I have been reading scanned books at the Internet Archive for more than ten years and have used them in some personal projects, including [1]. Like you, I have run into missing pages quite a few times, though perhaps not as frequently as you have. There are also sometimes problems with folded-over pages, blurred images, etc. If multiple scans of the same book are available, you might be able to find the missing pages in a different version.
Some of the original scanning institutions seem to have been more conscientious about the scanning process than others. The scans by the Getty Research Institute and by the Internet Archive itself, for example, seem better overall than some others. When I search for books that might be available in multiple versions, I usually add "-description:Google AND " at the beginning of the search string on the first pass to exclude Google’s scans, as they are often the worst quality.
Scanning is time-consuming and expensive when done well. Information about the Internet Archive’s scanning services is at [2] and [3]. An article about Google’s efforts is at [4].
[2] https://archive.org/scanning
[3] https://archive.org/details/partnerdocs
[4] https://www.theatlantic.com/technology/archive/2017/04/the-t...
I'm researching a historical book in my spare time, and every online scanned book I've tried to use, from multiple projects, has included at least one pair of missing pages where they've presumably turned two at once.
I really hope people aren't expecting these archives to be definitive or a replacement in case the books are lost.