Problems and Technical Issues with Rosetta@home

Message boards : Number crunching : Problems and Technical Issues with Rosetta@home

To post messages, you must log in.

Previous · 1 . . . 158 · 159 · 160 · 161 · 162 · 163 · 164 . . . 232 · Next

AuthorMessage
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104258 - Posted: 15 Jan 2022, 18:00:22 UTC - in response to Message 104256.  

I have an Intel i5 9400F and it runs rosetta python on it without me making any change.Recently I could not connect neither Firefox nor Edge and I found someone had activated a proxi, I don't know how, I am the only user. I disactivated proxi and everything works on Windows 11. Was one of the frequent Microsoft updates to do it?
Tullio
That seems very odd and if I were you I'd check for viruses and malware, a proxy should never appear without you doing so, and could be an attempt to reroute your internet traffic to grab passwords - unless you're using some kind of Norton Internet security? What was the address of the proxy? I use AVG for viruses and Malwarebytes for Malware. Both are free.
ID: 104258 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104260 - Posted: 15 Jan 2022, 18:26:08 UTC - in response to Message 104256.  
Last modified: 15 Jan 2022, 18:28:07 UTC

I have an Intel i5 9400F and it runs rosetta python on it without me making any change.Recently I could not connect neither Firefox nor Edge and I found someone had activated a proxi, I don't know how, I am the only user. I disactivated proxi and everything works on Windows 11. Was one of the frequent Microsoft updates to do it?
Tullio
One of your computers reports:
Microsoft Windows 11 Core x64 Edition, (10.00.22000.00)
ROTFPMSL, so Windows 11 is actually 10.0.2? This is getting as bad as Firefox version 96.
ID: 104260 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104261 - Posted: 15 Jan 2022, 18:51:38 UTC
Last modified: 15 Jan 2022, 18:55:51 UTC

Connecting outside of GRC did not help.

No response from David Baker yet.

Perhaps WUs like this one, where two of us screwed up, they'll see a lot of and look into the problem:
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=1306550771
(Note, "Too many total results")
ID: 104261 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Greg_BE
Avatar

Send message
Joined: 30 May 06
Posts: 5550
Credit: 5,554,708
RAC: 58
Message 104263 - Posted: 15 Jan 2022, 20:24:28 UTC - in response to Message 104261.  
Last modified: 15 Jan 2022, 20:28:30 UTC

Connecting outside of GRC did not help.

No response from David Baker yet.

Perhaps WUs like this one, where two of us screwed up, they'll see a lot of and look into the problem:
https://boinc.bakerlab.org/rosetta/workunit.php?wuid=1306550771
(Note, "Too many total results")



DB....told you..no one and I do mean no one at RAH responds to anything.
Your just SOL until WE figure out the problem.

Have you checked your VM environment and cleaned that up if needed?
The ARC error I already told you what my research came up with and you checked that. So I don't know what else to tell you.
I can only say now its beyond anything we understand here and your better off asking in the BOINC forum over at the Oracle Vbox forum.

But your not alone, due to this nonsense with bloating files I have blown up 10 tasks.
I just put a app_config with project_max in place to stop that.
I just ordered a 500GB m.2 drive that will be here later in the week. That will be dedicated to BOINC. If python blows through that much space then there is some serious BS going on.
ID: 104263 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104264 - Posted: 15 Jan 2022, 20:50:00 UTC - in response to Message 104263.  

I can't believe the CEO doesn't care about 6 in 7 machines not doing his research. I'll be emailing the admin dude if I don't get a response soon.

I've tried on many machines reinstalling VB, different versions, cleaning it up (I assume you mean deleting the inaccessible images? Let me know if there's something else I should be doing, this normally fixes the error "VB environment needs cleaning up" I get in Boinc for other projects).

I'd like to know what others are finding. Am I unusual in having 6 machines that simply do not run any of them properly? My 1 good machine only blew up one of them, all the others are fine. Apart from the Ryzen 9, the other bad machines are all quite old (DDR3 spec).
ID: 104264 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jim1348

Send message
Joined: 19 Jan 06
Posts: 877
Credit: 51,526,729
RAC: 2,981
Message 104265 - Posted: 15 Jan 2022, 21:06:49 UTC - in response to Message 104264.  

I can't believe the CEO doesn't care about 6 in 7 machines not doing his research. I'll be emailing the admin dude if I don't get a response soon.

I concluded some time ago that the great Dr. David Baker (who should get a Nobel Prize) turned over the management of Rosetta to others a long time ago.
And with the advent of AI, he can do more in-house anyway. As long as enough of the peripheral work is done by others, that is good enough for him.

He has more valuable uses for his time. And we have other projects if we want them.
ID: 104265 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104266 - Posted: 15 Jan 2022, 21:17:22 UTC - in response to Message 104265.  

I can't believe the CEO doesn't care about 6 in 7 machines not doing his research. I'll be emailing the admin dude if I don't get a response soon.

