Merge branch 'improve_monitoring_even_more' into 0.6

9 files changed, 51 insertions, 24 deletions

diff --git a/puppet/modules/site_check_mk/files/agent/local_checks/mx/check_leap_mx.sh b/puppet/modules/site_check_mk/files/agent/local_checks/mx/check_leap_mx.sh
index 2958d2ed..b8687c9a 100755
--- a/puppet/modules/site_check_mk/files/agent/local_checks/mx/check_leap_mx.sh
+++ b/puppet/modules/site_check_mk/files/agent/local_checks/mx/check_leap_mx.sh
@@ -29,5 +29,5 @@ else
   fi
 fi
 
-echo "$exitcode $CHECKNAME stale_files=$total ${STATUS[exitcode]}: $total stale files (>=$MAXAGE min) in $WATCHDIR."
+echo "${exitcode} ${CHECKNAME} stale_files=${total} ${STATUS[exitcode]}: ${total} stale files (>=${MAXAGE} min) in ${WATCHDIR}."
 
diff --git a/puppet/modules/site_check_mk/files/agent/logwatch/bigcouch.cfg b/puppet/modules/site_check_mk/files/agent/logwatch/bigcouch.cfg
index 3f6cc413..28f333b0 100644
--- a/puppet/modules/site_check_mk/files/agent/logwatch/bigcouch.cfg
+++ b/puppet/modules/site_check_mk/files/agent/logwatch/bigcouch.cfg
@@ -4,13 +4,17 @@
  I undefined - -.*201$
  I 127.0.0.1 undefined.* ok
  I 127.0.0.1 localhost:5984 .* ok
- C Uncaught error in HTTP request: {exit,
- C Uncaught error in HTTP request: {exit,normal}
+ # https://leap.se/code/issues/5246
+ I Shutting down group server
+ # ignore "Uncaught error in HTTP request: {exit, normal}" error
+ # it's suppressed in later versions of bigcouch anhow
+ # see https://leap.se/code/issues/5226
+ I Uncaught error in HTTP request: {exit,normal}
+ I Uncaught error in HTTP request: {exit,
  C Uncaught error in HTTP request: {error,
  C Response abnormally terminated: {nodedown,
  C rexi_DOWN,noproc
  C rexi_DOWN,noconnection
  C error
  C Connection attempt from disallowed node
- W Shutting down group server
  W Apache CouchDB has started
diff --git a/puppet/modules/site_check_mk/files/agent/logwatch/logwatch.cfg b/puppet/modules/site_check_mk/files/agent/logwatch/logwatch.cfg
index c4acae40..4f16d1bd 100644
--- a/puppet/modules/site_check_mk/files/agent/logwatch/logwatch.cfg
+++ b/puppet/modules/site_check_mk/files/agent/logwatch/logwatch.cfg
@@ -8,7 +8,7 @@
 /var/log/messages
 # Patterns are indented with one space are prefixed with:
 # C: Critical messages
-# W: Warning messages 
+# W: Warning messages
 # I: ignore these lines (OK)
 # The first match decided. Lines that do not match any pattern
 # are ignored
diff --git a/puppet/modules/site_check_mk/files/agent/logwatch/syslog/couchdb.cfg b/puppet/modules/site_check_mk/files/agent/logwatch/syslog/couchdb.cfg
index c92b5af7..f546135a 100644
--- a/puppet/modules/site_check_mk/files/agent/logwatch/syslog/couchdb.cfg
+++ b/puppet/modules/site_check_mk/files/agent/logwatch/syslog/couchdb.cfg
@@ -1,4 +1,7 @@
  C /usr/local/bin/couch-doc-update.*failed
  C /usr/local/bin/couch-doc-update.*ERROR
- W epmd: got partial packet only on file descriptor
+# on one-node bigcouch setups, we'll get this msg
+# a lot, so we ignore it here until we fix
+# https://leap.se/code/issues/5244
+ I epmd: got partial packet only on file descriptor
 
