When doing radio interferometry observations, one needs to combine observations from a number of different telescopes, at different locations, to get good quality data. This of course means that there is a potential for a lot of different things going wrong, as one needs to deal with multiple sites. Some of the problems that may come up, are completely out of your hands. However, even if one is having problems with the observations for these reasons, as the good scientists we are, we investigate the problems in detail. This is to make sure that it was indeed out of our hands, and not just us overlooking some obvious step we could have taken to avoid the issue in the first place. A situation that occurs more frequently than one would imagine.  Such inspection can also allow us to at least put the blame for loss of observations in the right place. Whether this means it will not happen again, is a completely different story. A beautiful example of this is the following excerpt from a paper by N. Bartel and colleagues on radio observations of the galaxy M82:

“No data were taken at station D during the period 0830 to 1630 GST due to the presence of a red racer snake (Coluber constrictor) draped across the high-tension wires (33,000 V) serving the station. However, even though this snake, or rather a three-foot section of its remains, was caught in the act of causing an arc between the transmission lines, we do not consider it responsible for the loss of data. Rather we blame the incompetence of a red tailed hawk (Buteo borealis) who had apparently built a defective nest that fell off the top of the nearby transmission tower, casting her nestlings to the ground, along with their entire food reserve consisting of a pack rat, a kangaroo rat, and several snakes, with the exception of the above-mentioned snake who had a somewhat higher density. No comparable loss of data occurred at the other antenna sites.”

At the very least, that snake will not be interrupting the observations again in the future. If you are interested in the actual outcome of the observations, the paper can be read in full here.