I concluded some time ago that the great Dr. David Baker (who should get a Nobel Prize) turned over the management of Rosetta to others a long time ago.
And with the advent of AI, he can do more in-house anyway. As long as enough of the peripheral work is done by others, that is good enough for him.

He has more valuable uses for his time. And we have other projects if we want them.


But enough work is not being done. The number of tasks being completed on the server status is abysmal compared to pre-VB.
ID: 104266 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jim1348

Send message
Joined: 19 Jan 06
Posts: 877
Credit: 51,526,729
RAC: 2,981
Message 104269 - Posted: 15 Jan 2022, 21:41:11 UTC - in response to Message 104266.  

But enough work is not being done. The number of tasks being completed on the server status is abysmal compared to pre-VB.

You can advise Dr. Baker of that if you can get his attention on the matter at all. The fact that it hasn't been corrected seems to indicate that he does not consider it to be a big problem.
Good luck.
ID: 104269 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104271 - Posted: 15 Jan 2022, 22:39:50 UTC - in response to Message 104269.  
Last modified: 15 Jan 2022, 22:42:47 UTC

But enough work is not being done. The number of tasks being completed on the server status is abysmal compared to pre-VB.

You can advise Dr. Baker of that if you can get his attention on the matter at all. The fact that it hasn't been corrected seems to indicate that he does not consider it to be a big problem.
Good luck.
Maybe they don't need the work done quickly and it's not worth the effort to sort it. For example the covid work over in WCG is being given out very slowly because apparently the scientists can't interpret the results that quickly.

Or maybe they know about the problem and the programmers are trying to fix it, but they don't have the staff to discuss it with us.

In the mean time I'll try contacting through the admin email address if I get no response from David, and keep running my working machine on the pythons, and I'll retry the other machines every so often to see if it's fixed.

It's very odd that it works fine on one of my machines and not the others. I'll try asking over in LHC what those errors mean.
ID: 104271 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Jim1348

Send message
Joined: 19 Jan 06
Posts: 877
Credit: 51,526,729
RAC: 2,981
Message 104272 - Posted: 15 Jan 2022, 22:53:48 UTC - in response to Message 104271.  

I am glad that you are making the effort, but the problems that I see and you see may be different.
I have no problem getting the work, but I run my machines 24/7 and see any hangs. A lot of people reboot every day and probably don't.
And I am on Linux, whereas Windows does not seem to have the "Vm job unmanageable" problem, though there still remains a bit of the "0 CPU" problem for everyone.

They may be getting enough work back to keep them happy. We don't know. But keep at them.
ID: 104272 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104273 - Posted: 15 Jan 2022, 22:55:21 UTC - in response to Message 104272.  
Last modified: 15 Jan 2022, 23:34:44 UTC

I am glad that you are making the effort, but the problems that I see and you see may be different.
I have no problem getting the work, but I run my machines 24/7 and see any hangs. A lot of people reboot every day and probably don't.
And I am on Linux, whereas Windows does not seem to have the "Vm job unmanageable" problem, though there still remains a bit of the "0 CPU" problem for everyone.

They may be getting enough work back to keep them happy. We don't know. But keep at them.
Thanks for your vote of confidence. My good machine can do them 24/7 without a problem, only one WU I've seen fail. It's on Windows 10. But my 6 bad machines get the 0 CPU thing with every single WU.
ID: 104273 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
.clair.

Send message
Joined: 2 Jan 07
Posts: 183
Credit: 23,178,785
RAC: 204
Message 104275 - Posted: 16 Jan 2022, 2:07:50 UTC

A strange thing happened to me today
I ran out of pythons :)
not one
even in the que
the big system is happily running 35 R4.2 tasks all together
rosetta is only using 11.2GB of disk space
and total system use of ram is only 17.5GB
Wot fun
it wont last
Tomorrow will be different [or rather later today]
off to bed
ID: 104275 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104276 - Posted: 16 Jan 2022, 2:23:45 UTC - in response to Message 104275.  
Last modified: 16 Jan 2022, 2:24:20 UTC

Ran out? You mean the server had none to give you? I can see 5000 now, alledgedly. Although my machine has been getting Sidock instead, maybe there were none left briefly.

[checks logs]

No, the server was inaccessible from about 00:30 to 02:00 UTC.

Two o'clock! Christ. I need to go to bed, I have an addiction to Fallout 4.
ID: 104276 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Greg_BE
Avatar

Send message
Joined: 30 May 06
Posts: 5550
Credit: 5,554,708
RAC: 58
Message 104277 - Posted: 16 Jan 2022, 9:59:40 UTC - in response to Message 104265.  

I can't believe the CEO doesn't care about 6 in 7 machines not doing his research. I'll be emailing the admin dude if I don't get a response soon.

