Taste and odor (T & O) issues in drinking water have become serious problems which cannot be ignored by customers. Several studies have confirmed that microbes in water can biotransform halophenols (HPs) to haloanisoles (HAs) with earthy and musty flavors via microbial O-methylation. In this paper, the formation of 2-chloroanisole (2-CA), 2,4-dichloroanisole (2,4-DCA), 2,4,6-trichloroanisole (2,4,6-TCA), 2,3,6-trichloroanisole (2,3,6-TCA) and 2,4,6-tribromoanisole (2,4,6-TBA), and the microbial variation during the microbial O-methylation were investigated in annular reactors (ARs) with three coupon materials. For precursors, 42.5% of 2-CP and 68.9% of 2,4-DCP decayed during the reaction. Among the five HAs, the formation rate constant followed an order of 2,4,6-TCA > 2-CA > 2,4,6-TBA > 2,4-DCA ~ 2,3,6-TCA, while [HA]max followed a totally opposite one. The simulated flow velocity had no significant effect (p > 0.05) on HA formation. Ductile iron (DI) AR could produce more HAs than stainless steel (SS) and polyvinyl chloride (PVC) ARs. The final HA molar concentration followed an order of 2,3,6-TCA > 2,4-DCA > 2,4,6-TBA ~ 2,4,6-TCA > 2-CA, which might be explained by multiple factors including HP's dissociation degree, halogen atom's steric hindrance and specificity of HP O-methyltransferases. During the reaction, the microbial biomass dramatically increased 6.8–9.0 times in bulk water but dropped significantly on coupon biofilms. The effect of HPs significantly changed the bacterial communities on coupon in terms of composition and diversity, and declined the relative abundance of HA-producing bacteria, while fungi and their HA-producing genus showed better resistance ability towards HPs. By using Pearson correlation analysis, a significant correlation (p = 0.0003) was found between [HA]max and initial coupon biofilm biomass. Finally, a linear relationship was established between initial total biomass and HA formation potential.