Theoretically but there are often other things at play. I know the story is older but since about 2015 raid5 has been dead to me, mostly because at current drive sizes a raid5 rebuild takes so long your chance of a cascade failure and losing a second drive which makes it a "send to a recovery lab" risk. Anywhere you would use raid5 just do raid6.

To add to the comments of cascading failure: if a drive goes bad, another drive from the same manufacturing batch is disproportionately likely to go bad. RAID arrays are often built with drives from the same batch, since they were bought at the same time from the same vendor. This means array failures include multiple drives more often than you'd expect.

Yes, the array itself was fine; was just a dumb action on my part given how brittle the system was.

If a second drive fails after the first while rebuilding (which happens more often with larger and slower drives), the data is lost.