From what I understand, the problem was because the second burn was when the rocket was going tail-first through the atmosphere at hypersonic speed, and much fa

From what I understand, the problem was because the second burn was when the rocket was going tail-first through the atmosphere at hypersonic speed, and much faster than a Falcon 9 booster. That's a problem because the amount of igniter fluid depends on the air pressure in the rocket chamber, and it was not possible to determine this exactly beforehand through either testing or modeling. So they took a guess and it turned out wrong, but they got lots of data on the rocket chamber pressures, and will adjust accordingly in the future.