The splitters are showing green, and there's plenty of data to split, but no work has been produced for several hours (2-4/s doesn't count). Another couple of hours & we'll be out of work.

Yeah, something odd going on there...been seeing it for a few hours.

AP stopped pretty much dead in it's tracks a couple of hours ago, even though splitters show on and data is there.
And the MB splitters have been dawdling as well.
No smoking guns that I can find though.

---

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

Well...
Looks like it's hit bottom.
My caches just started to run down.

My only guess is something DB related, as the languishing splitters are on several different servers.

---

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

Well...
Looks like it's hit bottom.
My caches just started to run down.

My only guess is something DB related, as the languishing splitters are on several different servers.

Ya, it's a shame. Things were running really good. Nice mix of MBs and APs and just a few shorties.

---

The splitters are showing green, and there's plenty of data to split, but no work has been produced for several hours (2-4/s doesn't count). Another couple of hours & we'll be out of work.

So, so frustrating.

My triple 670 rig is dry of GPU work; others to follow shortly.

Anyway that the earthquake shown on the daily graph
might have had something to do with the current circumstances?

Same here. Running down quickly. It's just a pity that the limits are so low and we don't have caches otherwise we could keep working whist it was sorted out ...

---

They probably were just on the edge of being VLARs. I think that next time this happens I might run "reschedule" and see if they get shunted to the cpu ... just curious to see what happens ...

With "Fred's Rescheduler"you can adjust the AR of what to count as "shorties" and VLAR's by plugging the values into the config.xml file.

Here is the one I use, as you can see, I count anything less than AR=0.25 as a VLAR. I think the official value is around AR=0.15.

config.xml file for Fred's Rescheduler.
Place in the Resched's home directory.

<config>
   <seti>
      <vlar>0.25</vlar>   
      <vhar>1.127</vhar>
   </seti>
<debug>
      <log_all_tasks>1</log_all_tasks>
      <rsc_fpops>1</rsc_fpops>
   </debug>
</config>

T.A.

MB just started flowing through again ...

---

Man, I like the random number generator the past few days.

Before, my APs were typically getting between 650-725, but the past 3 days, the average is 857. There's two in the low 700s, but most are in 800s, with three in 900 and one at 1004.88. I got a batch of high-blanked tasks last week though and I'm just now starting to chew through them, so that's got a bit to do with it. Run times aren't even that much longer than usual..not even +5%. I'm fine with that.

---

Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)

I will only say it once today:

Give me AP's. I need AP's.

The kitties are crunching up all the MBs they can to clear the decks for ya.

---

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

... and how a cache as well ... just enough to get us through the next outage would be fine ...

---

I will only say it once today:

Give me AP's. I need AP's.

The kitties are crunching up all the MBs they can to clear the decks for ya.

I'm helping to by crunching as fast as I can. I'm doing three at a time on my 660 TI they are averaging 17 minutes and 41 seconds. I'm using the latest GPU optimised app.

---

Just had several uploads backup, not to mention quite a few Scheduler requests resulting in "Couldn't connect to server" responses.
Hopefully just a passing glitch, not to be repeated.

---

Grant
Darwin NT

Thankfully the blockage on uploads only lasted a few minutes.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

\Looking at the sate of the tapes being split it won't be very long before a new batch are loaded and Sten will have his APs to chew on.

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

So, anyone got any theories about the present network traffic?
Inbound 35+Mb/s, and outbound 550+Mb/s- sustained for several hours now.
And after the outage yet another record for outbound traffic- 664M/s.

---

Grant
Darwin NT

\Looking at the sate of the tapes being split it won't be very long before a new batch are loaded and Sten will have his APs to chew on.

Looks like lotsa AP set to split now....
Hope Sten can grab enough of them to make his day.

---

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

So, anyone got any theories about the present network traffic?
Inbound 35+Mb/s, and outbound 550+Mb/s- sustained for several hours now.
And after the outage yet another record for outbound traffic- 664M/s.

AP's being thrown out as fast as they're made?

They do take up a lot of bandwidth when there is a glut of them.

Cheers.

I don't know what befuddles me more: the two steep, cliff-like drop-offs in the green part of the cricket graph, or the fact that I'm making the first post in this thread (WITH the graph looking like that, no less) in 28 hours.

I could understand the drops if we had run out of ready to send, but there are plenty of both MB and AP.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

I don't know what befuddles me more: the two steep, cliff-like drop-offs in the green part of the cricket graph, or the fact that I'm making the first post in this thread (WITH the graph looking like that, no less) in 28 hours.

I could understand the drops if we had run out of ready to send, but there are plenty of both MB and AP.

I don't understand the sharp drop in the Cricket graph either...
But, that notwithstanding, my caches have remained full almost up to the limits, so no panic here.

---

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