I concluded some time ago that the great Dr. David Baker (who should get a Nobel Prize) turned over the management of Rosetta to others a long time ago.
And with the advent of AI, he can do more in-house anyway. As long as enough of the peripheral work is done by others, that is good enough for him.

He has more valuable uses for his time. And we have other projects if we want them.



EXACTLY

The minute he stopped writing blogs here about the research, I knew he had moved on.
When the grad students stopped writing here and the mod left I knew this was going to be a side project
Then with all the problems, DEK doesn't respond to anything and we have to figure it out on our own.
We have a spammer in the new section, they don't care.
We have problems, they don't care.
This is just like TACC, its a side project now.
Out of those 2 million tasks, just 5,000 are assigned here.

So why be surprised? Us old guys are used to this. We have seen the decline over the years.
If you got one machine that can run Python keep that running and put your other machines to work on something that does work. I would have given up long ago and moved on.
ID: 104277 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile dcdc

Send message
Joined: 3 Nov 05
Posts: 1828
Credit: 106,943,056
RAC: 3,252
Message 104278 - Posted: 16 Jan 2022, 13:11:44 UTC - in response to Message 104277.  
Last modified: 16 Jan 2022, 13:12:17 UTC

I don't think he has exactly moved on - the world of protein simulation and design has become incredibly successful so he is in huge demand. It has been a long road of steady improvement, but the huge breakthrough send to have been Alphafold 2, which had then also led to RoseTTAFold.
E.g. https://www.nature.com/collections/dfejabhghd/

They will be struggling to keep up. And please don't email David Baker about Rosetta issues - the admin account is much more reasonable. It's a bit like emailing Bill Gates about problems with Windows these days! But I get that the lack of communications is very frustrating, especially given how unstable the VirtualBox tasks are.

It doesn't help that BOINC development is no longer funded because it doesn't deal with VirtualBox or GPU tasks very well IMHO.

And it also doesn't help that we don't have a mod on these forums any more.

D
ID: 104278 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
tullio

Send message
Joined: 10 May 20
Posts: 62
Credit: 612,922
RAC: 25
Message 104279 - Posted: 16 Jan 2022, 15:02:08 UTC - in response to Message 104260.  

One of my computers, that with an AMD Ryzen 5, stil lhas Windows 10. Two others, one Intel and one AMD, have Windows 11. I use the Windows Defender as antivirus, updated almost daily. When I had another Virus defender some LHC projects wuld not run.
Tullio
ID: 104279 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104282 - Posted: 16 Jan 2022, 17:16:08 UTC - in response to Message 104279.  

One of my computers, that with an AMD Ryzen 5, stil lhas Windows 10. Two others, one Intel and one AMD, have Windows 11. I use the Windows Defender as antivirus, updated almost daily. When I had another Virus defender some LHC projects wuld not run.
Tullio
"Still has windows 10"? Maybe I need to leave the updates running to get more chance. I disabled those pesky things because a reboot every other day is beyond a joke. I just do a manual update every 3 months. The last time was a week or so ago, and all I got was the oldest machine saying "we're working on compatibility with this processor" and the 2nd oldest saying "this is not compatible". The decent machines didn't even get a mention.
ID: 104282 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104283 - Posted: 16 Jan 2022, 17:17:30 UTC - in response to Message 104278.  

Please don't email David Baker about Rosetta issues - the admin account is much more reasonable. It's a bit like emailing Bill Gates about problems with Windows these days! But I get that the lack of communications is very frustrating, especially given how unstable the VirtualBox tasks are.
Well I wouldn't have if he had decent staff in here. Too late now anyway. I'll send the same message to the admin account. I've also asked in the virtualbox forum. LHC have been helpful but the images are different to theirs so they don't want to spend time looking into it.
ID: 104283 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile Greg_BE
Avatar

Send message
Joined: 30 May 06
Posts: 5550
Credit: 5,554,708
RAC: 58
Message 104284 - Posted: 16 Jan 2022, 18:29:21 UTC

And the "fun" continues.
6 plus hours for 25 mins cpu time. 4 hrs plus for completion.
CPU usage .41% or less
Just jettisoned 5 tasks.
ID: 104284 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Peter Hucker of the Scottish Boinc Team
Avatar

Send message
Joined: 12 Aug 06
Posts: 1278
Credit: 5,835,886
RAC: 2,367
Message 104285 - Posted: 16 Jan 2022, 18:39:16 UTC - in response to Message 104284.  

And the "fun" continues.
6 plus hours for 25 mins cpu time. 4 hrs plus for completion.
CPU usage .41% or less
Just jettisoned 5 tasks.
I wonder if there's some way we can gather stats for how many are returned broken? The info is public, but not collated AFAIK.
ID: 104285 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Previous · 1 . . . 158 · 159 · 160 · 161 · 162 · 163 · 164 . . . 232 · Next

Message boards : Number crunching : Problems and Technical Issues with Rosetta@home



©2022 University of Washington
https://www.bakerlab.org