The sequence of events raises questions about whether the launch should have gone ahead in spite of the leak.

Dr Adam Baker, who is head of Rocket Engineering, a UK company specialising in rocket propulsion systems, says he understands why the launch took place but says it would have been better to have got to the bottom of the cause of the leak and repaired it.

“There is a risk of trying to get things too perfect and ending up taking too long and it being too expensive and as a result, public and political support disappears,” he said.

“But what I feel is that they may not have sufficiently considered the worsening of the leak after launch. This is something that Nasa and Boeing probably should have done.”

That would have been tremendously expensive, as it would have involved taking the rocket off the launch pad and taking the propulsion system out of the spacecraft.

Another issue for Nasa’s review is why these issues were not identified in any of the two previous uncrewed flight tests of Starliner, according to Dr Simeon Barber, who is a space scientist at the Open University.

“The problems we have seen in the last few weeks are not the kind we would have anticipated at this stage of the development programme for Starliner,” he says.

“The whole point about this was to test what putting astronauts in the loop of controlling the spacecraft would do in terms of performance. Instead, we seem to be dealing with rather more fundamental issues that really should have been ironed out by now.”