diff --git a/puppet/modules/site_check_mk/files/agent/logwatch/syslog/stunnel.cfg b/puppet/modules/site_check_mk/files/agent/logwatch/syslog/stunnel.cfg
index a4e428b4..eb3131f2 100644
--- a/puppet/modules/site_check_mk/files/agent/logwatch/syslog/stunnel.cfg
+++ b/puppet/modules/site_check_mk/files/agent/logwatch/syslog/stunnel.cfg
@@ -1,6 +1,9 @@
 # check for stunnel failures
- C stunnel:.*Connection refused
-# this is a temporary failure and happens very often, so we 
-# ignore it
+#
+# these are temporary failures and happen very often, so we
+# ignore them until we tuned stunnel timeouts/logging,
+# see https://leap.se/code/issues/5218
  I stunnel:.*Connection reset by peer
  I stunnel:.*Peer suddenly disconnected
+ I stunnel:.*Connection refused
+
diff --git a/puppet/modules/site_check_mk/files/agent/logwatch/syslog/tapicero.cfg b/puppet/modules/site_check_mk/files/agent/logwatch/syslog/tapicero.cfg
index 9983d27c..7f43a18c 100644
--- a/puppet/modules/site_check_mk/files/agent/logwatch/syslog/tapicero.cfg
+++ b/puppet/modules/site_check_mk/files/agent/logwatch/syslog/tapicero.cfg
@@ -3,5 +3,11 @@
 # instances, so we ignore it
 # see https://leap.se/code/issues/5168
  I tapicero.*RestClient::PreconditionFailed:
+# until we can analyze multiline patterns with
+# check_mk logwatch, we unfortunatly need ignore this
+# line too, which shows up in combination with the
+# above PreconditionFailed error.
+# see https://leap.se/code/issues/4821#note-5
+ I tapicero.*Creating database.*failed due to:
  C tapicero.*failed
  W tapicero.*Couch stream ended unexpectedly.
diff --git a/puppet/modules/site_check_mk/manifests/agent/logwatch.pp b/puppet/modules/site_check_mk/manifests/agent/logwatch.pp
index 38514af2..4c249e6f 100644
--- a/puppet/modules/site_check_mk/manifests/agent/logwatch.pp
+++ b/puppet/modules/site_check_mk/manifests/agent/logwatch.pp
@@ -4,8 +4,9 @@ class site_check_mk::agent::logwatch {
   # see https://leap.se/code/issues/5135
 
   file { '/usr/lib/check_mk_agent/plugins/mk_logwatch':
-    source => 'puppet:///modules/site_check_mk/agent/plugins/mk_logwatch.1.2.4',
-    mode   => '0755'
+    source  => 'puppet:///modules/site_check_mk/agent/plugins/mk_logwatch.1.2.4',
+    mode    => '0755',
+    require => Package['check-mk-agent']
   }
 
   # only config files that watch a distinct logfile should go in logwatch.d/
@@ -13,6 +14,7 @@ class site_check_mk::agent::logwatch {
     ensure  => directory,
     recurse => true,
     purge   => true,
+    require => Package['check-mk-server']
   }
 
   # service that share a common logfile (i.e. /var/log/syslog) need to get
diff --git a/puppet/modules/site_check_mk/manifests/agent/mrpe.pp b/puppet/modules/site_check_mk/manifests/agent/mrpe.pp
index 37df7f74..7222e4e1 100644
--- a/puppet/modules/site_check_mk/manifests/agent/mrpe.pp
+++ b/puppet/modules/site_check_mk/manifests/agent/mrpe.pp
@@ -8,7 +8,8 @@ class site_check_mk::agent::mrpe {
   }
 
   file { '/etc/check_mk/mrpe.cfg':
-    ensure => present
+    ensure  => present,
+    require => Package['check-mk-server']
   } ->
   file_line { 'Apt':
     line => 'APT    /usr/lib/nagios/plugins/check_apt',
diff --git a/puppet/modules/site_check_mk/manifests/server.pp b/puppet/modules/site_check_mk/manifests/server.pp
index 0c98cd5a..e544ef0d 100644
--- a/puppet/modules/site_check_mk/manifests/server.pp
+++ b/puppet/modules/site_check_mk/manifests/server.pp
@@ -17,33 +17,41 @@ class site_check_mk::server {
 
   # override paths to use the system check_mk rather than OMD
   class { 'check_mk::config':
-    site          => '',
-    etc_dir       => '/etc',
-    nagios_subdir => 'nagios3',
-    bin_dir       => '/usr/bin',
-    host_groups   => undef,
-    require       => Package['check-mk-server']
+    site              => '',
+    etc_dir           => '/etc',
+    nagios_subdir     => 'nagios3',
+    bin_dir           => '/usr/bin',
+    host_groups       => undef,
+    use_storedconfigs => false,
+    require           => Package['check-mk-server']
   }
 
-  Exec['check_mk-reload'] -> Service['nagios']
+  Exec['check_mk-reload'] ->
+    Exec['check_mk-refresh-inventory-daily'] ->
+    Service['nagios']
 
   file {
     '/etc/check_mk/conf.d/use_ssh.mk':
       content => template('site_check_mk/use_ssh.mk'),
-      notify  => Exec['check_mk-refresh'];
+      notify  => Exec['check_mk-refresh'],
+      require => Package['check-mk-server'];
     '/etc/check_mk/all_hosts_static':
       content => $all_hosts,
-      notify  => Exec['check_mk-refresh'];
+      notify  => Exec['check_mk-refresh'],
+      require => Package['check-mk-server'];
     '/etc/check_mk/.ssh':
-      ensure => directory;
+      ensure  => directory,
+      require => Package['check-mk-server'];
     '/etc/check_mk/.ssh/id_rsa':
       content => $seckey,
       owner   => 'nagios',
-      mode    => '0600';
+      mode    => '0600',
+      require => Package['check-mk-server'];
     '/etc/check_mk/.ssh/id_rsa.pub':
       content => "${type} ${pubkey} monitor",
       owner   => 'nagios',
-      mode    => '0644';
+      mode    => '0644',
+      require => Package['check-mk-server'];
     # check_icmp must be suid root or called by sudo
     # see https://leap.se/code/issues/5171
     '/usr/lib/nagios/plugins/check_icmp':