• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Rosetta@Home - 12/23 - 12/27: Stats

BlackMountainCow

BlackMountainCow

Diamond Member
:beer: Please welcome

ComputerManiac and
Wiz and
THE TERMINATOR and
Neurodog

to the TeAm :beer:

TA[R@H] is now 88 members strong.


Need some stats? Take a look at Free-DC, BOINC-Stats and BOINC-Synergy

Got some questions? Try the Official R@H FAQ (work in progress).



Yesterday's highest overall climber: Neurodog (21,241 spots)

Yesterday's highest TeAm climber: Neurodog (25 spots)



Yesterday's production for TA: 54,204 (day before: 61,003 = -6,799) :thumbsdown:

TA's total score: 1,764,675 creds

TA's world rank is: #6



Upcoming Threats: none

Upcoming Conquests: Catalyst (13.6 days)




Milestone Makers 12/23 - 12/27:
  • 27th:

    OhioDude passed 25,000 😎
    TA_GeoffS passed 10,000 😀
    aaronpriest passed 10,000 😀
    kcjeeper passed 5,000 😛
    Bespin Reactor Shaft passed 1,000 🙂
  • 26th:

    Sofa King passed 75,000 :laugh:
    Soni passed 1,000 🙂
    aceofspades230 passed 1,000
    rabidjade passed 1,000 🙂
    Shame passed 1,000 🙂
  • 25th:

    Dalephi passed 10,000 😀
    sduguid passed 5,000 😛
    Strikermike passed 1,000 🙂
    mondobyte passed 1,000 🙂
  • 24th:

    mrwizer passed 75,000 :laugh:
    jw.middleton passed 10,000 😀
    bjm129 passed 5,000 😛
  • 23th:

    TeAm Enterprise passed 250,000 :shocked:
    NesuD passed 5,000 😛
    Dain passed 1,000 🙂


Yesterday's top 15 producers:
  • Rebel Alliance (9,764 creds)
  • TeAm Enterprise (9,635 creds)
  • OhioDude (3,444 creds)
  • BadThad (2,818 creds)
  • TA_TheReaper (2,695 creds)

  • Dalephi (2,470 creds)
  • mrwizer (2,290 creds)
  • jw.middleton (2,100 creds)
  • Sofa King (1,310 creds)
  • TA_GeoffS (1,182 creds)

  • Christian Diepold (1,084 creds)
  • TA_JC (1,078 creds)
  • TA_EvilWobbles (1,073 creds)
  • sduguid (988 creds)
  • Insidious (969 creds)


Mini Races

Sorry, looks like I missed the end of the 75k race. But from reading up on the forums I learned that Peter retired from the race due to some probs with Rosetta on his PCs. I guess we'll just declare Sean aka mrwizer the winner of the mini race to 75k and Rise and Peter come in second 😉 But what about this race to 50K with Sid, JC and EvilWobbles? Is that a deal? All your RACs are pretty close and this would be a true race! :evil: Crazee PMed me that you're all in. So, here are the stats to start from:

  • #12 Insidious ... ... 39,611 (RAC: 975)
  • #13 TA_JC ... ... ... 38,408 (RAC: 1,073)
  • #14 TA_EvilWobbles .. 38,080 (RAC: 952)
Go, go, go! :thumbsup:



SitRep:
  • We passed BOINC Synergy and are already way ahead of them. For the record, We're leading US-D by over 200k now, BOINC Synergy by 135k. :evil: The distance to Catalyst is down to 13.6 (- 0.1 day) days)
  • Congrats to all milestone makers in the last couple of days. No milestone is better than another, but I just have to mention OhioDude for passing 25K in no time, Sofa King for passing 75k, mrwizer for passing 75k and winning the mini race and last ut not least,TeAm Enterprise for passing a quarter of a million points = 250,000 creds!
  • Due to the growth of TA[R@H] I expanded the top 10 producers to top 15. That's for you Sid 😉 j/k. Any comments/complaints/requests on that?
  • There are still some of these bad DEFAULT_xxxx_205_xxx WUs out there. I just received one. Looks like some of them got reissued somehow. So watch your queues! According to the Devs, they should be ALL gone by now, but some very few survived.
  • According to the R@H forums, there are quiete a lot of these "30 secs error out" WUs out there as well. If you're on a bandwith limit or a worried about your PC sitting idle, the Devs suggest you suspend R@H for the time being. I personally don't have any problem with these short WUs. Ok, they "waste" 30 secs of my CPU, but on an average day, that doesn't ammount to more than 5 Minutes. A "quick" round of BF2 will do my production more harm than those short WUs. But that's something everybody has to decide for his or herself. 🙂
  • In order to avoid problmes like these in the furture, the so called "Ralph" team (Rosetta alpha) was proposed. It's basically a closed alpha/beta testing group that get new apps/WUs before they go mainstream. "Ralph" is still just an idea, but one heavily appreciated by the Devs most likely to start soon.
  • Once more, thx to Crazee for stepping in while I was high for the first time of my life. Legally! 😛 But the pain is gone now, I got some brand new teeth and I'm back on the stats. Thx all! 🙂



