The plant disease early warning is an essential way for protecting eco-environment and improving the level of vegetable quantity and quality safety in solar greenhouses. To realize the early warning, the cucumber disease model and decision support system were developed for precision cucumber protection in greenhouses as follow four parts. Firstly, we combined the leaf wetness sensors and estimation model based on canopy relative humidity (RH threshold model) in to the leaf wetness duration (LWD) monitoring method. The errors were around 1~2 h; compared with the LWD that was over 3 h, the monitoring effects of the method were acceptable. Secondly, we developed the primary infection situation early warning models of important diseases, such as cucumber downy mildew in solar greenhouses. The model was evaluated by over a 4-year (2006-2009) dataset in the field. The results showed that it could warn the primary infection and disease occurrence date with a probability of 95% and more than 2 days before disease appearance. Thirdly, considering the characteristics of multi-warning sources of cucumber downy mildew in solar greenhouses, the warning source traceability model for cucumber downy mildew in solar greenhouses was constructed for system ease-realization using chain-styled theory of disaster. At last, the decision support system of cucumber disease early warning in solar greenhouses was developed by Visual Studio 2005 and SQL Server 2000, which added more functions of ongoing decision making for revising cultural practices than traditional record-keeping systems. The method and system shows promise for increasing adoption for IPM in China, and can provide decision support for early warning of cucumber diseases in solar greenhouses.