ID: 5190
Title: The Livestatus command LOG works for the Nagios core now, too.
Component: Livestatus
Level: 1
Class: New feature
Version: 1.5.0i1
Previously only the CMC understood the Livestatus command LOG. This works
for the Nagios core now, too.
Note: For technical reasons the command is available under the name _LOG,
too, when the Nagios core is used, but this name is discouraged when using
Livestatus. On the other hand: If you use the Nagios command pipe, you have
to use _LOG. This tiny inconsistency is a bit unfortunate, but it is caused
by the way how Nagios handles custom commands.
ID: 4753
Title: Inactive browser windows/tabs are not updated until focused again
Component: Multisite
Level: 2
Class: New feature
Version: 1.5.0i1
This change should be a performance improvement to reduce the load of the
sites webserver.
While a Check_MK GUI page is opened in the browser several parts are updating
themselfs in the background, for example sidebar snapins, dashlets and views.
A lot of users open several instances of the GUI in different browser tabs
(or windows) and leave them inactive in the background. They open different
tabs and keep the others in background or even leave their computer running
and just lock their user session.
In such situation the browser keeps updating the page elements even when no
one looks at the information.
We have now implemented a mechanism to detect whether or not a page is
visible or hidden to the user. In later case the updates are stopped until
the user opens the page again, then the updates are immediately done.
This change should be seamless for the user in most cases because the updates
are immediately done when the user focuses the page again, but it might happen
that you see a "grayed out" page for a short time. This grayed out page
visualizes that the information is outdated. If you should see this for several
seconds there is something wrong. But you can always tell your browser to
reload the page which should resolve this.
ID: 5402
Title: Number of apache processes can now be configured via Global Settings
Component: WATO
Level: 2
Class: New feature
Version: 1.5.0i1
It is now possible to configure the apache configuration option {MaxClients} via
WATO (Global Settings > Site Management > Apache process tuning).
This option is often changed to tune the site apache to either consume less memory
(reduce the number of maximum processes) or being able to handle more users in the
GUI simultaneously.
The option can be configured globally for all sites or in the site specific globals,
depending on your needs.
The defaul value of this option is read from the existing configuration files
(etc/apache/apache.conf and etc/apache/conf.d/*.conf) and defaults to {64}.
ID: 5403
Title: RRDCached tuning can now be configured via Global Settings
Component: WATO
Level: 2
Class: New feature
Version: 1.5.0i1
It is now possible to configure the tuning settings of the rrdcached via
WATO (Global Settings > Site Management > RRDCached tuning).
This option is often changed to tune the disk IO behavious of the RRDCached. Details
about this can be found here: http://mathias-kettner.de/cms_graphing.html
The option can be configured globally for all sites or in the site specific globals,
depending on your needs.
The default values of the options are read from the existing configuration files
(etc/rrdcached.conf and etc/rrdcached.d/*.conf).
ID: 5191
Title: Fixed memory corruption in check_http.
Component: Site Management
Level: 1
Class: Bug fix
Version: 1.5.0i1
There was a bug in the password store handling, leading to memory
corruption. This has been fixed.
ID: 5306
Title: dell_compellent_folder: Fixed crash because of missing library
Component: Checks & agents
Level: 1
Class: Bug fix
Version: 1.5.0i1
In the original check plugin the connection to df.include has been
missing. This is a problem only since 1.4.0, when check plugins run in their
own namespace.
ID: 5405
Title: LDAP: Removed missleading and confusing "No persistent connection" option
Component: WATO
Level: 1
Class: Bug fix
Version: 1.5.0i1
The LDAP connector had an option to tell it that the connection to the LDAP server
should not be reused. This made the connector reconnect to the LDAP server for each
LDAP query.
Because enabling this option can result in a much longer synchronization run, the
option has now been dropped.