What can cause a resistor to become non ohmic?
all resistors have a range of conditions to be ohmic. One of the most important is the temperature: heat them enough and they will not meet the design specifications (until they melt, of course).
A bulb filament is an example because it operates at such a high temperature. The resistance is therefore a function of the current, which means that there is no constant value of r.
excessive heat is one thing – for example by making the resistor work well above its dissipation index.
Rectifiers, there were the copper oxide rectifiers. You can form a diode junction from copper oxide. if this occurred somewhere in the connections of the construction of a resistor, it would become non-ohmic – at least to a certain extent and particularly visible to the weaker applied voltages.
Excessive power dissipation is the usual problem. if you have a resistance of 1 watt that burns at 2 watts, a failure (non ohmic condition) is imminent (smoke in general).
other possibilities are manufacturing defects (very rare) or a bad installation process (more likely).
In my experience, it is usually a failure of something associated with the resistor that caused excessive power dissipation or a design fault that caused too much heat.
consider a filament bulb with a tungsten filament wire. it has a small finite resistance at room temperature.
By the time you give it a supply voltage, it starts to heat up and its resistance increases. resistance continues to increase with temperature until it stabilizes at its target design value to emit light of specified intensity. the element can not be verified for the ohms law.
take a water resistance because it is heated by the current, its resistance changes. the basic condition of the ohms law is therefore not maintained.
Varistors have a resistance that varies with voltage. they are non-ohmic.
Resistances behave essentially as non-ohmic when a current or voltage causes a significant change in their values.