Hey all!

Just a short query. I'm running Seti@home and Climateprediction.net, but recently, Seti@home seems to have taken over. It runs two WU's at a time, but climateprediction.net gets pre-empted.

Is there anyway to get them to co-exist peacefully???

Thanks in advance!

---

How do you have the resource shares defined for the projects, 50/50 or something different? Can you find the short- and long-term debt figures? The raw numbers are in the client_state.xml file, or you can see them if you run BOINCView or BOINC Debt Viewer (sorry no links right this moment). It sounds like your installation thinks you need to process more SETI right now than CPDN.

MJ

---

Hi Mikey,

If Climateprediction has been up and supplying WUs to you for the past couple of weeks, with the troubles SETI has had where they couldn't supply work, then SETI is far behind in the amount of work that you told the projects they could do, via your share ratios. SETI is just making up for the lost time, I think.

If you sit back and let them work for a few more days, I think you will see them start to "take turns" again, after SETI catches up. If not, then post another query and we'll take a look for you.

Happy crunching!

---

Okay... here's some clarification, using the answers I've received so far.

Both projects have 50% resource share.(it's listed as 100)

In client state.xml, under the sections on long/short term debt, it says
(for climateprediction.net)
76023.883951
45790.595952

(for Seti@home)
0.000000
-45790.595952

...so I guess it is a debt issue, right. Seti has been offline for a while, and so Boinc is making up for lost time.

If anyone can clarify these numbers, or put units to them, please let me know. Thanks in advance.

Cool, thanks guys.

---

Okay... here's some clarification, using the answers I've received so far.


...so I guess it is a debt issue, right. Seti has been offline for a while, and so Boinc is making up for lost time.

If anyone can clarify these numbers, or put units to them, please let me know.

Yeah, you've got it pegged. The units are seconds. I'd expect SETI to have a positive long-term debt, and CPDN to be negative, wonder if you've got them ID'd backwards? In any case, over a few days they'll equalize and start to take turns more politely. 8-)

MJ

---

In any case, over a few days they'll equalize and start to take turns more politely.

Not really. I noticed that the long-term debt takes a quite long time to normalize after such outages. I personally shut down BOINC, open client_state.xml in notepad (or similar), set all containers in that file to 0.0 and restart BOINC. Works fine to me.

---

since 29-Jun-2004:

In any case, over a few days they'll equalize and start to take turns more politely.

Not really. I noticed that the long-term debt takes a quite long time to normalize after such outages. I personally shut down BOINC, open client_state.xml in notepad (or similar), set all containers in that file to 0.0 and restart BOINC. Works fine to me.

I think that the more projects you're running, the longer it takes. My slow computer only has SETI and Protein Predictor running, and after the outages was back in balance quickly. This faster one, though, since I'm running six projects including CPDN, has interesting debt values to watch.

@Mikey: After sitting on this a day, has SETI given CPDN a chance? If not, I'm starting to wonder if it's gotten suspended somewhere...

MJ

---

Hello again!

Thanks again for the top-notch assistance!

Yeah, Seti is still refusing to share. I'll leave them to work it out for a while, and if they're not getting along, I'll edit client_state.xml

as of this morning (approx 24 hours later) the debt sections read:
(for CPDN)
86400.000000
176664.767828

(for SETI)
0.000000
-176664.767828

Now if these figures are in seconds, and 86400 of them is exacly 24 hours, this means that Boinc has determined CPDN is currently 24 hours ahead and has put the brakes on, right?

Is there any significance in the fact that the long term debts are exactly equal and opposite, or is it just a coincidence??

(Apologies for the additional questions)

---

Is there any significance in the fact that the long term debts are exactly equal and opposite, or is it just a coincidence??

(Apologies for the additional questions)

Yes. They should always sum to zero.

---