Keepalives

erpo41 asked for information regarding Tahoe-LAFS' use of keepalives over their TCP connections. Brian Warner pointed out Tahoe-LAFS sends a keepalive packet once every eight (8) minutes. This setting is configurable via "timeout.keepalive" as noted in "docs/configuration.rst".

Switching from Introducers to Gossip

Brian proposes switching from introducers to gossip. Brian's idea is for all nodes to offer a "grid-control" service. "grid-control" allows you to publish Announcements, either your own or ones you are forwarding for someone one else. Brian proposes several approaches to this concept:

• fully-connected mesh: every node makes a Foolscap connection to every grid-control provider they hear about, subscribe to hear about all announcements, and publish any announcements that the other side doesn't already know about.
• opportunistic: clients only connect to storage servers, and storage servers don't make outbound connections to anybody, but if you do happen to be connected to someone who also offers "grid-control", then connect to their grid-control object too and exchange Announcements
• cluster-of-Introducers: normal nodes don't offer grid-control, but multiple Introducers do, and all of them know about each other. All nodes connect to all grid-control providers (which means all Introducers).
• one Introducer: this is just a degenerate cluster-of-Introducers

grid-updates

darrob announced the release of grid-updates 1.1.3. grid-updates is Tahoe-LAFS helper program. Originally a shell script, darrob rewrote the program in Python, so it runs on all platforms supported by Tahoe-LAFS. grid-updates helps manage volunteer grids. It's core feature is downloading and uploading the node's introducer list and has a new feature of downloading lists for shares and repairing them. darrob is testing the program against 1.9.2a1. The program is packaged for Windows, Arch and Debian.