When I update Rosetta I get the following in the log:

4/3/2020 1:48:45 PM | Rosetta@home | Reporting 28 completed tasks
4/3/2020 1:48:45 PM | Rosetta@home | Requesting new tasks for CPU
4/3/2020 1:48:47 PM | Rosetta@home | Scheduler request failed: Error 400

The 28 reported tasks remain in my queue as "Ready to Report".

I'm not having this problem on other projects. Is it me or Rosetta?

Edit / PS: Thanks to the moderator who moved this to the proper forum.

---

Front page says 560k tasks queued
Server status page says 15k ready to send
Lower on server status page, all 15k are Android tasks. No Rosetta nor MiniRosetta available to download

We're out for the moment

Oh, cancel that. Server status page just updated and tasks already coming down. Sorry

And ran out, and came back and ran out again.
Back to being hand to mouth

---

I am currently running Rosetta 4.07 on my computer. But I see that the latest version is Rosette 4.12 on the site. I also read that the update is automatic. While the release date of 4.12 is April 1, the day before yesterday, is there anything I can or should do to speed the update along to my computer?

Thanks,

Ross

---

I am currently running Rosetta 4.07 on my computer. But I see that the latest version is Rosette 4.12 on the site. I also read that the update is automatic. While the release date of 4.12 is April 1, the day before yesterday, is there anything I can or should do to speed the update along to my computer?

Thanks,

Ross

It will download the appropriate application based on the work unit being crunched. So you don't need to do anything, you'll get the update automatically when it's needed on your end.

My PC does not get any WU. I mean ANY. Why?

Each work-unit is associated with a particular version of Rosetta. If you don't have 4.12 yet, that could be because you're storing more work than necessary, which means you're running older tasks, which mean you haven't gotten to the new application yet. My computer started getting 4.12 work about 18 hours ago.

If you have a good reason to store more work than the default, such as using an intermittent internet connection, then you'll just have to be patient. If your internet service is always on, try the defaults ("Store at least 0.1 days of work." "Store up to an additional 0.5 days of work.") and be patient for your client to catch up.

Shorter storage helps the project achieve shorter turn-around times, so it's also a win for the researchers.

I am currently running Rosetta 4.07 on my computer. But I see that the latest version is Rosette 4.12 on the site. I also read that the update is automatic. While the release date of 4.12 is April 1, the day before yesterday, is there anything I can or should do to speed the update along to my computer?

Of your two PCs, one has 4.12 tasks now, the other shows 3 tasks from March, before 4.12 came out, 2 of which never completed and timed out and the last one not returned but close to deadline - and no credit at all.
Have you run anything yet on that one? Doesn't look like it

---

I am currently running Rosetta 4.07 on my computer. But I see that the latest version is Rosette 4.12 on the site. I also read that the update is automatic. While the release date of 4.12 is April 1, the day before yesterday, is there anything I can or should do to speed the update along to my computer?

Thanks,

Ross

Just wait. Many workunits were set up to try to use 4.07 if possible, and those haven't run out yet. When you get a workunit set up to prefer 4.12, it will be downloaded automatically. It comes in a rather large file, which took over an hour to download for me.

My PC does not get any WU. I mean ANY. Why?

One of your computers is running Windows XP. I'm not sure if the latest versions of the Rosetta tasks are still compatible with that version of Windows.

Another is running Android, but with only enough memory to run one Rosetta task at a time. It should still do something if you tell BOINC not to try running more than one task at a time, though.
Also, it's still running a rather old version of BOINC; it's unclear if that's also a problem.

Another is running Windows 7. It's unclear if that's a problem.
Also, it's not clear if it has enough memory to run more than 3 tasks at a time.

Another one, also running Android, does not appear to have enough memory to be good at running even one Rosetta task at a time.
It's also running a rather old version of BOINC.

Another one is also running Windows XP, and may not have enough memory to run even one Rosetta task at a time.

Rosetta@Home recently got enough new users that they cannot keep from running out of tasks several times a day. My computer is now running fewer Rosetta tasks than usual, because it's hard to get any.

Only just joined Rosetta after the death of ET. I am not getting any work at all. Why not? Last few messages:
04/04/2020 01:24:57 | Rosetta@home | update requested by user
04/04/2020 01:24:58 | Rosetta@home | Sending scheduler request: Requested by user.
04/04/2020 01:24:58 | Rosetta@home | Requesting new tasks for CPU
04/04/2020 01:24:59 | Rosetta@home | Scheduler request completed: got 0 new tasks
04/04/2020 01:24:59 | Rosetta@home | No tasks sent
04/04/2020 01:25:09 | Rosetta@home | Sending scheduler request: To fetch work.
04/04/2020 01:25:09 | Rosetta@home | Requesting new tasks for CPU
04/04/2020 01:25:11 | Rosetta@home | Scheduler request completed: got 0 new tasks
04/04/2020 01:25:11 | Rosetta@home | No tasks sent

---

Only just joined Rosetta after the death of ET. I am not getting any work at all. Why not?

Because there isn't any (ignore the number in the "Tasks ready to send" on the Serve Status page- it doesn't seem to relate to the work available to download in any way, and the "Unset" number in the Tasks by application list gives & rough idea, but it can be several hours between updates of that value. Would be nice if it were every 5-10min or so).

---

Grant
Darwin NT

same issue - added a few machines today, some got WU's and some didn't. I thought maybe there was an account based throttle, where because I have a bunch downloaded already maybe I couldn't get more until some were reported back. But just totally guessing, no idea how the allocations work.

