We investigate here populations of cataclysmic variables (CVs) in a set of 288 globular cluster (GC) models evolved with the MOCCA code. This is by far the largest sample of GC models ever analysed with respect to CVs. Contrary to what has been argued for a long time, we found that dynamical destruction of primordial CV progenitors is much stronger in GCs than dynamical formation of CVs, and that dynamically formed CVs and CVs formed under no/weak influence of dynamics have similar white dwarf mass distributions. In addition, we found that, on average, the detectable CV population is predominantly composed of CVs formed via typical common envelope phase (CEP) ($\gtrsim70$ per cent), that only $\approx2-4$ per cent of all CVs in a GC is likely to be detectable, and that core-collapsed models tend to have higher fractions of bright CVs than non-core-collapsed ones. We also consistently show, for the first time, that the properties of bright and faint CVs can be understood by means of the pre-CV and CV formation rates, their properties at their formation times and cluster half-mass relaxation times. Finally, we show that models following the initial binary population proposed by Kroupa and set with low CEP efficiency better reproduce the observed amount of CVs and CV candidates in NGC 6397, NGC 6752 and 47 Tuc. To progress with comparisons, the essential next step is to properly characterize the candidates as CVs (e.g. by obtaining orbital periods and mass ratios).

Potentially interesting gravitational-wave candidates (outliers) from the blind all-sky searches have to be confirmed or rejected by studying their origin and precisely estimating their parameters. We present the design and first results for the followup procedure of the {\tt Polgraw} all-sky search pipeline: a coherent search for almost-monochromatic gravitational-wave signals in several-day long time segments using the $F$-statistic method followed by the coincidences between the candidate signals. Approximate parameters resulting in these two initial steps are improved in the final followup step, in which the signals from detectors are studied separately, together with the network combination of them, and the true parameters and signal-to-noise values are established.