Caches pretty much back to full now. Amazing how much WU demand has grown, if my perception is correct. Right, the splitters churning out 55+/sec is not enough to keep up with demand. That's huge ...

That Einstein does not have any GPU data could very well explain it.

---

Some people's caches are huge...

My GPU caches are huge, the project could probably be down for close to a day before those run dry. But on my bigger crunchers, the first of the CPU cores seem to run out of work about an hour or two before the average outage completes, and progressively more and more of them idle out, until they are all just loafing around. Whatta bunch of lazy so and so's! ;-)

---

Seems like you could use a rescheduler option that would move GPU tasks to your CPUs on an as needed basis when the CPUs run out. Something like that could perhaps keep them all busy through the outage.

I am using I believe the latest version of the rescheduler, running every 30 minutes, is this the one that you are referring to? Also, was it just me or was the website just down?

---

I am using I believe the latest version of the rescheduler, running every 30 minutes, is this the one that you are referring to? Also, was it just me or was the website just down?

Yeah, web site has been struggling.

The rescheduler we have doesn't address a lot of options we would all like to see, I think. This is one of them.
No knock on anyone who's written such is intended!
If you want to go into more depth, let's move this to the Rescheduling queue ...

---

I am using I believe the latest version of the rescheduler, running every 30 minutes, is this the one that you are referring to?

Sorry, I was just mentioning that such an option would be one you could use, if it existed. I'm guessing that efMer's old Rescheduler, which has been mentioned several times recently in other threads, had such capability, but I've never tried using it myself.

Also, was it just me or was the website just down?

I think it's been up and down a couple times today. Besides that, the Replica DB is currently (a/o 6 Oct 2016, 2:00:05 UTC) 85,743 seconds behind the Master, which does not bode well for the task detail pages. It does appear to be catching up, though.

Looks like the Replica is down again ...

---

Looks like the Replica is down again ...

Well, if we're lucky, perhaps it finally got caught up and now they're backing it up. Probably just wishful thinking, though.

Looks like the Replica is down again ...

Well, if we're lucky, perhaps it finally got caught up and now they're backing it up. Probably just wishful thinking, though.

Nope, it did not. I was about 40.000 sec behind master, then went offline...

Looks like the Replica is down again ...

Well, if we're lucky, perhaps it finally got caught up and now they're backing it up. Probably just wishful thinking, though.

Nope, it did not. I was about 40.000 sec behind master, then went offline...

I saw 53,000 sec behind master then it went offline.... And its still offline, maybe they will switch the web to the master, one would hope.....

come on replica! Breathe! (Need data for some code polishing :D)

---

"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.

come on replica! Breathe! (Need data for some code polishing :D)

Would be nice if someone in Berkeley at least let us know if they're working on it and/or what the issue might be. Then again, communicating useful information to the great unwashed masses is not their strong suit.

come on replica! Breathe! (Need data for some code polishing :D)

Would be nice if someone in Berkeley at least let us know if they're working on it and/or what the issue might be. Then again, communicating useful information to the great unwashed masses is not their strong suit.

With Eric ill plus the longstanding manpower shortage it is quite understandable.
I really, really miss the replica.

---

come on replica! Breathe! (Need data for some code polishing :D)

Would be nice if someone in Berkeley at least let us know if they're working on it and/or what the issue might be. Then again, communicating useful information to the great unwashed masses is not their strong suit.

I have a shower every time it rains.

---

Member of the People Encouraging Niceness In Society club.

Starting to be a problem with more than stats. I have a box I changed venues on from "AP only" to "all tasks" almost 24 hours ago, still have not been able to get that change to stick. Web acknowledges the change, says its happened, but next web query doesn't reflect it. So, no new work and machine is now empty.

---

Starting to be a problem with more than stats. I have a box I changed venues on from "AP only" to "all tasks" almost 24 hours ago, still have not been able to get that change to stick. Web acknowledges the change, says its happened, but next web query doesn't reflect it. So, no new work and machine is now empty.

Edit preferences is still working but it will effect all computers that use that location.

---

Kevin

Starting to be a problem with more than stats. I have a box I changed venues on from "AP only" to "all tasks" almost 24 hours ago, still have not been able to get that change to stick. Web acknowledges the change, says its happened, but next web query doesn't reflect it. So, no new work and machine is now empty.

Edit preferences is still working but it will effect all computers that use that location.

Turns out it was a client problem not related to the servers. Something was stuck, but a reboot solved it. Sorry for the mis-info ...

---

come on replica! Breathe! (Need data for some code polishing :D)

Would be nice if someone in Berkeley at least let us know if they're working on it and/or what the issue might be. Then again, communicating useful information to the great unwashed masses is not their strong suit.

With Eric ill plus the longstanding manpower shortage it is quite understandable.
I really, really miss the replica.

Is Eric the only one who can add a message?
Has nobody had even 30 seconds during the last several days to at least acknowledge the issue?
Maybe they've found ET, decided they're not helpful, job done, shut down commenced, all gone home!

For a project so reliant on 'us' it sometimes appears that we are not deemed very important.

---

come on replica! Breathe! (Need data for some code polishing :D)

Would be nice if someone in Berkeley at least let us know if they're working on it and/or what the issue might be. Then again, communicating useful information to the great unwashed masses is not their strong suit.

With Eric ill plus the longstanding manpower shortage it is quite understandable.
I really, really miss the replica.

Is Eric the only one who can add a message?
Has nobody had even 30 seconds during the last several days to at least acknowledge the issue?
Maybe they've found ET, decided they're not helpful, job done, shut down commenced, all gone home!

For a project so reliant on 'us' it sometimes appears that we are not deemed very important.

Unfortunately its more about funding, and at the moment Eric is ill

come on replica! Breathe! (Need data for some code polishing :D)

Unfortunately its more about funding, and at the moment Eric is ill

. . I hope he is not terribly ill and I wish him a speedy recovery.

. . Then when he is better we can hassle him about our concerns :)

Stephen

.
{ not that that will offer much incentive to get better :) }

.