In the mean time I added some Milkyway@Home work.

Only just joined Rosetta after the death of ET. I am not getting any work at all. Why not? Last few messages:
04/04/2020 01:24:57 | Rosetta@home | update requested by user
04/04/2020 01:24:58 | Rosetta@home | Sending scheduler request: Requested by user.
04/04/2020 01:24:58 | Rosetta@home | Requesting new tasks for CPU
04/04/2020 01:24:59 | Rosetta@home | Scheduler request completed: got 0 new tasks
04/04/2020 01:24:59 | Rosetta@home | No tasks sent
04/04/2020 01:25:09 | Rosetta@home | Sending scheduler request: To fetch work.
04/04/2020 01:25:09 | Rosetta@home | Requesting new tasks for CPU
04/04/2020 01:25:11 | Rosetta@home | Scheduler request completed: got 0 new tasks
04/04/2020 01:25:11 | Rosetta@home | No tasks sent

Your computer looks suitable, but there are enough new members here that the server usually has very few if any tasks available.

Keep trying.

same issue - added a few machines today, some got WU's and some didn't. I thought maybe there was an account based throttle, where because I have a bunch downloaded already maybe I couldn't get more until some were reported back. But just totally guessing, no idea how the allocations work.

In the mean time I added some Milkyway@Home work.

Your computers are hidden, so I can't see enough information to tell how many of them are suitable for Rosetta work.

However, there are so many new users here that there are usually very few if any tasks available for download.

Milkyway@Home looks like a good source of short tasks, to fill in the times between any tasks you get here.

I'm replying to myself because, while there seem to be a lot of people having trouble getting work, nobody else seems to be getting the 400 error. This is happening on only one of the two machines I have and that machine is having no trouble with the other projects I've tested. (I'm running Rosetta exclusively now except for those tests.) This problem started suddenly yesterday or possibly the day before. It had been working fine.

I've rebooted to no avail. The only other thing I can think of to try is to reset the project. I'm reluctant to do that for fear of losing the 41 (at this point) tasks listed as "Ready to Report".

If anybody has any advice I'd appreciate it. If not, tomorrow morning (it's late where I am now) I'll go for the nuclear option and reset.

Thanks.

---

With your computers hidden it's pretty much impossible to offer any useful advice.
A quick Google comes up with that in the past that error has been associated with older BOINC Manager versions & invalid Certificates, in those cases upgrading to a more recent BOINC version sorted it out. In others it has been due to an incorrectly set clock (wrong year), or the proxy they are using causing issues, but whether or not any of that applies to you, i've not got the slightest idea.

---

Grant
Darwin NT

With your computers hidden it's pretty much impossible to offer any useful advice.
A quick Google comes up with that in the past that error has been associated with older BOINC Manager versions & invalid Certificates, in those cases upgrading to a more recent BOINC version sorted it out. In others it has been due to an incorrectly set clock (wrong year), or the proxy they are using causing issues, but whether or not any of that applies to you, i've not got the slightest idea.

Thank you, Grant.

The BOINC version is the latest and greatest and my clock is fine. I don't use a proxy ... or didn't think I did. A bank I have an account with though suggested a a day and a half ago that I download a security app so, foolishly, I did. I uninstalled it and all seems to be fine.

If you're ever in my neighborhood (MA, USA), PM me and I owe you a beer!

- Dick

---

A bank I have an account with though suggested a a day and a half ago that I download a security app

Yeah, that'll do it too. Some AV programmes also have a nasty habit of deleting BOINC Tasks as they download because they don't like the look of them, or think the programme's behaviour is supicious.
Glad you got it sorted.

If you're ever in my neighborhood (MA, USA), PM me and I owe you a beer!

I'm one of those very rare Aussies that don't drink. But a Coke (original i think they are over your way) would hit the spot.
:-)

Although it's going to be quite a while before international travel is on the cards for the average person again.

---

Grant
Darwin NT

Hi especially @Grant (SSSF)

Where I am wrong?
I need 2x more time to finish the tasks and 50% GFLOPS on similar i7-8700K CPU

Compare:
- https://boinc.bakerlab.org/rosetta/host_app_versions.php?hostid=3933928
- https://boinc.bakerlab.org/rosetta/host_app_versions.php?hostid=3914491

Thanks in advance.

Could be the Tasks in question?
On my system all Tasks are running to the Target time (other than the odd one that bails out early), and apart from a glitch with some Tasks a few days back that paid out bugger all Credit (and the few early exits), Credit has generally been inline with Runtime.

That was then.

As time has gone on, it looks like the amount of Credit per Task has dropped quite a bit.
I was getting around 400 per 8 hour Task.
Now it's probably around 280 (a 216 low up to 340 high, so far), FLOPs per Application has dropped by around 3% for the current applications with the present Tasks, which is tracking pretty closely with the drop in Credit awarded.

---

Grant
Darwin NT

A bank I have an account with though suggested a a day and a half ago that I download a security app

Yeah, that'll do it too. Some AV programmes also have a nasty habit of deleting BOINC Tasks as they download because they don't like the look of them, or think the programme's behaviour is supicious.

I use the Windows Defender AV program, available free from Microsoft if you have a sufficiently recent version of Windows.

It does not have that nasty habit.

However, a Google search for where to download it gave only three sites, none of which will currently respond.