Network traffic has dropped right off & there is plenty of work ready to send, so i guess almost everyone has finally hit the task limit.

Good, maybe they'll lift the limit so we can build a cache.

Doesn't look like it...
It would appear that they may have increased the limit...say, from 20 to maybe 50 or something, but folks are still reporting hitting the in progress limit.
So they haven't opened up the spigot all the way just yet.
Maybe tomorrow?

---

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

Has anyone here generated any -12 errors? I have, 6 in the last 7 days... Out of thousands of good results...

I had 11 -12 errors with the 1200+ uploaded today. That's really unusual for my box, 1 or 2 errors a day in about 400 WU's is normal. And usually there caused by running out of memory (Because I forgot to snooze SETI when watching video's or playing a game) and not -12 errors.

Network traffic has dropped right off & there is plenty of work ready to send, so i guess almost everyone has finally hit the task limit.

Good, maybe they'll lift the limit so we can build a cache.

Doesn't look like it...
It would appear that they may have increased the limit...say, from 20 to maybe 50 or something, but folks are still reporting hitting the in progress limit.
So they haven't opened up the spigot all the way just yet.
Maybe tomorrow?

Hopefully. I just hope they can do it remotely because nobody will come in to work on the weekend. Especially a holiday weekend.

Hopefully. I just hope they can do it remotely because nobody will come in to work on the weekend. Especially a holiday weekend.

Bull#h!t.
They've done that on many occasions when the system hasn't been able to be restarted remotely, inspite of attitudes such as yours.

---

Grant
Darwin NT

Has anyone here generated any -12 errors? I have, 6 in the last 7 days... Out of thousands of good results...

Last -12 I reported was 26th of June. Reported one in May, too.

Martin

Will they take Monday off for the 4th Holiday? If they do I hope the project stays up through the weekend.

---

Lets see, No work for gpus(or any for that matter), -12 errors, anonymous platform, 2.3 dll files anybody?

---

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

Hopefully. I just hope they can do it remotely because nobody will come in to work on the weekend. Especially a holiday weekend.

Bull#h!t.
They've done that on many occasions when the system hasn't been able to be restarted remotely, inspite of attitudes such as yours.

Chill out.. I'm not saying anything bad about the staff. I'm going by what Matt said about going on a three week vacation starting today and also he mentioned a while back (Not long ago, don't feel like going through all his posts) that basically we shouldn't expect them to come in one weekends. Hell, maybe it was Eric that mentioned it. My memory isn't that great anymore.

I have great respect for all of the staff at SETI and I'm well aware they work their a**es off for this project and have come in numerous weekends.

If you're asking are we using the 2.3dlls, yes I am. As I said I only had the one -12 error over the outage. I now have the new limit of 60 WUs spread between my two CPUs and one 9500GT GPU. So you don't have to ask, I'm running optimized without the flops count added.

@forward...They probably will take Monday off but hopefully they will be able to raise the quotas remotely if they haven't got it set to raise automatically.

---

PROUD MEMBER OF Team Starfire World BOINC

Unfortunatly some of us have hosts that will process more than 20 WU's at a time. Case in point is a dual Intel X5680 system that I have with 2 Fermi cards running two instances per card. That is a total of 28 tasks at a time that could be processed if this limitation was lifted. Also I am building a system with a Quad sockets and each cpu has 8 cores and hyperthreading - and two fermi cards to boot. That will be 68 tasks at a time processing.

If this behavior is temporary and expected that is fine but if this is the way it is going to be I am seriously considering other options for my resources. This system is being built as a mega-cruncher and has a significant price tag associated with it. If it cant be used to its full potential I won't bother.

I'm not saying I'm dumping S@H or anything like that but do want to make it known that there are individuals with systems that are more than able to process more than 20 WU's at a time and a higher limit should be implemented.

Todd

I can understand the frustration, but it's still a big improvement on the usual "turn on the tap" recovery from a long outage.

The usual process if you remember is that the server and network link get hammered to death for 12 hours or more, uploads fail, and therefore you cant download any new work anyway.

