One pack draining quicker doesn't have to be a problem, just means that you don't use the full capacity of the higher capacity pack. Where people go wrong is that they have the voltage cutoff set too low and consistently over-discharge one of the packs, which damages it causing it's capacity to become worse every time it is cycled, becoming more severely over-discharged with every cycle until it is so degraded in capacity that it becomes noticeable..

Voltage vs discharge curve is non linear so what may see like an alarming voltage mismatch may not be a significant mismatch in capacity. For example, 3.6V/cell is ~90% discharged, 3.4V is ~95% discharged, 3.2V is ~97% and 3.0V is ~100%. So you could have one pack at 3.6V/cell and the other at 3.2V/cell and you have only 'missed out' on about 7% of your overall capacity. In that case you might find that a safe cut off setting is around 3.5 or 3.6V/cell. LiPos only become damaged below 3.0V, so as long as you aren't discharging any cells that far, no harm done, they will maintain good capacity and give you plenty of use. On the contrary, if one pack is say 3.8V/cell when the other hits 3.2V/cell, that is probably an unacceptable mismatch but this implies that the packs are wildly mismatched in capacity to begin with (one has about double the capacity of the other) - you're only going to get the capacity of the 'lower' pack, which tempts the user to set an aggressive LVC which operates the lower pack at dangerously low voltages, reducing it's life and causing it's capacity to drop even further, making it run into low voltages faster and faster every time until it is badly degraded and unusable. I.e. 3.8 and 3.2V quickly becomes 4.1V and 2.9V... the pack is being damaged but the LVC is blind to it because the higher capacity pack is maintaining voltage. If you note the cell voltages after every run, you should be able to catch this before it happens and worst case you just have to start using the packs separately or increase your cutoff voltage and accept a slight reduction in overall usable capacity.

Internal resistance does not have much effect unless you are drawing stupid currents because LiPos have extremely low resistance to begin with. The main concern is the overall capacity of the packs under normal use and as long as it is within a reasonable tolerance (say, 20% of each other), and you keep and eye on your cell voltages, series packs should be no issue.

As mentioned above, the exact same potential problems can occur within single packs, just that you're keeping all the cells together so they tend to age similarly (as the user is basically forced to only use them together) and they might be a little better matched out of the box. Most packs would be lucky to have their cells matched better than 5% out of the box anyway, so all it takes is an end user to set a too aggressive LVC and the pack will see premature death the same way that too aggressive LVC will kill series packs.