Rain
In a space habitat, as on Earth, rain would be useful for watering the plants, washing dust off paths, cleaning the air etc.. For the largest habitats imagined, it has been suggested that rain would fall from naturally forming clouds. I'm not sure about all of the ramifications of rotational gravity, light coming from odd directions etc., in such a situation. I'll restrict this bit to dealing briefly with smaller habitats where rain would be more obviously artificial.
In a toroidal or extended toroidal colony, the obvious thing is to have sprinklers in the 'ceiling' of the colony (Figure 1). These could be individually controllable to allow rain to fall wherever on the colony habitable surface it was wanted.
Figure 1. 
Please note that in this and the following diagrams, the path of
raindrops would in reality be more or less curved for an internal
observer, not only because of 'normal' ballistics but also
because of the effects of pseudogravity produced by rotation. The
degree of this additional curvature would depend on colony
diameter and rotation rate.
In a spherical or cylindrical colony, or similar, without a ceiling, the sprinklers could be located on a structure specifically set up for the purpose either in an equivalent position to that of the non-existent ceiling (Figure 2) or running down the rotational axis of the colony (Figure 3).
Figure 2. 
Figure 3. 
The latter would have the advantage of being at the zero gravity position of the colony, which should reduce the structural problems involved. If a colony had axial artificial lighting, the support structure could be shared for both functions. Both positions would, again, allow the location of rainfall to be chosen, although for the axial pipe the precision would probably be slightly less than for pipes closer to the habitable surface as the initial 'fall' would be slower and the overall distance larger. It may be necessary to propel the water out of the nozzles vigourously, both to enhance precision of location and to avoid adherence of water to the structure. The latter could still be something of a problem with this approach.
If one wished to avoid having a sky full of rain pipes, sprinklers could operate from ground level (or on buildings) (Figure 4) to give a sort of 'Hollywood' rain effect. This works fine for agriculture and horticulture on Earth, but it might be less acceptable in more populated areas. Obviously the position of rain could, again, be chosen at will.
Figure 4. 
To give rain from above without the sprinklers being above, it may be possible to inject water, from the centres of the end caps, along the rotational axis of a colony, either as 'slugs' (Figure 5) or jets, or as fine mist propelled by artificial wind. Although injected into a region of low to zero pseudogravity, it should then disperse and drift away from the axis of rotation to a point at which it would fall to the colony inner surface as rain.
Figure 5. 
In this case, the choice of site of rainfall onto the colony habitable surface would appear to be difficult. Whether it would be possible, by dextrous manipulation of such parameters as size of water body, spin on said body, speed of injection into the core and angle of injection from the axis of rotation, to predetermine where in relation to the colony surface the water body would break up and fall as rain, is beyond my capacity to say. Possibly the distance along the rotational axis would be easier to determine than which arc of that transverse section should receive rainfall. A large, vertical axis spinning, fluid-filled cylinder might give a partial simulation of the problem, with injection of coloured fluid droplets of varying size, density and viscosity at the axis, but there are probably better approaches. I'd welcome any suggestions*.
* Jack Muskett has suggested that, as an alternative to rain, raising atmospheric humidity and cooling specific areas of the inner surface would allow directed water deposition by condensation.