Long cell protrusions, which are effectively one-dimensional, are highly dynamic subcellular structures. The lengths of many such protrusions keep fluctuating about the mean value even in the steady state. We develop here a stochastic model motivated by length fluctuations of a type of appendage of an eukaryotic cell called flagellum (also called cilium). Exploiting the techniques developed for the calculation of level-crossing statistics of random excursions of stochastic process, we have derived analytical expressions of passage times for hitting various thresholds, sojourn times of random excursions beyond the threshold and the extreme lengths attained during the lifetime of these model flagella. We identify different parameter regimes of this model flagellum that mimic those of the wildtype and mutants of a well-known flagellated cell. By analysing our model in these different parameter regimes, we demonstrate how mutation can alter the level-crossing statistics even when the steady state length remains unaffected by the same mutation. Comparison of the theoretically predicted level crossing statistics, in addition to mean and variance of the length, in the steady state with the corresponding experimental data can be used in the near future as stringent tests for the validity of the models of flagellar length control. The experimental data required for this purpose, though never reported until now, can be collected, in principle, using a method developed very recently for flagellar length fluctuations.