At least this way the completed units all uploaded, and SOME new work was sent out to most crunchers. A big improvement over the chaos that most people expected. A few have got caught out with only CPU or GPU, but at least they have some work, and can report / request more without the usual hassles.

I see the limit has been raised to 60 in progess now, with a corresponding jump in the network traffic, so I assume things are starting to get back to normal.

Hopefully full caches for everyone before Tuesday?

Ian

Network traffic has dropped right off & there is plenty of work ready to send, so i guess almost everyone has finally hit the task limit.

Good, maybe they'll lift the limit so we can build a cache.

Doesn't look like it...
It would appear that they may have increased the limit...say, from 20 to maybe 50 or something, but folks are still reporting hitting the in progress limit.
So they haven't opened up the spigot all the way just yet.
Maybe tomorrow?

Looks like the new limit is 60. Two machines have gotten up to that many WUs. The third already had more (good cache going into the shutdown), so it's not getting any more for a while.

---

Lets see, No work for gpus(or any for that matter), -12 errors, anonymous platform, 2.3 dll files anybody?

@ VW Bobier

Since you cpu has 4 cores, as long as you have more than 2* cores = 8 wu waiting to upload, Boinc will not ask for work. As soon as it hits 8 or less then Boinc will start requesting work.

I saw in another posting by you that you were still uploading cpu work.

---

Boinc....Boinc....Boinc....Boinc....

As of right now, my three rigs have work. I ran Rescheduler on my CUDA rig so I now have CUDA work as well. For now, that's all that's needed and I'm happy.

Now if the servers were to crash tonight, I'd only have about 12 hours of work on the CUDA rig even though my cache is set for 5 days. I've got the voodoo going.

Hopefully, whatever new load control experiment that's being run proves successful so we can stock up on work prior to the scheduled outage next Tuesday. It did make for a successful restart after 3 days being down.

---

Join the PACK!

As of right now, my three rigs have work. I ran Rescheduler on my CUDA rig so I now have CUDA work as well. For now, that's all that's needed and I'm happy.

Now if the servers were to crash tonight, I'd only have about 12 hours of work on the CUDA rig even though my cache is set for 5 days. I've got the voodoo going.

Hopefully, whatever new load control experiment that's being run proves successful so we can stock up on work prior to the scheduled outage next Tuesday. It did make for a successful restart after 3 days being down.

Well...good news is the splitters just started back up and are building ready to send again.
And the waiting for assimilation total is starting to come down.

So I guess I'll just have to tell the kitties to start clicking their ruby slippers again...............

---

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

Network traffic has dropped off again so it looks like everyone's hit the task limit once again.

---

Grant
Darwin NT

ya know, everybody has a certain amount of pride that might not let them admit when their wrong.. But you did jump on me and accuse me of something I didn't do.. Least you could say would be "Sorry" or even a simple "my mistake".

AFAIK, this project has never asked for more than "spare" CPU cycles to get the job done. If this has changed, please let me know. Building more and more powerful systems then complaining when work is not available to feed them seems against the original concept of the project.

It's my opinion that we should be happy when the admimistrators of the project allow us to crunch for them. Isn't that why we all originally joined the project. That was the way it was when I joined back in 1999 :)

---

Looking at the bandwidth usage on Cricket and reading the posts, my uneducated guess is that they have a script running that is monitoring the server loads and was designed to spoon feed everyone. As the load balances out, I suspect it will incrementally increase the cache limits until they are removed completely.

That would be my solution to the ridiculous pounding we've seen after outages in the past. This approach ensures that everyone gets some work while the big crunchers with the large caches are prevented from saturating the system and gumming up the works. They've tried various load balancing schemes in the past that didn't work, but this one seems to be working.

---

Boinc V7.2.42
Win7 i5 3.33G 4GB, GTX470

new limit on seti tasks=42?

---

Janice

This is after the short break for database server crash, the replica is still off-line.

And the title msg is still there, so no changes there.