Anaerobic digestion is one of the most attractive and most cost-effective technologies to transform the Organic Fraction of Municipal Solid Waste (OFMSW) into energy. According to total solid contents (TS) of the waste, three main types of technologies have been developed: wet (≤ 10% TS), semi-dry (10-20% TS) and dry (≥ 20% TS) digestion processes. Dry anaerobic digestion is attractive because of the low amount of added water to the solid waste and, consequently, the reduction of the digester size. However, this technology suffers from biological and technological restrictions due to the excessive amount of solids. The aim of this study is to investigate the link between microbial community structure of solid waste anaerobic digestion and the moisture content as well as the macroscopic degradation performances of the overall process. To this end, cardboard was used as a model substrate since it represents the largest proportion of OFMSW. A TS gradient from wet to dry digestion was tested in batch series operated under mesophilic conditions: 10, 15, 20, 25, 30, 35% TS. Four replicates were performed per assay. Biogas production was analysed weekly. VFA measurements and microbial sampling were performed after 298 days of operation. Considering first the global performance, two main groups were distinguished: The first group, group A, gathering reactors operated at TS ≤ 25% and two replicates at TS = 30% (named 30a) showed a good methane production (140 to 180 ml CH4/g VS) with no VFA accumulation. A second group, group B (two replicates at 30% TS; named 30b), and reactors at 35% showed a lower methane yield (20-40 ml CH4/g VS) and a VFA accumulation (33 g VFA/kg), indicating a strong inhibition of methanogenesis. Thus, a value of 30% TS represented a threshold dryness affecting anaerobic digestion in the present operating condition. Microbial communities were studied by SSCP of the V3 region of the 16S rRNA gene for all samples to relate the macroscopic observation to the microbial structure. According to the bacterial SSCP fingerprints, group A and B were statistically distinguished using an analysis of variance. Three SSCP peaks were only observed in group B, when methanogenesis was blocked. These microorganisms corresponded therefore to potential bio-indicators of dry anaerobic digestion dysfunctions. Regarding the community of Archaea, three groups of microorganisms were statistically extracted. Group A was split in two subgroups: one single peak was dominant for semi-dry anaerobic digestion (10-15% TS), and two other for dry anaerobic digestion (20-30% TS). One single peak was dominant for group B representative of dry anaerobic digestion dysfunctions (≥ 30% TS). To conclude, this study clearly highlighted several bioindicators of anaerobic digestion malfunction. In order to test their validity, other environmental conditions are currently tested like the change in the inoculum and the surface of exchange "liquid" solid.