Panic Mode On (23) Server problems

Message boards : Number crunching : Panic Mode On (23) Server problems
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 . . . 13 · Next

AuthorMessage
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 913109 - Posted: 2 Jul 2009, 0:11:52 UTC - in response to Message 913073.  

My server has said that there are no jobs available for the last twenty-four hours. Since I have uploaded completed tasks successfully and can access this site from the BOINC Manager with no problem, am I to assume that there is simply no work available at this time?
Currently I have one task - an Astropulse task that predicts two hundred forty-four hours until completion. Since I am running only the one, will that task be completed more quickly than if I were running the usual four? Or will it be completed in the usual time frame. I know that I am displaying my ignorance here of how the system works but that's alright. Once someone answers that question and explains it to me I will be just a little less ignorant.
Thank you for your assistance.

Tuesday was the weekly maintenance outage. There is usually a backlog of requests for work after the outage, and not everyone gets work immediately.

Unfortunately, S@H tasks are currently all single threaded so the task will take just about as long as it normally would (there are some very minor time savings due to the lack of RAM contention and task switches, but these would subtract only a few seconds to a few minutes over the course of the task described).

If everything works as usual, your computer should be getting more work sometime today.

I do not specifically know what single-threaded means but have a good idea. So now I can just relax about this and worry about it if there are not some new tasks in the next day or so.
Thank you.

Sorry. Single threaded means that there is one thread of execution. A thread of execution cannot be split across CPUs.


BOINC WIKI
ID: 913109 · Report as offensive
Profile Nightgunner5
Avatar

Send message
Joined: 12 Aug 09
Posts: 1
Credit: 15,453
RAC: 0
United States
Message 926661 - Posted: 17 Aug 2009, 1:11:37 UTC

I keep getting

8/16/2009 7:58:40 PM	SETI@home	Sending scheduler request: To fetch work.
8/16/2009 7:58:40 PM	SETI@home	Requesting new tasks
8/16/2009 7:58:45 PM	SETI@home	Scheduler request completed: got 0 new tasks
8/16/2009 7:58:45 PM	SETI@home	Message from server: (Project has no jobs available)


Over and over and over and over...
ID: 926661 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 926672 - Posted: 17 Aug 2009, 2:33:09 UTC - in response to Message 926661.  

I keep getting

8/16/2009 7:58:40 PM	SETI@home	Sending scheduler request: To fetch work.
8/16/2009 7:58:40 PM	SETI@home	Requesting new tasks
8/16/2009 7:58:45 PM	SETI@home	Scheduler request completed: got 0 new tasks
8/16/2009 7:58:45 PM	SETI@home	Message from server: (Project has no jobs available)


Over and over and over and over...

The servers are running, but the splitters are not keeping up with demand. On each connection you may or may not get any tasks.


BOINC WIKI
ID: 926672 · Report as offensive
David Goodspeed

Send message
Joined: 26 Nov 07
Posts: 1
Credit: 23,213
RAC: 0
United States
Message 926803 - Posted: 17 Aug 2009, 18:11:10 UTC

I also receive work with gaps frequently. What is more disturbing to me is when a significant percentage of my work credits seem to vanish, like recently I was credited with just over 200 when a brief check of the log indicated I should have received more than 650 in a 24 hour span. Maybe I should be checking closer.
ID: 926803 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 926809 - Posted: 17 Aug 2009, 18:44:41 UTC - in response to Message 926803.  

I also receive work with gaps frequently. What is more disturbing to me is when a significant percentage of my work credits seem to vanish, like recently I was credited with just over 200 when a brief check of the log indicated I should have received more than 650 in a 24 hour span. Maybe I should be checking closer.


Two things to keep in mind are 1) credit is only granted for validated work. If you have a machine that is overheating or otherwise malfunctioning causing the results to be off when compared to a quorum, you will not receive credit for the invalid work, and 2) not all claimed credit turns into granted credit of the same value. For instance, CUDA apps always over-claim their credits due to a bug in the accounting of FLOPs, and if a CUDA result is compared to a CPU result from the same workunit, the granted credit is the lesser of the two.

