search

dCache / dcache

dCache - a system for storing and retrieving huge amounts of data, distributed among a large number of heterogenous server nodes, under a single virtual filesystem tree with a variety of standard access methods
https://dcache.org
277 stars 132 forks source link

In multi-core setup some components become unavailable #7519

Open kofemann opened 4 months ago

kofemann commented 4 months ago

An incorrect path selection makes message delivery impossible in setup, where one of the cores loses connection to another component. For example:

flowchart TD
    D[Door] --> c1(CORE 1)
    D[Door] --> c2(CORE 2)
    D[Door] --> c3(CORE 3)

    c1(CORE 1) --> p(pool)
    c2(CORE 2) -.x .- p(pool)
    c3(CORE 3) --> p(pool)

the message from Door to Pool sent over CORE2 will not arrive, despite alternative routes.

cfgamboa commented 2 months ago

@kofemann could you please remind me what is the status of this bug/fix deployment?

kofemann commented 2 months ago

We are still investigating.