This repository has been archived by the owner on Jan 21, 2022. It is now read-only.
-
-
Notifications
You must be signed in to change notification settings - Fork 344
Implement garbage collector for sockets #38
Labels
Comments
@ngosang , please update issue title to respect content. Mention something like garbage collector for sockets. |
Done |
I think there is a related stack trace:
|
boramalper
added a commit
that referenced
this issue
Apr 25, 2017
magneticod: * Stalled DisposablePeers are shut down after 120 seconds (fixes #47 and #38) * Improved performance, thanks to @ngosang, @ad-m, and all other contributors (fixes #48) * Default logging level is now set to INFO (since many users would freak out after seeing "peer failed" messages...) magneticow: * Search speed improved A LOT, like "A L O T"! (fixes #8) (for the curious, the problem was that magneticow initialised database *every* *single* *time* a request is made because "the global application context" (`flask.g`) didn't work for some reason I don't know. Now it's blazing fast! * A bit of logging added to assist the user. This is especially important as magneticow now takes a bit of time to initialize itself (as reasonably expected) and will not be able to handle any requests until it's complete. * A faster but possibly less accurate (and not-guaranteed behaviour dependent) calculation of the number of torrents. * Thousands separator for the torrent count in the homepage.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Peer should be destroyed after socket timeout or after several seconds without receiving data.
Now you are only destroying the peer if metadata is received
https://github.com/boramalper/magnetico/blob/master/magneticod/magneticod/__main__.py#L121
in other cases, number of peers (and sockets) increases forever until you reach OS limit and process crash.
The text was updated successfully, but these errors were encountered: