hausman vs xtoverid test

Hi there,

I'm a new user to stata so please forgive the really basic question here. I received this output when i keyed in the hausman test: "hausman cannot be used with vce(robust), vce(cluster cvar), or p-weighted data"

I have used the robust command for both the fixed effects and random effects model

FE: xtreg lnprice bar street nocondom rich regular alcohol, fe robust
RE: xtreg lnprice bar street nocondom rich regular alcohol attractive school age,
re robust

i understand that the -xtoverid - command can be used in replacement of the hausman test. But I have three questions:

A) how should i interpret the data of xtoverid

Test of overidentifying restrictions: fixed vs random effects
Cross-section time-series model: xtreg re robust cluster(id)
Sargan-Hansen statistic 125.867 Chi-sq(6) P-value = 0.0000

B) why can't i use the hausman test when heteroskedasticity is accounted for in the fe and re tests?

C) since the re test naturally assumes that the error terms is not correlated with regressors, does that mean that it assume that heteroskedasticity does not exist. Hence, the -robust- command is useless for this model?

I hope my questions are well-phrased. Thank you in advance!