Skip to content

Upgrade from shlink 3.7.4 to 4.1.1 fails #2143

Answered by acelaya
nellen asked this question in Help wanted
Discussion options

You must be logged in to vote

Actually, I just remembered about something.

When not using redis, Shlink uses an in-memory cache storage which is shared by all Shlink instances running in the same server. This can be a problem, so Shlink 3.7 introduced a way to define a unique cache prefix per instance, via CACHE_NAMESPACE config option or env var (in your config the value is "Shlink").

What's happening here is probably that your newest Shlink 4 instance is "picking-up" class metadata entries from the cache that were generated by Shlink 3.

Try setting a unique value there, and that should solve the problem.

See #1862 for context.

Replies: 1 comment 5 replies

Comment options

You must be logged in to vote
5 replies
@nellen
Comment options

@acelaya
Comment options

@acelaya
Comment options

Answer selected by nellen
@nellen
Comment options

@nellen
Comment options

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
2 participants