This is a continuation of another thread.

I have been having problems with my laptop backing off comunications with BOINC for a week. This happens when I am working on my laptop but not connected to the internet for hours at a time. When BOINC fails to find a scheduler four consecutive times, it backs off for an entire week. Here are the final messages:

12/20/2006 8:36:33 AM|rosetta@home|Scheduler request failed: couldn't resolve host name
12/20/2006 8:36:33 AM|rosetta@home|Deferring scheduler requests for 1 minutes and 0 seconds
12/20/2006 8:37:33 AM|rosetta@home|Fetching scheduler list
12/20/2006 8:37:38 AM|rosetta@home|Scheduler list fetch failed: http error
12/20/2006 8:37:38 AM|rosetta@home|4 consecutive failures fetching scheduler list - deferring 604800 seconds
12/20/2006 8:37:38 AM|rosetta@home|Deferring scheduler requests for 1 weeks, 0 days, 0 hours, 0 minutes and 0 seconds

This is the BOINC version of evil itself!!

I did change the settings to => Run based on preferences, but that didn't help. The messages above are the latest result from that experiment. I didn't used to have this problem, not sure when it started

---

(Click for detailed stats)

Just thought, since this is a boinc problem you could use the new help system
http://boinc.berkeley.edu/help.php

If you want to go to the man at the top, look for David Anderson ;-)

You'll need skype (which is free).

all I can think of is there is a boinc bug occuring.

(i.e could be fixed by using the current development version 5.8.x)

---

Team mauisun.org

So what BOINC version is doing that?

If you basically only have network access when you are at the PC, you might work around the problem by suspending network activity when you leave your machine. Then get connected and select that the network is always available. This is done in your Activity tab.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

So what BOINC version is doing that?

If you basically only have network access when you are at the PC, you might work around the problem by suspending network activity when you leave your machine. Then get connected and select that the network is always available. This is done in your Activity tab.

The version is 5.4.9. Actually, I have network access when I am at home and not at school doing homework and such for class. I am suspending network activity as you suggest, and seeing what happens over the next five or six hours. If this works, I will have to remember to go to the activity dialog when I return and plug in again.

---

(Click for detailed stats)

So what BOINC version is doing that?

The version is 5.4.9.

You might try 5.4.11 Gerry and see if the problem persists.

So what BOINC version is doing that?

The version is 5.4.9.

You might try 5.4.11 Gerry and see if the problem persists.

It seems that Feet's idea seems to have done the trick: I've been watching to see if the same problem has been replicated, and it has not returned after nearly twelve hours of crunching. I think I will have to remember to go to the activity dialog when I return and plug in again. Apart from that, if I remember correctly, the only difference between 5.4.9 and 5.4.11 is a bug fix to a Norton Antivirus problem.

---

(Click for detailed stats)

the only difference between 5.4.9 and 5.4.11 is a bug fix to a Norton Antivirus problem.

Close.

Mcafee

By disabling the network access, you're just preventing BOINC from making those attempts to connect, and thus avoid the 4 strikes and you're out for a week. I still do not understand that one. Mine does not behave that way. You might flag the project for "no new tasks", work through what you've got, get it all reported, and then reset the project. Perhaps there's some historical data that makes BOINC think your history warrents such a one week delay. The project reset should get you back to settings with no such history (i.e. more like everyone else).

If that doesn't change it, since you have one more thing to remember, you want to insulate yourself a bit so that you continue to crunch even if you forget once and a while to enable BOINC to connect to the network. You can do this simply by increasing your "cache" of work. This is done in General Preferences by increasing the "connect to network every ... days" setting. If you have 1.0 days of work for example, then you would have time to forget one day when you get back from class, and still have work to keep you busy through morning.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

By disabling the network access, you're just preventing BOINC from making those attempts to connect, and thus avoid the 4 strikes and you're out for a week. I still do not understand that one. Mine does not behave that way. You might flag the project for "no new tasks", work through what you've got, get it all reported, and then reset the project. Perhaps there's some historical data that makes BOINC think your history warrents such a one week delay. The project reset should get you back to settings with no such history (i.e. more like everyone else).

If that doesn't change it, since you have one more thing to remember, you want to insulate yourself a bit so that you continue to crunch even if you forget once and a while to enable BOINC to connect to the network. You can do this simply by increasing your "cache" of work. This is done in General Preferences by increasing the "connect to network every ... days" setting. If you have 1.0 days of work for example, then you would have time to forget one day when you get back from class, and still have work to keep you busy through morning.

Will try and see what happens. :)

---

(Click for detailed stats)

Actually, after I did as you suggested, I thought the problem might be fixed after I unplugged R@H for about 7 hours to see what the messages were going to look like. Everything seems to work fine: it just kept backing off for a few hours at most, usually 30 minutes to a couple of hours. My cache, by the way, has been 1 day for probably over a year now. Sure enough, now that E@H is off line, now the problem is back with E@H: 4 failures and now I can't crunch for a week unless I reset or manually communicate like below in this thread.

I was thinking this morning that it might be that once boinc sees that the servers are not communicating, as with E@H now, they go into a "project is down" mode, meaning they query the server once per week until the project is up again, much like P@H. Don't know why this is so for my computer in such a short period of time, though. About the only thing I can think of is to uninstal/reinstall. But then again, since there is a new version coming out soon, I think I will wait to do that when I do the upgrade on that host. Any other suggestions?

---

(Click for detailed stats)

Actually, after I did as you suggested, I thought the problem might be fixed after I unplugged R@H for about 7 hours to see what the messages were going to look like. Everything seems to work fine: it just kept backing off for a few hours at most, usually 30 minutes to a couple of hours. My cache, by the way, has been 1 day for probably over a year now. Sure enough, now that E@H is off line, now the problem is back with E@H: 4 failures and now I can't crunch for a week unless I reset or manually communicate like below in this thread.

I was thinking this morning that it might be that once boinc sees that the servers are not communicating, as with E@H now, they go into a "project is down" mode, meaning they query the server once per week until the project is up again, much like P@H. Don't know why this is so for my computer in such a short period of time, though. About the only thing I can think of is to uninstal/reinstall. But then again, since there is a new version coming out soon, I think I will wait to do that when I do the upgrade on that host. Any other suggestions?

Give 5.8.0 a go now ;-)

---

Team mauisun.org

Actually, after I did as you suggested, I thought the problem might be fixed after I unplugged R@H for about 7 hours to see what the messages were going to look like. Everything seems to work fine: it just kept backing off for a few hours at most, usually 30 minutes to a couple of hours. My cache, by the way, has been 1 day for probably over a year now. Sure enough, now that E@H is off line, now the problem is back with E@H: 4 failures and now I can't crunch for a week unless I reset or manually communicate like below in this thread.

I was thinking this morning that it might be that once boinc sees that the servers are not communicating, as with E@H now, they go into a "project is down" mode, meaning they query the server once per week until the project is up again, much like P@H. Don't know why this is so for my computer in such a short period of time, though. About the only thing I can think of is to uninstal/reinstall. But then again, since there is a new version coming out soon, I think I will wait to do that when I do the upgrade on that host. Any other suggestions?

Give 5.8.0 a go now ;-)

I'm sort of queezy about test versions. Is the new version pretty stable now?

---

(Click for detailed stats)

I'm an official alpha tester, I've not seen anything that would be considered a "showstopper". I am checking out an issue with 5.7.5 for linux though.

At worst you just uninstall and reinstall the previous version. No work should be lost, so you risk practically nothing in giving it a "go" as it were.

tony

I run 5.8.0 since it was released, I didn't notice any bugs so far. It's stable.

---