Adding upstream version 1.2.2.upstream/1.2.2

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

1 files changed, 65 insertions, 0 deletions

diff --git a/doc/03-Configuration.md b/doc/03-Configuration.md
new file mode 100644
index 0000000..0b2a38d
--- /dev/null
+++ b/doc/03-Configuration.md
@@ -0,0 +1,65 @@
+# <a id="Configuration"></a>Configuration
+
+## Basics
+
+Open up the Icinga Web frontend and navigate to:
+
+`Configuration > Modules > graphite > Backend`
+
+Enter the Graphite Web URL. (e.g. `https://192.0.2.42:8003/`)
+
+The HTTP basic authentication credentials are only required
+if your Graphite Web is protected by such a mechanism.
+
+## Advanced
+
+Open up the Icinga Web frontend and navigate to:
+
+`Configuration > Modules > graphite > Advanced`
+
+### UI
+
+The settings *Default time range* and *Default time range unit* set the default
+time range for displayed graphs both in the graphs lists and in monitored
+objects' detail views.
+
+If you'd like to suppress the *No graphs found* messages, check *Disable "no
+graphs found"*. (This may cause unexpected blank pages in the graphs lists.)
+
+### Icinga 2 (Core)
+
+The settings *Host name template* and *Service name template* both are only
+required if you are using a different naming schema than the default Icinga 2
+is using. (As outlined [here](https://www.icinga.com/docs/icinga2/latest/doc/14-features/#current-graphite-schema))
+
+The setting *Obscured check command custom variable* is only required if there
+are wrapped check commands (see below) and the "actual" check command is stored
+in another custom variable than `check_command`.
+
+## Wrapped check commands
+
+If a monitored object is checked remotely and not via an Icinga 2 agent, but
+e.g. by check_by_ssh or check_nrpe, the monitored object's effective check
+command becomes by_ssh or nrpe respectively. This breaks the respective
+monitored objects' graphs as graph templates are applied to monitored objects
+via their check commands. (They fall back to the default template.)
+
+To make the respective graphs working as expected you have to tell the
+monitored object's "actual" check command by setting its custom variable
+`check_command`, e.g.:
+
+```
+apply Service "by_ssh-disk" {
+  import "generic-service"
+  check_command = "by_ssh"
+  vars.by_ssh_address = "192.0.2.1"
+  vars.by_ssh_command = "/usr/lib64/nagios/plugins/check_disk -w 20 -c 10"
+  vars.check_command = "disk"    // <= HERE
+  assign where host.name == NodeName
+}
+```
+
+## Further reading
+
+* [Templates](04-Templates.md)
+* [Troubleshooting](05-Troubleshooting.md)