Is what happened to Springer's Journal "K-theory" an evidence that journals' availability is not guaranteed?

After the editorial board of Springer's "K-theory" resigned in 2007, the journal page hosted by Springer became unavailable, including all published articles.

In the related chat group, Scott Morrison explained:

K-Theory is currently unavailable because of an ongoing lawsuit between Anthony Bak (the former chief editor) and Springer. It's a complete travesty. If you're interested in more details just ask.

However, there is no way to ask for details in that chat.

Despite of the availability problem being eventually solved through various (still commercial and paywalled) sources as per @DavidMichaelRoberts answers https://mathoverflow.net/a/165357/66983 and https://mathoverflow.net/a/287034/66983, the publications did remain unavailable for several years.

What I still don't understand:

1. How Springer as professional publisher, whose duty was supposed to guarantee its journal's preservation, failed to deliver on that important premise?

2. Lawsuits do happen, but how is it conceivable that Springer did not direct sufficient resources towards resolving whatever matters prevented it from re-establishing the service?

3. Why does the Springer's search still return no results as if the journal had never existed?

4. Does Springer or other non-OA publishers give any guarantee the same will not happen in the future with other publications?