There is no "official outage" read Matt's post

Speaking of outages, this Saturday (the 27th) we will be bringing the project down for the day as the colo is messing with power lines and while they are confident we shouldn't lose power during their upgrades we're going to play it safe and make sure our databases are quiescent.

They have decided to bring the project down "just in case". The COLO expects no irruptions.

---

Well, since nothing comes out of the splitters, my GTX980 will be cold in 3 hours. And even if they start splitting full bore, and I will be able to fill the cache to the brim, it will become dead cold again, 6 hours into tomorrows outage.

100 tasks for the GTX980, only lasts about 6 hours....

Maybe I missed it, but was any indication given as to how long an outage to expect?

Matt said "we will be bringing the project down for the day". So I expect that means pretty much all day. Depending on who is doing the work, campus or the power company, they might be finished by the evening. Then someone could remote in to power everything up.
I'm thinking we won't see the servers back up before 6PM local SETI time.

I suppose so.
Oddly, there was nothing posted on the IST page about it.

Not really. There is no scheduled down time. They have chosen to take things down.
"they are confident we shouldn't lose power during their upgrades we're going to play it safe"

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Ahhhhh...
That makes sense. Thanks for the clarification.

---

"Time is simply the mechanism that keeps everything from happening all at once."

After the new ap's from today I found 8 more (7567911, 7567922, 7567938, 7568510, 7568518, 7568644,7569479, 7572925)

All 35 identical in chronological order of building:

7567886, 7567911, 7567912, 7567913, 7567922, 7567924, 7567931, 7567938, 7567941, 7567951, 7568499, 7568501, 7568503, 7568504, 7568508, 7568510, 7568511, 7568512, 7568513, 7568516, 7568518, 7568519, 7568596, 7568597, 7568637, 7568640, 7568642, 7568643, 7568644, 7568646, 7569479, 7569519, 7572867, 7572890, 7572925

Last 5 are not identical to first 30.

As for this mysterious 30 identical hosts aborting AP's seems to be in end. No new task downloaded in last "feeding frenzy". All of those hosts has "Last contact - 12 Jun 2015"

So in between 30 June - 5 July there will be 414 AP tasks resend. Most of them in 2 July.

After the new ap's from today I found 8 more (7567911, 7567922, 7567938, 7568510, 7568518, 7568644,7569479, 7572925)

All 35 identical in chronological order of building:

7567886, 7567911, 7567912, 7567913, 7567922, 7567924, 7567931, 7567938, 7567941, 7567951, 7568499, 7568501, 7568503, 7568504, 7568508, 7568510, 7568511, 7568512, 7568513, 7568516, 7568518, 7568519, 7568596, 7568597, 7568637, 7568640, 7568642, 7568643, 7568644, 7568646, 7569479, 7569519, 7572867, 7572890, 7572925

Last 5 are not identical to first 30.

As for this mysterious 30 identical hosts aborting AP's seems to be in end. No new task downloaded in last "feeding frenzy". All of those hosts has "Last contact - 12 Jun 2015"

So in between 30 June - 5 July there will be 414 AP tasks resend. Most of them in 2 July.

They must have finished testing the server farm...........

---

"Sour Grapes make a bitter Whine." <(0)>

We may be back but not without some hiccups.

Getting the dreaded Scheduler request failed: HTTP service unavailable

On well, now it's a waiting game.

We may be back but not without some hiccups.

Getting the dreaded Scheduler request failed: HTTP service unavailable

On well, now it's a waiting game.

It's often like that when first coming back up after an outage.
The servers start getting hit whilst they are still sorting themselves out.
Should sort itself in a while.

---

"Time is simply the mechanism that keeps everything from happening all at once."

Seems a bit slow here since the forums came back online.

---

Savoir-Faire is everywhere!
The T1 Trust, T1 Class 4-4-4-4 #5550, America's First HST

Data Distribution State SETI@home # Astropulse # As of*
Results ready to send 304,948 0 7h
Current result creation rate 0.1445/sec -1.0000/sec 4m

I don't think I've ever seen a negative creation rate before.

---

"Sour Grapes make a bitter Whine." <(0)>

Data Distribution State SETI@home # Astropulse # As of*
Results ready to send 304,948 0 7h
Current result creation rate 0.1445/sec -1.0000/sec 4m

I don't think I've ever seen a negative creation rate before.

I've seen it here and there for AP, but not often.
Last SSP page is more current.
The 7 hour old stats are now updated.

---

"Time is simply the mechanism that keeps everything from happening all at once."

There is a heartbeat!!

Ok, maybe just an MBs here and there, lol...

slow and steady....

Times like this I still miss not having a simple OEM tab for the Event Log.

No hoops necessary...

Times like this I still miss our dearly departed Cricket graphs......

---

"Time is simply the mechanism that keeps everything from happening all at once."

It will be fine, well before Christmas, I'm sure...

Things should start to settle a bit over the next hour, save any major conniptions.

My last couple of rigs just a short while ago got out of their Boinc 'deep sleep' backoff mode and reported in. So the rest of the rigs on the project either probably have also, or should be shortly. And the SSP shows results received starting to drop back off accordingly.

The daily stats dump completed about 15 minutes ago, so that DB thrasher is out of the way for today.

Work is starting to come back in to the kitties, albeit in hit and miss fashion.
I am about 1,300 tasks short of a full boat here. I wish I still had the Cricket graphs to monitor what the traffic in and out of the servers is actually doing. That was a real loss to us Seti server fixated types.

Kitties want their kibbles back....LOL.

---

"Time is simply the mechanism that keeps everything from happening all at once."

There's a lot of work available (220,000 & the splitters are actually producing more, almost 40/s), but what I've seen so far it appears to be mostly VLARs.
Almost all of my CPU cache is now VLAR (over 90%) & most GPU requests for work result in "Project has no tasks available" messages; then every 15-30 requests a whole bunch of WUs finally come through.

---

Grant
Darwin NT

I wish I still had the Cricket graphs to monitor what the traffic in and out of the servers is actually doing.

You and me both.
It makes it a lot easier to see just how heavy the load actually is or isn't without having to try & take a wild arse guess based on the graphs.

---

Grant
Darwin NT

Yup, 91 out of 100 CPU MB are VLARs on my 1 machine.

Moving along pretty well now....
The kitties are starting to top off the tanks in fair fashion.

Meow!

---

"Time is simply the mechanism that keeps everything from happening all at once."

Had a VLAR storm here to only 1 CPU wu is not a VLAR. And my remaining wu's are ati5-nocal. I am going to ask. What exactly is a VLAR.

---

Had a VLAR storm here to only 1 CPU wu is not a VLAR. And my remaining wu's are ati5-nocal. I am going to ask. What exactly is a VLAR.

It stands for very long and range if my memory serves me correctly.

---