Forums slowed to a crawl & then the website went MIA. Was it a glitch or are we about to crash & burn again?

Not sure, but I seen that too. If I were a betting man, I'd default to my favorites from Hackers.

  Crash Override
 Acid Burn

The remaining resends will be processed quickly despite having long deadlines. The few hosts that will still be accepting work a few weeks from now will have empty caches so they will return the results very fast and also they are unlikely to just disappear if they haven't already done so by that time.

The remaining resends will be processed quickly despite having long deadlines. The few hosts that will still be accepting work a few weeks from now will have empty caches so they will return the results very fast and also they are unlikely to just disappear if they haven't already done so by that time.

Actually the hosts have probably signed up to projects with short turn round times and therefore Seti tasks will sit on the back burner until close to the expiry date.

with the cleared backlog, need to reactivate the server ghost script but in another way, TO NOT RESENT WU GHOSTED ON HOSTS WITH AntiVirus PB ^^ but to OTHERS ... if not, it will enter in an endless loop

---

Actually the hosts have probably signed up to projects with short turn round times and therefore Seti tasks will sit on the back burner until close to the expiry date.

Yep, had that happen to me yesterday, so suspended project until all Seti tasks completed.

Well. SETI was my only project, and will remain so. I'll keep the machines doing what they do until it no longer runs. Then I'll uninstall boinc.

I see the Scheduler has had issues again, although it appears to coincide with general status updating problems- there was a flat line of the graphs for a few other processes around the time the Scheduler started to fall behind again.
At least it's catching up fairly quickly.

---

Grant
Darwin NT

Here we get to the end fo this project, and I'm finally getting my windows laptop to play with my nix boxes. openssh is installed, and powershell on my nix boxes. Vim installed on my laptop. It's almost cool. Would have been fun playing with all of this with a running project :)

with the cleared backlog, need to reactivate the server ghost script but in another way, TO NOT RESENT WU GHOSTED ON HOSTS WITH AntiVirus PB ^^ but to OTHERS ... if not, it will enter in an endless loop

It won't be an endless loop. Ghost won't get new deadlines on resend, so they will expire on the original schedule and will then get resent to other users.

Assimilation queue hasn't been this small since early February.

I'm watching the assimilation queue - 2,221,676 at 07:40 UTC - and my personal total of valid but unpurged tasks - currently 8521.

In theory, they should both hit ground zero at the same time (near enough), but if they arrive separately it'll suggest another problem disguised in the noise.

Came across this computer: https://setiathome.berkeley.edu/show_host_detail.php?hostid=8873201
It has 15k tasks in progress (running BOINC 7.17.0). How is this possible ?

Tom

Came across this computer: https://setiathome.berkeley.edu/show_host_detail.php?hostid=8873201
It has 15k tasks in progress (running BOINC 7.17.0). How is this possible ?

Tom

Really doesn't matter how. Looking at what their system can actually do; means they're going to have a lot of timed out tasks.
https://setiathome.berkeley.edu/host_app_versions.php?hostid=8873201

Edit: not to mention that this system hasn't contacted the servers since the 8th.

Came across this computer: https://setiathome.berkeley.edu/show_host_detail.php?hostid=8873201
It has 15k tasks in progress (running BOINC 7.17.0). How is this possible ?

Probably cheater with faked number of GPUs (the system is claiming to have 64 GPUs, how likely is that?) to get around the limits set by the project staff to limit the load on the database. People doing this should be banned from the project IHMO, at least for a week or so. Some of the tasks can also be ghosts.

---

Came across this computer: https://setiathome.berkeley.edu/show_host_detail.php?hostid=8873201
It has 15k tasks in progress (running BOINC 7.17.0). How is this possible ?

Tom

Hasn't returned anything after the April 2nd but has kept contacting for several days longer. So most likely an empty cache but a lot of ghosts.

And, more specifically, that host hasn't contacted the servers since 8 April.

That's not a major problem - the servers have had plenty of other things to worry about recently - but looking at the Munin graphs in detail, the assimilation queue has declined from 4.82M to 2.07M in 36 hours - and at a very uniform slope. That process should be complete in a further 27 hours, around midday UTC Monday 13 April.

It's been pointed out elsewhere that the user has probably switched to another project, and BOINC will be giving that project priority until the resource shares even out. Fair enough, although with the owner having chosen anonymity, we can't see or link to their other project, if any.

But the user must have put some effort into building and configuring a beast like that, and presumably they did it specifically for the benefit of SETI@Home. In that case, I would expect them to pause the replacement project, and return to SETI crunching, as soon as the servers are ready to receive and process the returned tasks. Wingmates are waiting for the returns, too, and the project staff are waiting to get on with their analysis. Let the cleanup commence on Monday afternoon.

Once the assimilators have catched up, they should reenable the ghost resending mechanism on the servers (the full version of it), the servers should now be able to handle that additional load.

---

Once the assimilators have catched up, they should reenable the ghost resending mechanism on the servers (the full version of it), the servers should now be able to handle that additional load.

I'll second that motion.

Meow!

---

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

Ah, we're back.
I was getting nothing but a dozen or so messages about missing drives on Carolyn for a while there.

---

Grant
Darwin NT

Once the assimilators have catched up, they should reenable the ghost resending mechanism on the servers (the full version of it), the servers should now be able to handle that additional load.

I'll second that motion.

Meow!

Perhaps you can send a message to Eric like you did in the past (IIRC).

---