> Eh, that is not the supported way to disable services. The start > symlinks should be changed to stop symlinks to disable a service the > supported way.
From init's point of view a service can be in one of 3 states in each runlevel: "enabled", "disabled" or "ignored". "insserv -r <service>" moves a service to the "ignored" state for all run levels. For a Linux-HA cluster I need this 3rd state. The problem is that invoke-rc.d and init handle this 3rd state in a different way. This seems inconsistent to me. > I'm not that familiar with the logic of invoke-rc.d, so I will have to > spend some time to see if I can figure out what is going on there. I > agree that invoke-rc.d should not start disabled services. What > was/is the output from 'runlevel' when you upgraded the packages? N 2 More important is the state of heartbeat on this host: It was in standby mode, i.e. the services provided by the cluster were active on another host. Maybe you can imagine that it can cause a lot of trouble if a postinst script calls invoke-rc.d during an upgrade, and an unwanted second instance of the service is started. It could corrupt a common database or filesystem, for example. Regards Harri
signature.asc
Description: OpenPGP digital signature
