level: 1 +Component: checks +Compatible: compat +Version: 1.2.7i1 +Date: 1427882680 +Class: fix + +The check results in CRITICAL state when the last checkpoint time is in +the futute. This is possible after daylight changes or after a change of +time on the database server after the start of the instance. +This has been corrected. The check leads to a WARNING state when the last +checkpoint is in the future and the time is displayed as a result. diff --git a/ChangeLog b/ChangeLog index 9b471eb..6fe6419 100644 --- a/ChangeLog +++ b/ChangeLog @@ -315,6 +315,7 @@ * 2110 FIX: netapp_api_aggr: check did not take configured levels when using Nagios * 1954 FIX: fileinfo / fileinfo.groups: Fixed discovery function for fileinfo groups and equalize agent output of fileinfo agents... * 1904 FIX: mk_oracle: added processes check to ASM... + * 1905 FIX: oracle_recovery_status: Bugfix for checkpoints in the future... * 2111 FIX: hitachi_hnas_volume: fix for cases when size information of volumes is not available * 2190 FIX: jolokia_metrics.gc: Fixed exception in check if no warn/crit levels are defined * 2192 FIX: check_notify_count": Fix exception in PNP template in case of explit email addresses... diff --git a/checkman/oracle_recovery_status b/checkman/oracle_recovery_status index 5f42a03..bc2ffbc 100644 --- a/checkman/oracle_recovery_status +++ b/checkman/oracle_recovery_status @@ -10,7 +10,7 @@ description: of an archived redolog will move the checkpoint of the database. This is a usual way to monitor the apply of archived redologs on a Physical Standby Database without Data Guard. This is normaly needed for - Standard Edition Databases. There is an imporatnt difference between + Standard Edition Databases. There is an important difference between Primary and Standby Database. Primary will only create WARNINGs, when the level of levels is exceeded. A value for CRITICAL is only usable when the Database is in Physical Standby Mode. @@ -25,6 +25,9 @@ description: monitored. The rule describes the maximum time for an active backup mode for a datafile until a WARN or CRIT is generated. + If the last checkpoint time is in the future the check generates a WARN + and is displaying an information to check the time on the host system. + {Important Information} There is no checkpoint created if a database is in backup mode. It is expected, that the checkpoint time could also reach a configured diff --git a/checks/oracle_recovery_status b/checks/oracle_recovery_status index e98e34f..b1482e2 100644 --- a/checks/oracle_recovery_status +++ b/checks/oracle_recovery_status @@ -41,7 +41,7 @@ def inventory_oracle_recovery_status(info): def check_oracle_recovery_status(item, params, info): state = 0 offlinecount = 0 - oldest_checkpoint_age = -1 + oldest_checkpoint_age = None oldest_backup_age = -1 backup_count = 0 @@ -75,16 +75,26 @@ def check_oracle_recovery_status(item, params, info): if datafilestatus == 'ONLINE': checkpoint_age = int(checkpoint_age) - oldest_checkpoint_age = max(oldest_checkpoint_age, checkpoint_age) + if oldest_checkpoint_age is None: + oldest_checkpoint_age = checkpoint_age + else: + oldest_checkpoint_age = max(oldest_checkpoint_age, checkpoint_age) else: offlinecount += 1 if itemfound == True: infotext = "%s database" % (database_role.lower()) - if oldest_checkpoint_age == -1: + if oldest_checkpoint_age is None: infotext += ", no online datafiles found(!!)" state = 2 + + elif oldest_checkpoint_age <= -1: + # we found a negative time for last checkpoint + infotext += ", oldest checkpoint is in the future %s(!), check the time on the server" \ + % get_age_human_readable(int(oldest_checkpoint_age)*-1) + state = 1 + else: infotext += ", oldest Checkpoint %s ago" \ % (get_age_human_readable(int(oldest_checkpoint_age)))