Slow WU download speed

First off sorry if this is posted in the wrong section.


It seems the last two days I've gotten very slow download speeds from the server, to the point where I'm actually crunching faster than it takes to download.

These are the 8MB WUs from AstroPulse (if that makes a difference, other than size). Log says:

06/01/2013 13:36:12 | SETI@home | Temporarily failed download of ap_20oc12ac_B3_P0_00013_20130104_14697.wu: transient HTTP error
06/01/2013 13:36:12 | SETI@home | Backing off 5 hr 15 min 41 sec on download of ap_20oc12ac_B3_P0_00013_20130104_14697.wu


06/01/2013 13:40:54 | SETI@home | Started download of ap_20oc12ac_B4_P0_00072_20130104_03791.wu
06/01/2013 13:40:56 | | Project communication failed: attempting access to reference site
06/01/2013 13:40:58 | | Internet access OK - project servers may be temporarily down.



No proxy, and internet connection otherwise is okay.

Anyone else experiencing this also?
____________

Just highlight the download, and press 'Retry Now' or wait for the timer to count down, that's quite normal around here when the Hurricane link is maxed out:

Graphs for gigabitethernet2_3

Claggy

Thanks. It retires, while initial speed is 12KB/sec it just drops to a nothing after 10-15 seconds. I think average run time for WU dl is about 60 minutes.

Took a screen of it here https://dl.dropbox.com/u/11600062/boinc_transfer.jpg
____________

Thanks. It retires, while initial speed is 12KB/sec it just drops to a nothing after 10-15 seconds. I think average run time for WU dl is about 60 minutes.

Took a screen of it here https://dl.dropbox.com/u/11600062/boinc_transfer.jpg

Took a screen of it here https://dl.dropbox.com/u/11600062/boinc_transfer.jpg

If you go to the Activity Menu, and Suspend Network Activity for a moment, then resume it,

1) Boinc will resume the (mostly complete) downloads at the top of the que, 2) If DNS has switched your downloads will start on the other server, hopefully the faster one,

You should also check out <max_file_xfers_per_project> in Client Configuration. You're still using the default 2.

You should also check out <max_file_xfers_per_project> in Client Configuration. You're still using the default 2.

Thanks! Made myself a config file now.. didnt know that was possible.. hopefully this will help.. I guess I should do a readup of this config file.. seems alot of usefull settings.

Just returned to boinc and seti (still want the old client back tho).. if any other tips that would be appreciated :-)

Another trick is to run SIV, it has a BOINC Status page where you can configure it to retry downloads (But you're got to keep that page open, so i'm running two instances)

Claggy

...you can configure it to retry downloads...
My hero....

I currently have 10 APs on the transfer page. The Retry button is moaning.

I'm wondering if this is a Scandinavian problem, as I have the same issue. After a disable/enable network, the download seems speedy for a few seconds, but then quickly grinds to a halt.

Looking at the top 10 hosts, they are all getting what they need according to the increased RAC, except my hosts which are steadily crunching 0-10% of capacity due to failing downloads.

Uploads are superspeedy.

I don't have any proxy to test, so don't know where the bottleneck could be....
____________
Morten Ross

ONLY 54 minute - I haven't had an AP go through that fast for a few days, they are typically taking 90-150 minutes (excluding the time wasting, band width saturating, server clogging so called "retries" and "back-offs" - all these accomplish is to waste resources and frustrate users - and more than double the REAL download times, to more like 180-300 minutes).
____________
Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

I've noticed the prolonged downloads too, but noticed today that it's improved significantly. It's been a few months since I payed much attention to what the clients are doing, so not sure how unusual the slower DL's on my end has been.