My book Isc Nootan 12th reasons for the question that when a voltmeter is connected in parallel and if it had less resistance it would draw some current and potential drop across resistor to be measured will decrease thus an error.In order to avoid this voltmeter has very high resistance.What i dont get is if connect voltmeter across resistor the net resistance in circuit would decrease and thus current drawn should increase.This increase current only passes through voltmeter thus there would be no change in current flowing through resistor,then how is above reasonable?
This argument fails because the voltage drop across the parallel resistors (the one that you wanted to measure and the one of your voltmeter) have a non-linear effect on the current. So, specifically, your assertion that all of the "new" current flows through the voltmeter is wrong.
You can see this must be the case by thinking of the limit where the voltmeter has very small resistance. Then all of the current will flow the voltmeter with very little drop in voltage. But you have the same drop across your original resistor because they are in parallel. Hence adding the (badly designed) voltmeter in this thought experiment has dropped the voltage that you wanted to measure to nearly 0 V.
Any current the voltmeter draws has to come from the voltage source being measured. That changes the measured voltage slightly if it is a small current, more if it is a larger current.
In the example you quote, the voltmeter still measures the voltage across the resistor, but the voltage changes when it is disconnected.
If you want to measure the current through the resistor at the same time, you have a problem. Either some of the current you measure is not going through the resistor, because it goes through the voltmeter, or some of the voltage you measure is lost across the ammeter.
Voltmeters have the highest possible resistance, to draw as little current as possible. Ammeters have the lowest possible resistance so as to lose as little voltage as possible.