El 3/1/25 a las 10:09, Thomas Goirand escribió:
I was about to apply your patch, though reading it, I realized that even if the compress fails 5 times, the postinst will continue to execute instead of returning an error. That's not what we want... So I cannot apply the patch as-is.
That's great news indeed. Thanks! Does that mean that if I modify the patch so that it exits with error if the compress fails 5 times in a row then the patch will be ok for you? I'll start to work on that right now, I don't think it would be too difficult. (In either case, bear in mind that according to the tests I made, the failures are independent, so five failures in a row is extremely unlikely, like 0.05 to the power of 5). (BTW: I'll tell you later about some ideas I had to properly diagnose this in case we would still want to). Thanks.