Also bear in mind that once a result has met its quorum, the displayed result is removed after 24 hours on the website, but the credits remain in the account permanently.
ID: 926809 · Report as offensive
kcramer

Send message
Joined: 16 Aug 09
Posts: 1
Credit: 13,040
RAC: 0
United States
Message 926839 - Posted: 17 Aug 2009, 20:55:37 UTC

Hi,

I just downloaded seti@home yesterday and it keeps saying that there are no new jobs. I am successfully doing climate prediction as well, but seti is not working. below is a portion of my log:

8/16/2009 11:51:23 PM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/16/2009 11:51:23 PM|climateprediction.net|Not reporting or requesting tasks
8/16/2009 11:51:29 PM|climateprediction.net|Scheduler request completed
8/17/2009 12:07:34 AM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/17/2009 12:07:34 AM|climateprediction.net|Not reporting or requesting tasks
8/17/2009 12:07:40 AM|climateprediction.net|Scheduler request completed
8/17/2009 1:49:53 AM|SETI@home|Sending scheduler request: To fetch work.
8/17/2009 1:49:53 AM|SETI@home|Requesting new tasks
8/17/2009 1:49:58 AM|SETI@home|Scheduler request completed: got 0 new tasks
8/17/2009 1:49:58 AM|SETI@home|Message from server: (Project has no jobs available)
8/17/2009 4:38:20 AM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/17/2009 4:38:20 AM|climateprediction.net|Not reporting or requesting tasks
8/17/2009 4:38:25 AM|climateprediction.net|Scheduler request completed
8/17/2009 4:55:02 AM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/17/2009 4:55:02 AM|climateprediction.net|Not reporting or requesting tasks
8/17/2009 4:55:07 AM|climateprediction.net|Scheduler request completed
8/17/2009 9:28:06 AM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/17/2009 9:28:06 AM|climateprediction.net|Not reporting or requesting tasks
8/17/2009 9:28:11 AM|climateprediction.net|Scheduler request completed
8/17/2009 9:43:38 AM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/17/2009 9:43:38 AM|climateprediction.net|Not reporting or requesting tasks
8/17/2009 9:43:43 AM|climateprediction.net|Scheduler request completed
8/17/2009 2:18:31 PM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/17/2009 2:18:31 PM|climateprediction.net|Not reporting or requesting tasks
8/17/2009 2:18:36 PM|climateprediction.net|Scheduler request completed
8/17/2009 2:34:37 PM|climateprediction.net|Sending scheduler request: To send trickle-up message.
8/17/2009 2:34:37 PM|climateprediction.net|Not reporting or requesting tasks
8/17/2009 2:34:43 PM|climateprediction.net|Scheduler request completed
ID: 926839 · Report as offensive
deahan

Send message
Joined: 17 Aug 09
Posts: 1
Credit: 86
RAC: 0
United States
Message 926840 - Posted: 17 Aug 2009, 20:59:00 UTC - in response to Message 926839.  

the same thing is happening to me.
ID: 926840 · Report as offensive
Aurora Borealis
Volunteer tester
Avatar

Send message
Joined: 14 Jan 01
Posts: 3075
Credit: 5,631,463
RAC: 0
Canada
Message 926841 - Posted: 17 Aug 2009, 21:02:20 UTC - in response to Message 926839.  
Last modified: 17 Aug 2009, 21:02:59 UTC

As can be seen in many threads, work production hit a bottleneck this weekend. At the moment there is only a trickle going out. Hopefully the problem will be resolved during the Tuesday maintenance shutdown.
ID: 926841 · Report as offensive
Profile arkayn
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4438
Credit: 55,006,323
RAC: 0
United States
Message 927184 - Posted: 19 Aug 2009, 18:14:04 UTC

Carrying this over from the previous thread.

Vistro Wrote:

I have to visit every machine anyway to install. I have two flash drives so I can work on two machines at once, or have someone else help me. A staff member will be present for every install, for their purposes and mine.

My planned order is this:

1. Have staff member disable Fortres 101 to allow system changes
2. Insert flash drive
3. Install BOINC while simultaniously running CPU-Z. I should have to run CPU-Z once in the computer lab as they are all identical in every way.
4. Reboot if nessesary
5. Kill BOINC
6. Run batch file to attach then run BOINC
7. See step 5
8. Run optimized installer.
9. Run BOINC, leave manager open until I leave room, go on to next computer.

That can be reasonably done in 5 minutes per machine, right?



ID: 927184 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13797
Credit: 208,696,464
RAC: 304
Australia
Message 927186 - Posted: 19 Aug 2009, 18:17:02 UTC - in response to Message 927184.  

