commit | 79c97118b6a73336e910897693fd71ba92f1dab1 | [log] [tgz] |
---|---|---|
author | Daniel Gruno <humbedooh@apache.org> | Thu Apr 25 09:53:55 2019 -0500 |
committer | Daniel Gruno <humbedooh@apache.org> | Thu Apr 25 09:53:55 2019 -0500 |
tree | f3614525a66198b96f9f61c115e34e8d4487d23d | |
parent | bf4ce84681bd2adb48930b6b0405948229eac74d [diff] |
Make scan_type optional based on ES major version This in part touches on #489 where in scrolling has changed between ES versions and needs to be accommodated properly. the scan used now is changed to a backwards compatible iterator, and thus .scroll is no longer needed by the caller (but still used internally). .scan is kept, so as to not ruin it for any user-made plugins.
Apache Pony Mail (Incubating) is a web-based mail archive browser licensed under the Apache License v/2.0 and built to scale to millions of archived messages with hundreds of requests per second.
Pony Mail allows you to browse and interact with mailing lists using Mozilla Persona or OAuth2 (Google, GitHub, Facebook etc) for authentication.
See https://lists.apache.org for a demo.
Pony Mail works in both public, private and mixed-mode, allowing you to have one unified place for all your communication, both public and private.
(Optionally see the detailed installation instructions for more information)
For a quick guide to installing Pony Mail, please see the guides for:
We'd LOVE if more people would contribute to Pony Mail! Any form of contribution is most welcome, whether it be programming, documentation, evangelism, marketing, or helping out other users.
To contribute to Pony Mail, follow these steps (also see this doc):
Pony Mail has been built for and tested with the mail archives of the Apache Software Foundation, which span more than 15 million emails sent across more than 20 years. To put things into perspective, importing all this on a modern machine (2xSSD with 64GB RAM) took around 12 hours and resulted in a performance at around 100 archive search requests per second per ES node, depending on mailing list size and available bandwidth.
This is a list of what we would love to get done: