Particularly with larger rockets, where database mass value errors get compounded due to the scale of the build, your best bet is to ignore the weights until after the rocket is complete. I'm assuming that your working with a HPR rocket because of the recovery harness materials you describe.
I say "ignore" the mass values - I know the temptation/need/obsession to sim the rocket early on is there. True, you'll get some ballpark sim estimates.
However - this is what I do and I've been quite pleased with the results.
"Build" the rocket in rocksim like you normally would. Don't mess with mass overrides for individual components, just go with whatever the database spits out.
After you've built the actual rocket, finish it (paint and all), and basically prep it for flight (minus motor,) weigh it on a scale. Then balance it horizontally and find the Cg in flight-ready configuration.
Then, go to "mass override" for the entire rocket and force in the physically measured mass and Cg values. Re-run the sims after this data is entered, and voila!
In my experience, particularly with "conventional designs" (i.e. those without tube fins, ring fins, or other configurations that are notoriously tricky for rocksim) you'll get some very good sim data to take to the field for motor/delay selection and predicted altitudes for various motors. Granted, you need to give the sim the appropriate launch conditions (a resonably close temperature and humidity, wind conditions {which I generally set as "breezy"; 8-14 mph wind})
I guess it's important to state that you need to take RockSim simulation data for what it's worth. No matter how "accurate" you set up the sim, if a given motor simulation predicts a peak altitude of say 1500', you're probably not gonna hit exactly 1500' in practice. If you go with the methodology I describe above, you'll be close enough that your flights should be successful and close enough (sometimes very close) to predicted that you'll be pleased.
