Skip to content

HybridCache L1 cache getting out of sync when using multiple instances #5674

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
Rick-Anderson opened this issue Nov 20, 2024 · 2 comments
Closed
Labels
area-caching-hybrid bug This issue describes a behavior which is not expected - a bug.

Comments

@Rick-Anderson
Copy link

Description

moved from dotnet/AspNetCore.Docs#34083, created by @dukesteen

Description

The current documentation for HybridCache does not adequately warn developers about the risk of L1 cache getting out of sync when invalidating cache entries from another instance. Consider the following example:

Instance 1
 - memorycache with tenant id 1 inside
 - redis cache nothing inside
Instance 2
 - memorycache with nothing inside
 - redis cache also nothing inside

Now instance 2 does an update on the tenant and invalidates both the redis and local cache, which gives you the following situation

Instance 1
 - memorycache with tenant id 1 inside
 - redis cache also has tenant id 1 inside
Instance 2
 - memorycache with tenant id 1 inside
 - redis cache also has tenant id 1 inside

Instance 1's layer 1 cache is now out of sync, and can respond to requests with the wrong data. This can be catastrophic if instance 2 removed access for a user while instance 1 is still serving requests for this user as if they were active.

Related issue: dotnet/aspnetcore#55308

Page URL

https://learn.microsoft.com/en-us/aspnet/core/performance/caching/hybrid?view=aspnetcore-9.0

Reproduction Steps

@dukesteen please provide public GitHub link with the simplest possible app that reproduces the problem.

Expected behavior

@dukesteen please provide

Actual behavior

@dukesteen please provide

Regression?

No response

Known Workarounds

No response

Configuration

No response

Other information

No response

@jose-escalona
Copy link

I found this issue while searching for documentation confirming this theoretical problem, is this really happening? Or is this new library implementing a solution to keep L1 cache sync between multiple instances? Still not clear to me.

I think that most of us are running multiple replicas of our services today in production environment, having this mechanism to keep sync between L1 cache for each of our replicas would make a huge difference, and definitely, make this library really valuable for everybody.

@IbrahimMNada
Copy link
Contributor

it has been discussed here a little --> #5517

@github-actions github-actions bot locked and limited conversation to collaborators Mar 30, 2025
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
area-caching-hybrid bug This issue describes a behavior which is not expected - a bug.
Projects
None yet
Development

No branches or pull requests

4 participants