On (04/10/09 10:18), Erik Nordmark wrote: > > Suppose the admin has setup a property for bge0 using ipadm set-prop, > and then bge0 disappears (due to DR or due a poweroff, unplugging the > card, and power on). My understanding is that the property is > persistent, hence it will remain in the system and will be applied > should bge0 reappear. > > But what does it look like to the admin while bge0 is missing? > > I think it makes sense for the property to appear in ipadm show-prop > (perhaps with some annotation such as "bge0*" instead of "bge0" to > indicate that it refers to a missing IP interface.)
right, and in the current model, "ipadm show-prop -P" will display the information pertaining to persistent information. > Note that the "missing" aspect is orthogonal to "persistent across > reboots vs. temporary" since with DR we could even have "temporary" > properties (should we decided we need to support such beasts) end up > referring to missing IP interfaces. > > Given that the property is maintained by the system and can be shown, it > would also make sense for the admin to at least be able to destroy it > (in case the admin really doesn't want it to be applied when bge0 is > plugged back in.) and probably also be able to change it. True. in fact "dladm delete-interface -P" should be the functional equivalent of "dladm create-interface -t"- both would not change the saved-config to hold the interface info, but will change the running config. > And for consistency reasons, since the system can have properties for > interfaces that have disappeared, it probably makes sense to optionally > allow the creation of properties for interfaces that does not (yet) > exist, but relying on a force flag to avoid a typo in the name of the IP > interface being interpreted as a reference to a non-existing interface. > For example: > # ipadm set-prop -p mtu=1400 -m ip bge1 > dladm: interface "bge1" does not exist; use -F to set property > for non-existent interfaces > # ipadm set-prop -F -p mtu=1400 -m ip bge1 > > > If that general approach is sensible for ipadm properties, it probably > also makes sense for IP address management, and it might make sense to > apply it to dladm as well. this part, while a neat idea, would need some careful design to be consistently available across all the adminstrative levels. Particularly, recovering from (or even reporting) badly constructed commands must be done with care.. --Sowmini _______________________________________________ networking-discuss mailing list [email protected]
