RC5 - Low # of WU per packet.

[Choralone]

Has anybody else noticed not being able to get RC5 packets with more than 8 WU from D.net lately?

I've been looking at my Pproxy logs and since the 2nd of July all my machines have been working through packets with less than 8 WU in them. I checked all the machines I have here at home and they're all setup to grab packets with 32 WU each in them.

D.net must be recycling part of the keyspace or something. It's not something to worry about, but it's making my daily logs 2 to 3 times their normal size.

 

[Sloth]

Noticed this a couple days ago. I was trying to get blocks for a machine that I have to leave for a week at a time and due to the 500 packet limit I was only able to get about 2500 units. Normally I can get closer to 15k at a time.


S.

 

[IJump]

I've noticed this for quite a while. I am not sure why, but it seems to be pretty common.

 

[sciencewhiz]

d.net is probably getting ready to exhaust a section of the keyspace. If that is true, they are filling in the holes and thats why you can't get any big packets. As soon as they open a new section, you should be able to get big packets again.

 

[teckno]

I have been getting 1 or 2 WU per packet for several weeks

 

[kmmatney]

I have set my work-fetch threshold numbers to low values on several machines. I am hoping this increases the chance of getting "good" workunits. I have a 25K WU discrepancy in my stats (and growing!) and I hope the smaller work loads help.

 

[Choralone]

At least I seem to be able to get a few larger packets in here in the last day or so. But only a few.

I'd hate to imagine the extra overhead all these small packets are causing d.net. I realize they have to issue them due to randoms and other things that break up the larger packets, but it still sucks. 🙁 Oh well.