Anybody else having....

Slagathor · Post by **Slagathor** » Thu Jun 17, 2004 5:07 pm

Issues getting WU's??????
This is with the original Seti, not BIONC.....

The last few days I have come home on lunch break and it's having transfer errors........

BCN · Post by **BCN** » Thu Jun 17, 2004 5:59 pm

Yeah, it seems so, but as i use client version seti driver and setispy, so still i have enough units to complete, but yes, sometimes I find few units pending for transfer for some time, quite some time.

hyperspace · Post by **hyperspace** » Thu Jun 17, 2004 8:47 pm

FWIW,

I just got two new WU's.

Billl · Post by **Billl** » Fri Jun 18, 2004 4:17 am

Yeah I'm having big time problems too. Can't get WU's either. Fortunately I keep a good supply on hand. But if they don't solve this pretty soon I'm gonna be out of them.

Billl

Billl · Post by **Billl** » Fri Jun 18, 2004 4:18 am

Yeah I'm having big time problems too. Can't get WU's either. Fortunately I keep a good supply on hand. But if they don't solve this pretty soon I'm gonna be out of them.

Billl

BCN · Post by **BCN** » Fri Jun 18, 2004 8:43 am

well, with cache of WUs that's Ok, but recently I keep noticing that I have 1-2 WUs pending sending the results or the same amount getting new, but then I get normal sometimes...

anyway I think I have 3 days of work stored on each PC

Dave Rave · Post by **Dave Rave** » Fri Jun 18, 2004 11:22 am

best bet average is 3-7 days
using a good old setiqueue

BCN · Post by **BCN** » Fri Jun 18, 2004 12:11 pm

oh by the way when you use caching clients like setique or setidrive, do they crunch the oldest WUs first? it should be logical, but who knows how they do that...

Billl · Post by **Billl** » Fri Jun 18, 2004 2:09 pm

BCN wrote:oh by the way when you use caching clients like setique or setidrive, do they crunch the oldest WUs first? it should be logical, but who knows how they do that...

Yes they do crunch oldest first. My question is how big should the cashe be? Some of my machines I have as much as 10 days others I have a lot less. And of course my newest fastest machine is almost out

Billl

Dave Rave · Post by **Dave Rave** » Fri Jun 18, 2004 10:12 pm

if you use setiq, and no cache on your machines, you set it for x to y days and it will store y times RPD and you don't have to play with it if oyu add more machines. it will figure out that you need more.

setiq will hand out fifo.
setihide i think does it that way, but i also think it looks at the date of the sah tape splitter being used and does the oldest real time wu, not necessarily the oldest you downloaded.
i think, but not sure sure.

kuun · Post by **kuun** » Mon Jun 21, 2004 12:06 am

i use seti hide

i've noticed problems lately as well

but i keep about 30=99 wu's cached

so i won't run out for awhile

Dave Rave · Post by **Dave Rave** » Mon Jun 21, 2004 12:09 am

setihide is ok for one machine
but when you get more than one, you have to manage each machine
hence setiq

and speaking of vicious, (but the wrong thread) could we use your power cord for rope, save on excess freight charges.

Billl · Post by **Billl** » Mon Jun 21, 2004 9:03 am

kuun wrote:i use seti hide

i've noticed problems lately as well

but i keep about 30=99 wu's cached

so i won't run out for awhile

Your probably keeping too many on hand given your usual output. If you keep too many and you don't send them back soon enough your wasting your time even crunching them. After so long it's assumed they were lost and reissued to another user. At least this was always my understanding? I think probably 10 days is the max you should hold any work unit. I use SETIDriver to manage mine on the individual machines. To me it just seems easier. I got caught earlier last week though with the newest machines with a lot smaller cashe then I normally maintain. All I do is take the average output per day on that machine and multiply by 10 to set the cashe size. It's worked well so far for me. Of course one never knows for sure since we never get any feed back from the project. Anyone have any thoughts?

Billl

Slagathor · Post by **Slagathor** » Mon Jun 21, 2004 4:41 pm

I raised the cach of WU's up to 10 on each of the AMD's.
And there all out......

I figured 10 would be enough to get a few back to work on......HA!!!!!!!!!!!!

my machines havent sat this idel since I put them together. There gonna think somethings wrong.....heh