:beer:😛:beer:

BMC
 
Congrats to the Milestone Makers. :beer:

Thanks for the thread, BlackMountainCow. :thumbsup:
 
WELCOME to Rosetta@Home ComputerManiac, Wiz, THE TERMINATOR and Neurodog (and Robor with TeAm Enterprise) Thank You for joining us! I hope you like it here! :beer: :beer:

ya take one little vacation...... (we tried to stay busy 😛 )

Milestones dropping like flies :thumbsup:

Thanks for the GREAT STATS and UPDATE BMC

and about this RACE..... I look at Free-DC this morning and see that JC piles 484 points in his first update since we started.... :shocked:

OK buddy.... get yer track shoes out of the closet..... IT's ON! :thumbsup: 😀

-Sid

(Thank you RA for approving my 3-day liberty! :beer: )

Crazee, Thanks for your efforts in BMCs absense... GREATLY APPRECIATED! :gift:

PS: I think top 15 is a TERIFFIC idea 😀
 
ooh, that looks like a good race :beer:

good to see you feeling better BMC.

its a sad day (for me) when i don't hit the top 15 :Q
 
I've had a lot of hung rosetta clients ... one or two machines might be acceptable but ... to unhang 30-40 machines is a royal pain ...

Anyone have a clue when things will get stable for Rosetta so I can go into low maintenance mode ... alternatively, maybe this isn't a mature enough project and I should buzz off ...

mondo
 
Originally posted by: mondobyte
I've had a lot of hung rosetta clients ... one or two machines might be acceptable but ... to unhang 30-40 machines is a royal pain ...

Anyone have a clue when things will get stable for Rosetta so I can go into low maintenance mode ... alternatively, maybe this isn't a mature enough project and I should buzz off ...

mondo
Click to expand...

They were hung?

That isn't normal. even if you were to get a 'Default xxxx 205 xxx' work unit, they don't hang machines. (they just crunch for about 10 hours then move on to the next work unit.... the project devs. posted credit WILL be given if it happens to you)

The other WUs that were messed up, crunch for about 20 seconds or less then move on to the next work unit.

I think there might be something else at work if your machines are hanging up. I'll try to help if you can post a little more detail.

-Sid

The only thing that was really affected was that dial-up users were having to connect pretty frequently to refil their queues to make up for the (extremely) fast turnaround when the 20 WUs were emptying so fast.

I have been hands off through all of this except to check queues for the 205 WUs and abort them. (but like I said above even if you didn't do that, your clients should not stop operating unless something else is wrong)
 
Originally posted by: mondobyte
I've had a lot of hung rosetta clients ... one or two machines might be acceptable but ... to unhang 30-40 machines is a royal pain ...

Anyone have a clue when things will get stable for Rosetta so I can go into low maintenance mode ... alternatively, maybe this isn't a mature enough project and I should buzz off ...

mondo
Click to expand...

There was a batch of work units that got out but those are out of the system or so they say. I haven't had a real problem with hung clients but there is some out there that abort after a few seconds and if you get a lot of them and you have your cache set too low you can run out of work but those are slowly working their way out of the system as well. The RA's cache is set to 3 days bty and as far as I know none of us have ran out of work yet althought it did get kind of close.
I would say check on the Rosetta forums but there are some real a$$holes over there these days ( Mostly ex Fadders )
I would hate to see you leave the project as we are really picking up speed and we could sure use your hordes against XPC, DPC, and FreeDC.
 
Originally posted by: Insidious
Originally posted by: mondobyte
I've had a lot of hung rosetta clients ... one or two machines might be acceptable but ... to unhang 30-40 machines is a royal pain ...

Anyone have a clue when things will get stable for Rosetta so I can go into low maintenance mode ... alternatively, maybe this isn't a mature enough project and I should buzz off ...

mondo
Click to expand...

They were hung?

That isn't normal. even if you were to get a 'Default xxxx 205 xxx' work unit, they don't hang machines. (they just crunch for about 10 hours then move on to the next work unit.... the project devs. posted credit WILL be given if it happens to you)

The other WUs that were messed up, crunch for about 20 seconds or less then move on to the next work unit.

I think there might be something else at work if your machines are hanging up. I'll try to help if you can post a little more detail.

-Sid

The only thing that was really affected was that dial-up users were having to connect pretty frequently to refil their queues to make up for the (extremely) fast turnaround when the 20 WUs were emptying so fast.

