hello,
i have install on one of my Servers
"boinccmd, built from BOINC 6.4.5"
and set 2 day work puffer.
OS: Suse Linux 10.3
i found in http://setiathome.berkeley.edu/hosts_user.php that this server have old workunits, deadline 15 Jan, and new ones with deadline in february.

now boinc runs every time the newest one and the old ones still in puffer. Why i have this problem with this Server?
to resolve this problem i set the next days --nomorework but this is not the best resolution

---

LanOpa
Dabei seit 18.12.2008

Are these WU's on your PC? If they arent in your BOINC/projects/setiathome folder then they may be Ghost WU's

or did you have an old install that had WU's left on it. The new BOinc won't see the old WU's if you installed in a new folder or in a different location. The only way I know of doing the old WU's is to reinstall in the old location, run until they are gone then delete the old BOINC files.

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

...now boinc runs every time the newest one and the old ones still in puffer.

Which buffer? Do you see them in BOINC Manager's Tasks view or only on the website?

Why i have this problem with this Server?
to resolve this problem i set the next days --nomorework but this is not the best resolution

If you are speaking of tasks in the Tasks view, then it is no problem! As your header line states correctly, it's a FIFO buffer, and it is perfectly normal that tasks with an earlier deadline arrive later than others with a deadline farther in future.

You obviously confuse deadline dates and arrival dates (which are not shown, you have to search for them in the log files).

This question arises often. If you do a search (upper left corner on this page) for "deadline", you'll find plenty, including answers.

Gruß,
Gundolf

[edit] Sorry, you probably don't use BOINC Manager, but there are also commandline options to retrieve the list of local tasks. [/edit]

---

Computer sind nicht alles im Leben. (Kleiner Scherz)

SETI@home classic workunits 3,758
SETI@home classic CPU time 66,520 hours

I should have looked before... I didnt see anything unusual on your linux machines. and I agree you may get Wu's that have an earlier return date then older WU's the ones with the nearer expiration date will usually go first if they are running high priority

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope

there are no changes on this computer, only the changes on this page here up to 2 days workspace.

http://setiathome.berkeley.edu/results.php?hostid=4719042

Taskid -> Send -> my report -> state
1115443380 -> 7 Jan 2009 11:15:35 UTC -> 9 Jan 2009 11:58:29 UTC -> Over
2 days work around, it's ok, i have normaly 2-3 days.

next
Taskid -> Send -> my report -> state
1115651931 -> 7 Jan 2009 16:23:09 UTC -> 30 Jan 2009 14:51:47 UTC -> In progress
in progress, ok is a new one i think today it is ready.

next on next page:
Taskid -> Send -> my report -> state
1110211023 -> 2 Jan 2009 6:07:14 UTC -> 25 Jan 2009 4:41:22 UTC -> In progress
ups, is older the my first example task, nearly 7 days saved lokal!
and there are more on page 2 and 3 of this server task.

---

LanOpa
Dabei seit 18.12.2008

ok, there are a heavy problem
i have only 2 wu on my Server saved but there more in task list by this site.

i haven't install boinc new, i think this is a problem by the new Year down.

is it possible to relaod this files or must this files timeout?

---

LanOpa
Dabei seit 18.12.2008

I'd just let it run out since its in all likelihood that your Seti client wont see the additional WU's if you move them over to the current folder

---

In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope