Your model is unable to converge to a consistent support configuration for your load case 2 - your first operating case. If you press F2 during this step, the program will show the nonlinear restraints that are not converging in the latest iteration. (Your screen shot shows 5 restraints as unsettled.)
If you remove the friction from the +Y support at 490, the model will run. Looking at those results you can estimate how much friction you are "missing" here.

[Your iteration loop goes something like this...
1-friction reduces sliding, reduced sliding drops normal load, liftoff - remove Y restraint, rerun
2-without Y support there is no friction, slide is free, pipe position drops, return Y restraint, rerun
3-back to 1]

A few comments on your input:
++ These two restraints:
+X GAP=nn, mu=NN and -X GAP=nn, mu=NN
can be simplified to:
X GAP=nn, mu=NN
++ If GAP=0.00, it need not be specified
++ I question your termination at 490, there appears to be no moment restraint. Is it really a free end?