Deluge Forum

Hi guys,

I am using Deluge 1.3.15-8 on a Synology DS214 NAS. Now granted this is not the fastest box out there - however, once I get to 50 or more torrents, the web UI or the client just cannot cope. It is slow to update/refresh, add new torrents, force re-check, move files etc. The delay is of the order of a few minutes to 30 minutes to never. At the moment I've got about 600 torrents. Rebooting the NAS seems to improve things for a few minutes. Just shutting down and restarting the daemon does not.

This makes using Deluge all but impossible which is a shame as otherwise I've been very happy with it.

It looks like the client/UI is not able to interrupt the daemon sufficiently frequently but I know very little about the two communicate (though I'm willing to learn )

Any ideas?

(I have a little programming experience - if anyone could give me some pointers as to where the problem might lie I might even have a bash at fixing it myself though it's possible it could be beyond my abilities or available time.)

many thanks

The issue is probably related to the low CPU performance in that NAS and the way Deluge saves the states of all torrents. When they're saved nothing else can run as there's only one thread. This save state happens every 10 minutes or so and if it does not finish within that time you could end up in a loop where it saves over and over again. At least that's what I've seen.

As I understand this will probably be fixed in the 2.0 release.

Oh, OK, thanks for that.

Any idea when 2.0 will be released (even beta)?

Hmmm - an additional find.

Shutting down the daemon within Synology shows the daemon as stopped - but examining the processes shows that it is still running and consuming as much CPU (around 50%) as it was before the shut down.

I dont recommend running deluge ATM with more than 300 torrents on a NAS, it wont work properly.

It seems to work (reasonably) fine in terms of seeding - but if you want to add or edit any torrent details, usually the only choice is to shut down the daemon and restart - you then have a few minutes window in which to work before it becomes unresponsive again.