That can be reasonably done in 5 minutes per machine, right?

I'd think more like 10minutes per machine, at best.

Grant
Darwin NT
ID: 927186 · Report as offensive
Profile Vistro
Avatar

Send message
Joined: 6 Aug 08
Posts: 233
Credit: 316,549
RAC: 0
United States
Message 927198 - Posted: 19 Aug 2009, 18:28:58 UTC - in response to Message 927186.  

Ah. There are steps that will allow me to move on to a new machine so I can work on multiple computers and bring the average time down.

That is, if the tech weenie says yes. I know one heck of a lot more than this guy and he gets paid more than me!

Dude only know how to operate macs. He almost called in for a higher-up to come in for something and the staff said "Actually, we have someone better", and then they called me out of a timed test. To unfreeze something they could have just killed with the task manager.

So this guy naturally hates me. I asked him "do you even know your way around half of these machines" and he looks at me like I was satan himself and said, in the most rich and stuck-up voice I've ever heard and said "I prefer a macintosh"

I wanted to punch the guy out so hard. I hate macs, I prefer PCs, but I still know how to kill something on a mac!
ID: 927198 · Report as offensive
Profile arkayn
Volunteer tester
Avatar

Send message
Joined: 14 May 99
Posts: 4438
Credit: 55,006,323
RAC: 0
United States
Message 927210 - Posted: 19 Aug 2009, 19:02:23 UTC - in response to Message 927198.  

I prefer my Mac, but have no problem running a mixed network in my house.

ID: 927210 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 927239 - Posted: 19 Aug 2009, 19:57:15 UTC - in response to Message 927198.  

Ah. There are steps that will allow me to move on to a new machine so I can work on multiple computers and bring the average time down.

That is, if the tech weenie says yes. I know one heck of a lot more than this guy and he gets paid more than me!

Dude only know how to operate macs. He almost called in for a higher-up to come in for something and the staff said "Actually, we have someone better", and then they called me out of a timed test. To unfreeze something they could have just killed with the task manager.

So this guy naturally hates me. I asked him "do you even know your way around half of these machines" and he looks at me like I was satan himself and said, in the most rich and stuck-up voice I've ever heard and said "I prefer a macintosh"

I wanted to punch the guy out so hard. I hate macs, I prefer PCs, but I still know how to kill something on a mac!


If you want to trim down your install time to around a minute just copy the data folder from boinc that is installed somewhere and put it on your flash drive.

Depending on the OS you could even use autostart and have it run a bat file to copy the contents from the flash drive to the machine and do all the hookie pokie for you. I do something simular to this for my machines. As I reimage them sometimes several times a day.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 927239 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 927240 - Posted: 19 Aug 2009, 19:58:49 UTC - in response to Message 927198.  

I asked him "do you even know your way around half of these machines" and he looks at me like I was satan himself and said, in the most rich and stuck-up voice I've ever heard and said "I prefer a macintosh"

File this under "how to make friends and influence people."

There is an almost inexhaustible supply of idiots out there. Pointing them out can be fun, but they tend to get annoyed.

Under the wrong circumstances, demonstrating that someone is an idiot can be a serious career limiting move.

ID: 927240 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 927244 - Posted: 19 Aug 2009, 20:02:49 UTC - in response to Message 927184.  

Carrying this over from the previous thread.

Vistro Wrote:

I have to visit every machine anyway to install. I have two flash drives so I can work on two machines at once, or have someone else help me. A staff member will be present for every install, for their purposes and mine.

I wasn't actually talking about the first visit, but what happens if there is a new optimized science app. (or an entirely new app from the project).

In other words, the second, third or fourth visits.

If you optimize, and there is a new, urgent, must be updated science app., you'll have to visit every machine again to update them.

If you don't optimize, BOINC will handle the updates for you.

ID: 927244 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 927246 - Posted: 19 Aug 2009, 20:09:00 UTC - in response to Message 927244.  

Carrying this over from the previous thread.

Vistro Wrote:

I have to visit every machine anyway to install. I have two flash drives so I can work on two machines at once, or have someone else help me. A staff member will be present for every install, for their purposes and mine.

I wasn't actually talking about the first visit, but what happens if there is a new optimized science app. (or an entirely new app from the project).

In other words, the second, third or fourth visits.

If you optimize, and there is a new, urgent, must be updated science app., you'll have to visit every machine again to update them.

