Untargeted sequencing of total RNA in wastewater, known as metatranscriptomics, has recently garnered interest due to its potential for detecting outbreaks and circulation of pathogens and their variants, especially those of RNA viruses. Both the liquid and settled solids fractions of wastewater influent have been successfully used and compared as starting material for community-level surveillance of select targets such as SARS-CoV-2 using quantitative PCR and to a limited extent, targeted sequencing. We seek to complement such comparative studies using a holistic, i.e. untargeted microbiome sequencing, approach to understand the differences in microbial diversity that can be captured between the two sample types, which may inform our decision surrounding choice of sample type for future routine PCR-based surveillance programs which aim to include a broader range of targets. We present a feasibility study for using total untargeted RNA sequencing for the differentiation of microbial abundance and diversity within liquid influent and settled solids samples. Raw wastewater samples were collected from two wastewater treatment plants—City of Davis and University of California at Davis—in January 2022 and March 2022. Extraction of total nucleic acids from liquid influent and settled solids were performed using protocols optimized for PCR-based surveillance of SARS-COV-2, followed by mRNA capture and untargeted library preparation for Illumina shotgun sequencing. For data analysis, we evaluated two open access tools requiring varying skill levels: Chan Zuckerberg ID (CZID), a cloud-based analysis and visualization pipeline with a web-based interface and Kraken2, a script-based pipeline that requires knowledge of the command line interface. We found no significant differences in overall microbial richness, alpha diversity and relative abundance between influent and settled solids samples, though beta diversity analysis suggested samples within a sample group (combination of sample type and location) were most similar to each other. A greater number of phages, novel viruses, and transcribed AMR genes were reported in UCD samples compared to COD. Overall, Kraken2 and CZID were concordant in their detection of the topmost relatively abundant species across domains. The total number of viral classified reads in the overall dataset was lower than expected, which may be addressed in future studies through the optimization of sample pre-concentration steps and the use of pathogen enrichment panels to perform targeted sequencing.