I have been hands off through all of this except to check queues for the 205 WUs and abort them. (but like I said above even if you didn't do that, your clients should not stop operating unless something else is wrong)
Click to expand...


If you did get some of the 205's you could easily think they are hung because they are so long and crunch so slow you can't even get one done by the return deadline.
 
Hey Sid.... you seem to be up-to-date on issues on the project.... I've got a few results over several machines that have significant credit claimed and no credit issued... do these get reviewed and credited?

For example:

Result ID 5158659
Name INCREASE_CYCLES_10_1dcj_226_1285_0
Workunit 4159919
Created 26 Dec 2005 15:25:09 UTC
Sent 27 Dec 2005 14:35:56 UTC
Received 27 Dec 2005 18:07:07 UTC
Server state Over
Outcome Client error
Client state Computing
Exit status -1073741819 (0xc0000005)
Computer ID 105469
Report deadline 24 Jan 2006 14:35:56 UTC
CPU time 12294.515625
stderr out
<core_client_version>5.2.13</core_client_version>
<message> - exit code -1073741819 (0xc0000005)
</message>
<stderr_txt>
***UNHANDLED EXCEPTION****
Reason: Access Violation (0xc0000005) at address 0x006F47D0 write attempt to address 0x0EA9F1C0
Exiting...
</stderr_txt>
Validate state Invalid
Claimed credit 21.8119003723207
Granted credit 0
application version 4.81

and

Result ID 5178269
Name INCREASE_CYCLES_10_1r69_226_1418_0
Workunit 4173225
Created 26 Dec 2005 21:06:02 UTC
Sent 27 Dec 2005 20:02:40 UTC
Received 28 Dec 2005 3:08:21 UTC
Server state Over
Outcome Client error
Client state Computing
Exit status -1073741819 (0xc0000005)
Computer ID 105973
Report deadline 24 Jan 2006 20:02:40 UTC
CPU time 17634.65625
stderr out
<core_client_version>5.2.13</core_client_version>
<message> - exit code -1073741819 (0xc0000005)
</message>
<stderr_txt>
***UNHANDLED EXCEPTION****
Reason: Access Violation (0xc0000005) at address 0x7C911E58 read attempt to address 0x3FB262C8
Exiting...
</stderr_txt>
Validate state Invalid
Claimed credit 27.5687379045808
Granted credit 0
application version 4.81

and

THE BIGGEST ONE

Result ID 5052221
Name DEFAULT_1di2_205_1_3
Workunit 3760455
Created 25 Dec 2005 4:28:39 UTC
Sent 25 Dec 2005 4:42:07 UTC
Received 26 Dec 2005 7:24:35 UTC
Server state Over
Outcome Client error
Client state Computing
Exit status -177 (0xffffff4f)
Computer ID 105973
Report deadline 22 Jan 2006 4:42:07 UTC
CPU time 89340.125
stderr out
<core_client_version>5.2.13</core_client_version>
<message>Maximum CPU time exceeded
</message>
<stderr_txt>
</stderr_txt>
Validate state Invalid
Claimed credit 142.184211617864
Granted credit 0
application version 4.81
 
I have a bunch of those as well Geoff. The staff has been on Holiday break and say they will sort it all out when they return ( Which will be next week I believe )

"The work units in batch 205 were set up to predict 1000 structures instead of 10, so they will all reach the run time limit of 12-16 hours before finishing and will eventually error out. WE WILL GRANT CREDIT TO PEOPLE WHO HAVE RUN AND ABORTED THESE WORK UNITS."

These are they only ones I have seen they say they will give credit for any other work units I don't know about.
 
Cool... thanks Jim... One of the make-it-or-break-it qualifiers for me for a project is whether or not I'm wasting my time...
 
Originally posted by: GeoffS
Cool... thanks Jim... One of the make-it-or-break-it qualifiers for me for a project is whether or not I'm wasting my time...
Click to expand...

I agree my friend.
I'm willing to be a bit more paitent with the Rosetta staff because they are on top of everything and as soon as they know of a problem they not only let the crunchers know they start working on a fix. They admit their mistakes and take steps so they don't make the same one twice.
I'm very impressed with the staff at Baker labs so far.
 
Thanks for the stats and welcome back, BMC.

Thanks for the special shout-out for my milestone, too! 🙂

BTW - I've gotta tooth that's killing me. Can you help me out? 😉🙂
 
Thanks BMC for stats.

I think once the holidays are over, things will get better over there at Rosetta. I think it was a lot to ask of them to get ambushed by SETI and FAD both ending at the same time and then a week later the holiday season started. I think things will smooth out soon enough.
 
Geoffs,

The '205' work unit you have is the one they say credit will be given. (I assume it will be delayed until they figure out just how to do that.)

