Wald test - m1 not found r(111) - model is being stored - perhaps no memory?

[Martin Imelda Borg]

Hi I'm working with a large dataset of 600,000 observations with weights I've run wald tests several time previously and have never had problems, I suspect that Stata doesn't have enough space 'to store, remember and retrieve the results' and that's why it's giving me this error I'm posting it here, just in case someone else has any other ideas

stcox var1 var2 var3, tvc(var1 var2) texp(ln(_t)) hr
estimates store m1  

stpm2 var1 var2 var3,scale(n) df(5) eform tvc(var1 var2) dtvc(3)  

estimates store m2  
test m2 m1
 //Stata error:  
m2 not found(r111)
//this is impossible as I have stored them (see above) 

//Lrtest  
lrtest m2 m1  
r(498)
LR test likely invalid for model with robust vce