The aim of this paper is to outline and critically analyse the ethical dilemmas faced by criminologists tasked with online community research. Online communities and online content serve as a valuable sources of criminological knowledge about online crime and deviance as well as formal and informal norm-making and means of social control. From discussion forum texts and blogs to multimedia posts in open and closed social networking groups, from visual and video materials on Instagram, Youtube, or Tiktok to organized crime group data exchanges in publicly inaccessible communication channels, there is great diversity and variety of the contents and forms of online communication enacted by online communities. Correspondingly, research projects are different – some focusing on the content as a linguistic object, others focusing on social relations, social network structure, and its ethnographic characteristics, while many fall in between. In addition, depending on the research goals and sensitivity of the research questions, researchers may opt for active interaction or passive (and sometimes covert) observation. Therefore there is no one-size-fits-all ethical solution for approaching online communities in criminology. Based on an in-depth analysis of methodological literature, the paper suggests that online community research is largely a matter of situational ethics, wherein researchers must make situation-aware ethical decisions about several key issues. In particular, they should aim to choose and provide arguments regarding: 1) expectations of publicity or privacy in publicly accessible information; 2) the need for informed consent or absence of such need; 3) ensuring balance between anonymity and authorship attribution; 4) securing collected data; and 5) correctly assessing risks to the researched individuals and communities, and the researchers themselves.