I'll leave My cache at 10 days as My two quads tend to burn through their respective caches rather quickly, Anything shorter and an outage will leave Me doing nothing as the way I'm setup locally doesn't favor other projects without making lots of errors. :(

---

The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's

Looks like they might have found the sweet spot.
Network traffic has been steady at around the 45Mb/s mark for eleven hours now (that's the best it's been since Feb). There isn't a Ready to Send buffer building up, but there is work available when requested (might take 2 or 3 requests) and there doesn't appear to be any major backlog building up of other processes (although the Purge figures are slowly climbing).

---

Grant
Darwin NT

Looks like they might have found the sweet spot.

It still needs more sugar for my taste.

Sun 23 Sep 10:30:24 2007|SETI@home|Deferring communication for 1 min 0 sec
Sun 23 Sep 10:30:24 2007|SETI@home|Reason: no work from project
Sun 23 Sep 10:31:55 2007|SETI@home|Deferring communication for 1 min 0 sec
Sun 23 Sep 10:31:55 2007|SETI@home|Reason: no work from project
Sun 23 Sep 10:33:00 2007|SETI@home|Deferring communication for 1 min 0 sec
Sun 23 Sep 10:33:00 2007|SETI@home|Reason: no work from project
Sun 23 Sep 10:34:05 2007|SETI@home|Deferring communication for 1 min 0 sec
Sun 23 Sep 10:34:05 2007|SETI@home|Reason: no work from project
Sun 23 Sep 10:35:11 2007|SETI@home|Deferring communication for 1 min 50 sec
Sun 23 Sep 10:35:11 2007|SETI@home|Reason: no work from project
Sun 23 Sep 10:37:11 2007|SETI@home|Deferring communication for 4 min 34 sec
Sun 23 Sep 10:37:11 2007|SETI@home|Reason: no work from project
Sun 23 Sep 10:41:52 2007|SETI@home|Deferring communication for 16 min 1 sec
Sun 23 Sep 10:41:52 2007|SETI@home|Reason: no work from project
Sun 23 Sep 10:58:12 2007|SETI@home|Deferring communication for 24 min 41 sec
Sun 23 Sep 10:58:12 2007|SETI@home|Reason: no work from project
Sun 23 Sep 11:22:31 2007|SETI@home|Deferring communication for 1 hr 55 min 57 sec
Sun 23 Sep 11:22:31 2007|SETI@home|Reason: no work from project

---

I've started receiving longer running Work Units, the network traffic's dropped right off, and the Ready to Send queue is starting to build (quite rapidly too).
Looks like Matt's directory shuffling & system tweaking has paid off- 25 hours of hammering & nothing fell over & no system-grinding-to-a-halt backlogs occured.
Hopefully they can continue their work on the servers so that when they finally make an official announcment of MultiBeam they can run more splitters without the system choking on the load.
It's looking promising.

---

Grant
Darwin NT

Daaaaaaaannnnnngggggg, Now you've jinxed it.

Daaaaaaaannnnnngggggg, Now you've jinxed it.

Very probably. If it can survive the above statement, then we know it's good.
:-)

---

Grant
Darwin NT

six splitters now. Hope the wheels don't fall off.

---

A problem?
Since this weeks backup outage, the "Ready to send" buffer hasn't bult up at all- the "Result creation rate" hasn't been any higher than around 9 results per second & now that the queue has dropped to zero it's only just moved up to 10 per second.

---

Grant
Darwin NT

8 splitters running:
Results ready to send 0 11m
Current result creation rate 12.67/sec 11m

Must be just keeping up with demand.

---

Not even I have been getting the No Work from project all day today.

---

Something's definately broken.
Network traffic is way down (after several spurts), result creation rate is less than it was this morning, and all i'm getting are "No work from project" messages.

---

Grant
Darwin NT

Something's definately broken.
Network traffic is way down (after several spurts), result creation rate is less than it was this morning, and all i'm getting are "No work from project" messages.

NWFP here as well.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

Clearly something has gone doubly wrong. It's not Friday yet, dammit!

---

Dublin, California
Team: SETI.USA

Clearly something has gone doubly wrong. It's not Friday yet, dammit!

It is, at least here in Europe.

---

Clearly something has gone doubly wrong. It's not Friday yet, dammit!

In less than 4 hours, for me, its going to be POETS day. The weekend starts soon.

P*ss
Off
Early
Tomorrows
Saturday

Clearly something has gone doubly wrong. It's not Friday yet, dammit!

It is, at least here in Europe.

Friday night here in Australia.

---

Grant
Darwin NT

Personally, I hope they've run out of these .406 Angle Range wus. That seems to be all my caches have in them, and have had in them for the last 5 days. It doesn't do alot of good collecting data on returned wus twice a day, when all you get is the same old AR. Especially, since what I need to see anything is a wide range of Angle Ranges...(on my AMD64 X2 6000, I have collected 50 samples since going to my "special" version, 47 of those are .406 AR. On my AMD64 X2 5200, all the data is .406 AR). G darn it, get that gondola moving (yes, I know...it's historical data)

Personally, I hope they've run out of these .406 Angle Range wus. That seems to be all my caches have in them, and have had in them for the last 5 days. It doesn't do alot of good collecting data on returned wus twice a day, when all you get is the same old AR. Especially, since what I need to see anything is a wide range of Angle Ranges...(on my AMD64 X2 6000, I have collected 50 samples since going to my "special" version, 47 of those are .406 AR. On my AMD64 X2 5200, all the data is .406 AR). G darn it, get that gondola moving (yes, I know...it's historical data)

That's the thing about the new, improved MB. Once they latch on to a tape with crappy WUs, they stick around a while.

---

"Freedom is just Chaos, with better lighting." Alan Dean Foster

I feel a need to post really large graphics...... If it doesn't make sense to me to invest the time to record the same "ole" AR over and over and over. I can't justify posting graphics (as much as I'd like to). I wanna chart something...gosh darn it. The market (premarket) doesn't open for 30 minutes and I'm bored. (yet, not quite bored enough to browse the Cafe for interesting topics).

Personally, I hope they've run out of these .406 Angle Range wus. That seems to be all my caches have in them, and have had in them for the last 5 days. It doesn't do alot of good collecting data on returned wus twice a day, when all you get is the same old AR. Especially, since what I need to see anything is a wide range of Angle Ranges...(on my AMD64 X2 6000, I have collected 50 samples since going to my "special" version, 47 of those are .406 AR. On my AMD64 X2 5200, all the data is .406 AR). G darn it, get that gondola moving (yes, I know...it's historical data)

I think your wish may have been granted - I've just started to pick up some shorties from 07mr07ah. That'll really help to keep those caches filled.

BTW, I've only just spotted Eric's post in the Hardware Donations thread (I think he would like someone to donate a hot spare NAS box, like, yesterday).

Basically, he's just discovered a RAID mis-configuration (as in, wrong choice of configuration) in the main WU storage. So he's got to move everything off, reconfigure the RAID, and move it all back again. It's gonna be another of those weekends, folks. But there should be some real gain at the end of the pain.