RTS over 800K. We have seen this before where the splitter cutoff mechanism doesn't trip correctly and the splitters continue to split but the scheduler doesn't recognize requests for work and sends out no tasks available messages. Looks like the scheduler picked up the pieces again and tasks are going out to refill caches.

That's the resends sending the RTS up Keith, not the splitters. ;-)

Cheers.

Really? We get 300K resends for every hour??

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Thanks for the confirmation.
Hopefully just a chron job that will clear in a bit.

Meow.

. . I don't think I have asked this before, but what is a Chron job?

Stephen

? ?

Well, the kitties will send the word out.
If Eric or Jeff are about and wish to attend to it, fine.
If it doesn't get fixed before tomorrow's outage, that is just fine too.

Meow.

. .Exactly ..

Stephen

. .

Simply put, a shell script that runs on a timed interval.

If you look in your /etc folder there are cron.hourly ... cron.daily .. etc
If you put shell scripts in those folders, they will be started by the system.

---

Thanks for the confirmation.
Hopefully just a chron job that will clear in a bit.

Meow.

. . I don't think I have asked this before, but what is a Chron job?

Stephen

? ?

cron job is just a script that is set to go off at a certain time.

. . Hi Brent,

. . Thanks for the explanation ...

Stephen

:)

Just got a bunch of gpu tasks after I went from 2 gpus to 5 gpus on one box :)

Tom

---

A proud member of the OFA (Old Farts Association).

I assume you are speaking of Host #8565979? Looks like you are only crunching on the gpus? Only see 500 tasks in progress and it would be 600 if you used the cpu.

But I also see your other Host #8610031 with 579 tasks in progress but only two gpus listed. Have you got almost 300 'ghosts" on that machine? Or is the new five gpu host and the website hasn't caught up yet with the changes?

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

I assume you are speaking of Host #8565979? Looks like you are only crunching on the gpus? Only see 500 tasks in progress and it would be 600 if you used the cpu.

But I also see your other Host #8610031 with 579 tasks in progress but only two gpus listed. Have you got almost 300 'ghosts" on that machine? Or is the new five gpu host and the website hasn't caught up yet with the changes?

https://setiathome.berkeley.edu/show_host_detail.php?hostid=8565979

Let me check. Its showing 5 gpus now so must have been a slow update.

I seem to be getting gpu tasks but no seti cpu tasks right now on it. I run a mixed bunch of projects on that box so sometimes the scheduler is convinced I don't need Seti CPU tasks.

https://setiathome.berkeley.edu/show_host_detail.php?hostid=8610031

This one is pure Seti. And it only has two GPUs. Seems to be full up. So far.

But here comes Wed morning. :)

Tom

---

A proud member of the OFA (Old Farts Association).

Host 8610031 was the one I was talking about. Either you have already started to bunker or it has 300 ghosts since its cache should only be 300. The task in progress shows 579.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)

Looks like the servers are slowing down just ahead of the shutdown. Cannot get any new tasks this morning on the east coast.

12/26/2018 6:34:49 AM | SETI@home | Reporting 4 completed tasks
12/26/2018 6:34:49 AM | SETI@home | Requesting new tasks for CPU and NVIDIA GPU
12/26/2018 6:35:11 AM | SETI@home | Scheduler request failed: Couldn't connect to server
12/26/2018 6:35:12 AM | | Project communication failed: attempting access to reference site
12/26/2018 6:35:13 AM | | Internet access OK - project servers may be temporarily down.

Roger

---

Looks like the servers are slowing down just ahead of the shutdown. Cannot get any new tasks this morning on the east coast.

12/26/2018 6:34:49 AM | SETI@home | Reporting 4 completed tasks
12/26/2018 6:34:49 AM | SETI@home | Requesting new tasks for CPU and NVIDIA GPU
12/26/2018 6:35:11 AM | SETI@home | Scheduler request failed: Couldn't connect to server
12/26/2018 6:35:12 AM | | Project communication failed: attempting access to reference site
12/26/2018 6:35:13 AM | | Internet access OK - project servers may be temporarily down.

Roger

Yep

---

Wed 26 Dec 2018 07:19:25 AM EST | SETI@home | Scheduler request failed: HTTP internal server error

<Panic ON>

---

I released my 2900ish bunkered tasks. I hope the project goes down soon :)

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

The data is flowing again, refill your caches while you can before the outage.

---

The data is flowing again, refill your caches while you can before the outage.

Thanks for the heads-up. What time does the outage usually start?

---

The data is flowing again, refill your caches while you can before the outage.

Figures. Right as I get on the plane haha. If it goes down at 8am ET like it has been recently, then it won’t be much consequence. I can remote in and I have a layover in Chicago at 10am ET, but it will probably be down by then.

---

Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

Beta down.

. . Interesting outage. It was the first 'false start' I have seen. It went down at 3:30pm UTC but was back up at 4:25pm, then went down again at 4:55pm. I know it was up because I received new work during that interval. Now it is back up at 11:00pm.

. . Hopefully it went smoothly despite that hiccough.

Stephen

:)

[edit]

. . Ruh Roh! It seems the upload/download servers are crumbling under the post outage onslaught ...

Going to have to be patient. Servers are timing out on connection requests.

---

Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)