Suggest to clear cookies on "Error: server is down."

Multiple institutions reported the issue in the thread in #folio-implementers Slack channel on 2024-10-03: https://folio-project.slack.com/archives/CAM25LTGV/p1727982997215719

I wish I could add more details but in our case it just started happening right after the upgrade to Quesnelia with staff doing nothing but starting the Chrome browser in the morning and attempting to open Folio. The night before they log out from Folio and close the browser. The only way to get rid of the error message is to clear the cookies and cache or just use the Incognito mode, which they do now. Not all staff gets that error message.

@Olga Kalachinskaya , @Julian Ladisch : do you know what circumstances are leading users to see these messages? There were no cookies in Poppy, so it’s surprising to me that clearing cookies would fix a problem that is caused by a P → Q upgrade.

https://folio-org.atlassian.net/browse/STCOR-896 appears closely related (er, exactly the same as) the behavior described here, however, and I am keenly interested in any details you can provide. I have found a few ways to improve how stripes-core handles the P → Q upgrade, but I have not yet been able to replicate the “server is down” problem, or to repeat if after a reload. I would be grateful for any details you can offer about what steps lead up to this error message appearing on screen, and what information, if any, appears in an error message or stacktrace on the JS console.