I'm not sure about the others. The WUs that were 'bad' and errored out after a (very) few seconds will get no credit... I'm sure (it would only be like 0.1 points anyway because of the extremely short time they worked.

I am concerned about these two. I see they are from batch 226 which is well after the 'bad' WUs we have been talking about so much. (the developers felt that issue was solved for batch 220 and later)

And like you... I would not sit comfortably with that amount of work going uncredited unless it was the rare exception among the other work that was getting done OK.

There is the talk (in the Rosey forum) about some volunteers that are willing to be alpha testers for new batches and this will help eliminate bad WU batches in the future.. KUDOS to them!

I don't really have good advise. From past experience, the Rosetta developers seem to be VERY good at resolving these issues quickly, so it probably isn't time to bail yet. (at least I'm not considering it atm)

Hang in there buddy!

-Sid
 
I took a look at my logs (5 hosts) and see that I haven't crunched many WUs from the 226 batch yet. I found 3 that had crunched to completion successfully, so at least it isn't the entire 226 series... I'll poke around in the Rosey forums and keep watching my logs... If I find anything out, I'll let you know.

-Sid
 
The problem basically boils down to this: With 4.81 the DEVs introduced a new way of randomizing the search patterns: a random seed. This random number is generated and then applied to a WU. But only afterwards they found out, that some of these random seed are so random that they crash the WU; hence you have these short 30 sec WUs.

They are more than aware of the prob and I bet it's the first thing they'll work on once they get back from the holidays. As Sid said, the "Ralph" team will test WUs and new apps in the furute to make sure this doesn't happen again (see also SitRep).
 
Looks like we finally got XPC's eye
😀

"Looks like we should be into the top 3 within 3 weeks, how long we stay there is another thing. TeAm AnandTech are slightly out producing us, but not by much, hopefully we can increase output and stay ahead of them, other than that, there isn't as yet anyone else in the top 10 that are moving at a great rate of knots... "
 
Thanks guys... I don't really care much for the ones that error out in less than a minute... losing the 0.05 points doesn't concern me much... it's those other ones that got my attention... I'll check all of them to see if there are other significant differences between claimed and credited...

Thanks again... this TeAm is fantastic! :thumbsup:

Geoff
 
Originally posted by: Freewolf
Looks like we finally got XPC's eye
😀

"Looks like we should be into the top 3 within 3 weeks, how long we stay there is another thing. TeAm AnandTech are slightly out producing us, but not by much, hopefully we can increase output and stay ahead of them, other than that, there isn't as yet anyone else in the top 10 that are moving at a great rate of knots... "
Click to expand...

MUHUHAHAhahahaaha :evil:

 
not to flog a dead horse but-

12/28/2005 10:49:00 AM|rosetta@home|Starting result 1hz6A_topology_sample_207_15636_10 using rosetta version 481
12/28/2005 10:49:02 AM|rosetta@home|Started upload of DEFAULT_1r69_219_7151_0_0
12/28/2005 10:49:06 AM|rosetta@home|Finished upload of DEFAULT_1r69_219_7151_0_0
12/28/2005 10:49:06 AM|rosetta@home|Throughput 27255 bytes/sec
12/28/2005 10:49:13 AM|rosetta@home|Unrecoverable error for result 1hz6A_topology_sample_207_15636_10 ( - exit code -1073741819 (0xc0000005))
12/28/2005 10:49:13 AM||request_reschedule_cpus: process exited
12/28/2005 10:49:13 AM|rosetta@home|Computation for result 1hz6A_topology_sample_207_15636_10 finished
12/28/2005 10:49:13 AM|rosetta@home|Starting result 1ogw__topology_sample_207_5146_7 using rosetta version 481
12/28/2005 10:49:29 AM|rosetta@home|Unrecoverable error for result 1ogw__topology_sample_207_5146_7 ( - exit code -1073741819 (0xc0000005))
12/28/2005 10:49:29 AM||request_reschedule_cpus: process exited
12/28/2005 10:49:29 AM|rosetta@home|Computation for result 1ogw__topology_sample_207_5146_7 finished
12/28/2005 10:49:29 AM|rosetta@home|Starting result 1hz6A_topology_sample_204_11721_8 using rosetta version 481
12/28/2005 10:49:42 AM|rosetta@home|Unrecoverable error for result 1hz6A_topology_sample_204_11721_8 ( - exit code -1073741819 (0xc0000005))
12/28/2005 10:49:42 AM||request_reschedule_cpus: process exited

edit- sorry that was a lot lionger than i thought, took out a couple. anyway, its almoat all 207 topology for me.
 
You have to stay after school and make it up 😛

you owe us 25 seconds.... 😀

(I think these get recycled until a certain number of hosts all fail them.... we should be close to the end)
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top