If you don't optimize, BOINC will handle the updates for you.


That is a good wishlist item. Process for 3rd party apps to update themselves.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 927246 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 927254 - Posted: 19 Aug 2009, 20:40:54 UTC - in response to Message 927246.  

Carrying this over from the previous thread.

Vistro Wrote:

I have to visit every machine anyway to install. I have two flash drives so I can work on two machines at once, or have someone else help me. A staff member will be present for every install, for their purposes and mine.

I wasn't actually talking about the first visit, but what happens if there is a new optimized science app. (or an entirely new app from the project).

In other words, the second, third or fourth visits.

If you optimize, and there is a new, urgent, must be updated science app., you'll have to visit every machine again to update them.

If you don't optimize, BOINC will handle the updates for you.


That is a good wishlist item. Process for 3rd party apps to update themselves.

What that means is that you'd have to have some kind of security mechanism so that BOINC could detect legitimate 3rd party application sites, avoid non-legitimate sites, and prevent BOINC from replacing legitimate applications with malware.

... and if someone tricked BOINC into upgrading improperly, they (the project and the University) would be in some way liable.

Now, I suspect that the third-party sites could build an update module into their apps and handle this outside BOINC, but it is a security issue.

Not everyone on the internet is nice, not everyone has good intentions.
ID: 927254 · Report as offensive
OzzFan Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Apr 02
Posts: 15691
Credit: 84,761,841
RAC: 28
United States
Message 927260 - Posted: 19 Aug 2009, 20:53:44 UTC - in response to Message 927174.  

What happens after you graduate and no longer have access to those computers that are using optimized apps that are deprecated at a later date?
ID: 927260 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 927274 - Posted: 19 Aug 2009, 21:35:09 UTC - in response to Message 927254.  

What that means is that you'd have to have some kind of security mechanism so that BOINC could detect legitimate 3rd party application sites, avoid non-legitimate sites, and prevent BOINC from replacing legitimate applications with malware.

... and if someone tricked BOINC into upgrading improperly, they (the project and the University) would be in some way liable.

Now, I suspect that the third-party sites could build an update module into their apps and handle this outside BOINC, but it is a security issue.

Not everyone on the internet is nice, not everyone has good intentions.


Is there a secure process now when updating the applications?

If someone took over a router or changed some dns server entries they could direct someone anywhere they like really.

Having a requirement like the apps being verisign signed could possibly work. Although $400/yr might be asking a bit much.

They could have a process for 3rd parties sending their apps to them. Then the updates would come from berkeley. Since they have loads of free time now. I'm sure they wouldn't have any problem managing that. :)
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 927274 · Report as offensive
1mp0£173
Volunteer tester

Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 927297 - Posted: 19 Aug 2009, 22:58:29 UTC - in response to Message 927274.  

What that means is that you'd have to have some kind of security mechanism so that BOINC could detect legitimate 3rd party application sites, avoid non-legitimate sites, and prevent BOINC from replacing legitimate applications with malware.

... and if someone tricked BOINC into upgrading improperly, they (the project and the University) would be in some way liable.

Now, I suspect that the third-party sites could build an update module into their apps and handle this outside BOINC, but it is a security issue.

Not everyone on the internet is nice, not everyone has good intentions.


Is there a secure process now when updating the applications?

If someone took over a router or changed some dns server entries they could direct someone anywhere they like really.

Having a requirement like the apps being verisign signed could possibly work. Although $400/yr might be asking a bit much.

They could have a process for 3rd parties sending their apps to them. Then the updates would come from berkeley. Since they have loads of free time now. I'm sure they wouldn't have any problem managing that. :)

The process now is reasonably secure. They are SETI@Home apps. on SETI@Home servers.

There is no practical way to "take over a router" and break that chain unless the router was very close to Berkeley. I can think of some nasty tricks to make trouble, but they're not very practical.

DNS is a better target, but if you're current on your patches, it's safe enough.

... and I think you'd have to hijack the whole zone, which is possible, but not very easy.

Verisign isn't very trustworthy, and that's a major flaw with code-signing.

I just don't see any project in any way "certifying" a 3rd party app. If we have to go someplace else to get them, then SETI@Home is not directly involved, and if we get something from someone dishonest, it's our fault, not theirs.
ID: 927297 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 . . . 13 · Next

Message boards : Number crunching : Panic Mode On (23) Server problems


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.