Tomasz Traczyk

Thanks @robertsLando. I don't have an ETA from my side either, unfortunately. I'll let you know if I start working on this to avoid duplicate effort. Can you confirm though...

> but your test doesn't reproduce this as in the test the PUBCOMP is successfully delivered in publish method. I don't think it's an issue: I think that the broker...

> Oh ok so you mean that in general he should always reply to a pubrel with a pubcomp even if there is no packet matching in store? I think...

Just to be clear - you mean a PR with a fix or a PR with steps demonstrating the problem ;)?

I've added [a PR](https://github.com/moscajs/aedes/pull/603), please have a look at your convenience. The tests are not passing atm, because it requires a [complementary change](https://github.com/tomekt-bolt/aedes-persistence/commit/eeb85d94226ed31eefaf3fad26c5bf3179bdfb7a) in `aedes-persistence`.

The failing tests are the ones I've added, which require the mentioned change in `aedes-persistence`, I can open another PR for that change, but don't know how to link it...

> Also this requires updates on mongodb and redis persistence (at least) `aedes-persistence-redis` works out of the box, as it updates the stored value with a freshly encoded packet: https://github.com/moscajs/aedes-persistence-redis/blob/9cc9864e512059ebb3add14ce7463544ad12b7fe/persistence.js#L335....

> `aedes-persistence-redis` works out of the box, as it updates the stored value with a freshly encoded packet: https://github.com/moscajs/aedes-persistence-redis/blob/9cc9864e512059ebb3add14ce7463544ad12b7fe/persistence.js#L335. OK, now I see that it probably _doesn't_ work out of...

Thanks @robertsLando for your comment. I will wait for some more feedback on the general direction of this change before investing in aligning the persistence layer implementations then.

I don't have any open PRs on persistences related to this issue. As I've mentioned above, I'd rather not invest in aligning the persistence without having a green light for...