updaterpms - LCFG updaterpms component
This component is used to manage the installed RPMs.
The start and run methods attempt to match the installed RPMs with those listed in the machine's RPM spec file.
The install method is used at install time to install RPMs onto a target root filesystem. This method is called with a parameter specifying the target root filesystem.
A test run, useful to see which RPMs will be deleted/installed/upgraded, can be called by invoking the run method as follows :-
om updaterpms run -- -t
It is often useful to disable the RPMs update process at boot time, eg for laptops (this has a side effect; see below). To achieve this you should add the following resource definition to the relevant machine's profile source.
!updaterpms.methods mREMOVE(start)
Doing this has the side effect that RPMs which are flagged for processing only at boot time will never be processed. To force processing of these RPMs, the run method should be called as follows :-
om updaterpms run -- -f
Indicates for which methods RPM processing will be carried out. Possible methods are start and run; the default is both these values.
The file containing the list of RPMs to be installed on this machine.
The list of directories or HTTP roots, known as the RPM repositories, containing RPMs to be installed. See updaterpms(8) for more details.
Extra flags to be used for updaterpms.
Temporary directory for downloading RPMs fetched over HTTP Defaults to /var/tmp
Specifies who should be emailed if the RPM processing fails. Only the last 50 lines of the log are emailed.
If this resource is set, the RPM specifications will be translated from the temporary syntax used for coping with RPMs with the same name but with different architectures, into the normal updaterpms syntax.
Fedora3, Fedora5, Fedora6, Scientific5
Alastair Scobie <ascobie@inf.ed.ac.uk>