Skip to content

Full varnish cache invalidation is prevented by hardcoded backlog=200Β #11

@avoinea

Description

@avoinea

Is there any good reason we have Queue maxsize hard-coded to 200?!

def __init__(self, factory=Connection, timeout=30, backlog=200,

self.queues[key] = queue_ = queue.Queue(self.backlog)

This limit is easily reached if you have a lot of related-items to invalidate, thus Varnish cache will be partially invalidated resulting in inconsistent content.

I think this should be set by default to 0 (infinity)

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions