Message boards :
Cafe SETI :
time outs on work units?
Message board moderation
Author | Message |
---|---|
dasy2k1 Send message Joined: 9 Jul 05 Posts: 65 Credit: 118,948 RAC: 0 |
do seti@home WUs have a deadline timeout after wich you will not recive the credit? i am asking thisas my einstein@home running on the same boinc warned me taht i was 1.58 days late and wouldent get any credit (in teh messages tab) needless to say as it was only 60% done and haveng used 40MB (seti uses about 5mb per WU) of disk space to do this and taken 3 days CPU time i aborted and have unatached einstein for teh moment... do seti progects do this? |
MJKelleher Send message Joined: 1 Jul 99 Posts: 2048 Credit: 1,575,401 RAC: 0 |
Look on the work tab in BOINC manager and it will show the deadline You can drag the edge of that column (in the heading) to see the whole "Report Deadline", not to mention the exact time that it's due. SETI has a deadline of 14 days after the work unit is assigned to you, ditto LHC. Eintein and Protein have 7 day deadlines. CPDN is something like 11 months. I don't know about the other current and upcoming projects. For more information than you'd ever really want to know about BOINC and the projects using it, check out the Wiki, created and managed by the inimitable Paul D. Buck. MJ |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
Think it's 10 days from the time you download a work unit to the And reported!! Uploading Boinc units is a two-way thing, you upload them and have to report them. But the general deadline is 14 days on Seti units. As far as I understood from JM7, the deadline is given out at the moment the unit is sent to you, so from that moment onwards you have 14 days to crunch it, upload it and report it. |
dasy2k1 Send message Joined: 9 Jul 05 Posts: 65 Credit: 118,948 RAC: 0 |
have now checked with a new work unit i just got on a machine i have just started boionic at and its definatly 14 days!!! as i am going on holiday soon i think i will set all machines to no new work as it takes me on average a week to compleate oen WU (i am only 17 and muy mum dousent like teh computers to be left on all night!!!) |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
You can type slower though. We don't have a time limit of 14 seconds to get the message in. ;) Boionic... now there's a new one. :) |
mikey Send message Joined: 17 Dec 99 Posts: 4215 Credit: 3,474,603 RAC: 0 |
Think it's 10 days from the time you download a work unit to the To expand a little, if a unit is not returned before the 14 day cycle expires it is automatically sent out to new computers for crunching, I think this can happen upto 7 times but is adjustable by Berkeley. IF you as a unser that got the unit the FIRST time around returns it BEFORE the next round of people start to retun the same unit, you WILL get credit. If however you return it AFTER any of the next round of people you will get no credit. Berkeley has started sending units out for the second and third and fourth, etc. times to computers that can crunch a unit in about one day, so you must be REAL FAST to get credit!! MOST of the time you will NOT get credit, once you are late you are usually REAL LATE!!! |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
If however you return it AFTER any of the next round of people you will get no credit. Berkeley has started sending units out for the second and third and fourth, etc. times to computers that can crunch a unit in about one day, so you must be REAL FAST to get credit!! MOST of the time you will NOT get credit, once you are late you are usually REAL LATE!!! That's bullshit. Sorry to say it like that, but only if there are a lot of computers returning errors, like the 4.13 clients do, then no one will get credit. Else it doesn't matter. First 3 in, they get the middle credit. Everyone else after that gets the same credit that the first 3 got. You only don't get credit as a 4rd or 5th if you return a unit with an error. Or outside the deadline. Please don't scare people like that. |
dasy2k1 Send message Joined: 9 Jul 05 Posts: 65 Credit: 118,948 RAC: 0 |
yeh now taht i have started bionc on anotehr of my machines which is much faster it seems to be running at 1% in just over 1 minute!!! and the graphics are smooth!! not like teh jerks in my other machine! my 3rd machine (a linux box) which can only run boinc in comand line wont retrive work units, it sais that it has got a sucessfull scheduler request then "No work from project" then defers comunication with project for x minutes y seconds.. what is going on here? |
MJKelleher Send message Joined: 1 Jul 99 Posts: 2048 Credit: 1,575,401 RAC: 0 |
my 3rd machine (a linux box) which can only run boinc in comand line wont retrive work units, it sais that it has got a sucessfull scheduler request Like this? 7/22/2005 7:30:45 PM|SETI@home|Sending scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi 7/22/2005 7:30:45 PM|SETI@home|Requesting 4448 seconds of work, returning 0 results 7/22/2005 7:30:46 PM|SETI@home|Scheduler request to http://setiboinc.ssl.berkeley.edu/sah_cgi/cgi succeeded 7/22/2005 7:30:46 PM|SETI@home|No work from project 7/22/2005 7:30:47 PM|SETI@home|Deferring communication with project for 3 hours, 18 minutes, and 7 seconds That's not a problem with your computer, it's that the project just doesn't have enough work to keep everybody fully occupied. Though if you check the Technical News page, you'll see that they're working on a solution, at least for now. The long term solution to keeping your computer busily crunching data is to attatch to more than one project. The intent is that, when one project is out of work, another will be able to use your cycles. There are several projects out there, happy to use your CPU if you will only let them. 8-) MJ |
dasy2k1 Send message Joined: 9 Jul 05 Posts: 65 Credit: 118,948 RAC: 0 |
yep... exactly like that though it tends to say that it has no work 2 times :-/ i have just halted that machine for the night as it was defering for 23H 59M 59S i may as well! give teh elctricity bill a bit of a break! mabie i will attach my einstein prodj to it then it can choose between both... i wish i could see some progress meter though even if its just a simple comand line output that says like 10% done 20% done etc |
MJKelleher Send message Joined: 1 Jul 99 Posts: 2048 Credit: 1,575,401 RAC: 0 |
mabie i will attach my einstein prodj to it then it can choose between both... i wish i could see some progress meter though even if its just a simple comand line output that says like 10% done 20% done etc I like Einstein as a project, who knows, maybe ET is communicating with gravity waves... Check for the system requirements on the project, though. I don't know if the minimum specs are different for Linux vs. Windows, but Einstein says it needs 70Meg of memory, and your computer shows about 61. MJ |
Jim Baize Send message Joined: 6 May 00 Posts: 758 Credit: 149,536 RAC: 0 |
What is your setting for "connect ever x days"? if it is a value too large it can cause problems also. Jim do seti@home WUs have a deadline timeout after wich you will not recive the credit? |
mikey Send message Joined: 17 Dec 99 Posts: 4215 Credit: 3,474,603 RAC: 0 |
If however you return it AFTER any of the next round of people you will get no credit. Berkeley has started sending units out for the second and third and fourth, etc. times to computers that can crunch a unit in about one day, so you must be REAL FAST to get credit!! MOST of the time you will NOT get credit, once you are late you are usually REAL LATE!!! Okay I think I didn't make it clear enough then! The 3rd, 4th time etc, is realting to the 3rd and 4th, etc. times it is sent out in 2 week intervals. NOT the 3rd , 4th, etc. computer, the 3rd, 4th, etc. 2 week cycles! I was expanding the idea of not enough computers returning results to get 3 of them to agree, with the credit number range. If only 2 of the original 4 computers return results but the 3rd and 4th computers do not the unit is sent out for a 2nd time. If those computers do not return a "valid" result then it is sent out for a 3rd, 4th, etc. time until 3 results, or more if the validator is slow, are within the acceptable range. I just found out this AM the resending can occur a max of 9 times before the unit is flagged for Berkley's attention. |
©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.