Very probably you have too many lags in your VAR model. If you lower number of lags, it should help you. In the beginning you should find out optimal number of lags. In R you can do it by selectVAR function for your data and the optimal number of lags (usually it is AIC) paste to p parameter in VAR function. This kind of thing happens especially when you have low number of observations or too big number of lags.