Unable to contact the Scheduler again,
21/12/2019 08:29:56 | SETI@home | Scheduler request failed: HTTP internal server error
and :Timeout was reached.

The large number of shorties going through at the moment won't be helping matters either.

---

Grant
Darwin NT

I just switched one of my Weekend Warriors to "all cpu" with a dinky little gpu on there that has no hope of processing most anything.

What do I get?

"Project down" try again in an hour.

Reminds me of "all dressed up and no place to go".

Tom

---

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

Well it won't let me report completed tasks, sometimes it says it may be down, or it says it can't connect to server.

---

Well it won't let me report completed tasks, sometimes it says it may be down, or it says it can't connect to server.

i offer to do the sacrifice.... Again.

---

I see that they brought the missing validators back up again after they were down. Don't see any improvement on validation tasks outstanding. They really need to whittle that down to something manageable. I haven't been able to pull up any task web page on any of my hosts since the outage. Don't know if the hosts are doing OK or not.

Hmm, I couldn't get tasks either, but I have no problem visiting my host task page like this one: https://setiathome.berkeley.edu/results.php?hostid=8846357. I don't have as many tasks as you guys do though.

---

The large number of shorties going through at the moment won't be helping matters either.

Results-received-in-last-hour 186,386- much more than 140k often causes issues. But that many, on top of the existing problems? No wonder the system has choked up.
At least the return rate will drop off with people unable to contact the Scheduler, then of course go through the roof again when they can, but hopefully by then some of the other backlogs will have cleared.

---

Grant
Darwin NT

If you want to help this situation the best thing to do would be to have a look through your task list and see if you have any ending with _2 or higher and return these so that "work waiting to be administered" can drop. I am aware this is not possible for everybody. Just my 2 cents.

---

If you want to help this situation the best thing to do would be to have a look through your task list and see if you have any ending with _2 or higher and return these so that "work waiting to be administered" can drop. I am aware this is not possible for everybody. Just my 2 cents.

No clue here how I could select tasks to prioritize reporting...

---

If you want to help this situation the best thing to do would be to have a look through your task list and see if you have any ending with _2 or higher and return these so that "work waiting to be administered" can drop. I am aware this is not possible for everybody. Just my 2 cents.

No clue here how I could select tasks to prioritize reporting...

On Windows you have to select the tasks then on the left-hand side you will see an option saying suspend. So suspend all tasks and then click each task that has _2 or higher and then click resume. The suspend/resume button is located under the graphics button. I imagine it would be a similar process using the graphical interface on Linux. Of course after they are all processed you will need to resume the other work units

---

If you want to help this situation the best thing to do would be to have a look through your task list and see if you have any ending with _2 or higher and return these so that "work waiting to be administered" can drop. I am aware this is not possible for everybody. Just my 2 cents.

Now that's easier said than done Speedy for some of us and seeing as I go through my GPU cache in 9-10hrs anyway the time required for me to go through all those tasks just isn't worth it. Even just doing the CPU tasks I consider a waste of time when they'll be returned in 2-3 days anyway ;-)

Though that'll work ok I guess for those with slower rigs using a minimal cache or those like yourself who share their time around other projects as well and not the straight out SETI only with the 3/0.1 day setting that I'm currently using. Once I get my personal experiments done with here I will initiate my "back up only projects" again, but not until then.

Cheers.

If you want to help this situation the best thing to do would be to have a look through your task list and see if you have any ending with _2 or higher and return these so that "work waiting to be administered" can drop. I am aware this is not possible for everybody. Just my 2 cents.

No clue here how I could select tasks to prioritize reporting...

. . Select all tasks with _0 or _1 at the end and suspend them, then when the _2 and above WUs are crunched unfreeze the others and let it report the completed units. If you want to go to that trouble. It should help clear some older tasks hanging around though.

Stephen

. .

Thanks Wingo I completely understand. All the best with your goals

---

Linux system out of GPU work.
Occasionally contacting the Scheduler, but "Project has no tasks available" is the response.

---

Grant
Darwin NT

Linux system out of GPU work.
Occasionally contacting the Scheduler, but "Project has no tasks available" is the response.

. . I notice you have not implemented the spoofed client ...

Stephen

? ?

I notice you have not implemented the spoofed client

Gave it a go, couldn't get it to run, went back to what worked. Probably needed to set the permissions for the new files.

---

Grant
Darwin NT

I notice you have not implemented the spoofed client

Gave it a go, couldn't get it to run, went back to what worked. Probably needed to set the permissions for the new files.

. . Maybe it was a permissions thing but maybe it was the dependencies. The reason that I only have it running on one of my major systems is that it was the only one already running libcurl4.0 (I think that is the dependency missing on the other 3 rigs).

. . If you go to the directory in which the spoof version files are stored and type sudo ldd and the filename (boinc, boinccmd, boincmgr) it will list all the dependencies and show any that are missing. There was only the same one missing on all of the other three rigs so I hesitated updating that file in case I needed to roll back. With the new limits the 2 slower rigs are AOK now anyway but I really should get it working on the highest performer to ensure its continued operation.

Stephen

. .

. . If you go to the directory in which the spoof version files are stored and type sudo ldd and the filename (boinc, boinccmd, boincmgr) it will list all the dependencies and show any that are missing. There was only the same one missing on all of the other three rigs so I hesitated updating that file in case I needed to roll back. With the new limits the 2 slower rigs are AOK now anyway but I really should get it working on the highest performer to ensure its continued operation.

Stephen

. .

Probably so. I did need a sudo apt install libcurl4 on mine.

---

First I was getting these:

Scheduler request completed: got 0 new tasks
Project has no tasks available

Now it's this:

Scheduler request failed: Couldn't connect to server
Project communication failed: attempting access to reference site
Internet access OK - project servers may be temporarily down.

But the rest of the site is working fine.

Is there something I can do? (Other than PANIC)

First I was getting these:

Scheduler request completed: got 0 new tasks
Project has no tasks available

Now it's this:

Scheduler request failed: Couldn't connect to server
Project communication failed: attempting access to reference site
Internet access OK - project servers may be temporarily down.

But the rest of the site is working fine.

Is there something I can do? (Other than PANIC)

---

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

Not too sure about the server status page numbers. It shows a return rate of 144k, but it's been over 4 hours since either of my systems were able to contact the Scheduler & get a response that wasn't one type of an error or another.

---

Grant
Darwin NT