Just too many problems for me to deal with. I am not qualified to comment on this very fine project and will continue to support this BOINC project in the future when the problems are put to bed. In the mean while I will support a number of other projects.

---

Just too many problems for me to deal with. I am not qualified to comment on this very fine project and will continue to support this BOINC project in the future when the problems are put to bed. In the mean while I will support a number of other projects.

Yep understandable. But you might want to come back already soon. There are betatests running with an improved application which autoaborts stuck units. Overall it does what it should also you can't expect no failing WUs at all. This is cutting-edge-research with real up-to-date science and therefore some failures have to be accepted. Btw you will get credit for your stuck WUs after a week or so.

regards tralala

P.S.: It is always a wise decision not to accumulate huge WU-Caches when just joining a project. In fact it is never advisable unless you only have sporadic internet connection time.

---

Btw you will get credit for your stuck WUs after a week or so.

Will the same thing happen with all the jobs that complete successfully but finish with 0 CPU time ? Not the most important thing, but it would be nice.

---

Just too many problems for me to deal with. I am not qualified to comment on this very fine project and will continue to support this BOINC project in the future when the problems are put to bed. In the mean while I will support a number of other projects.

It looks like you had several very long "hung work units" (they ran for 3 & 4 days, vs his 2hr preference, before he found and aborted them).

I know software guys are always promising "the next release". But they are presently testing changes on Ralph (I've seen them) which should be available on Rosetta in the next couple of days that will prevent these "hung work units" problems from, well, getting you hung up.

With the next release, these will work their way through the system normally, reporting what models they did complete successfully, and the fact that it seemed to run for an extended period (an hour) of time without progress and so it was ended.

This will help the project resolve any future problems with the application or the work units, and help participants like yourself not to lose 4 days of crunch time, which further helps the project get more valid models crunched and you to get more credits.

Please check the applications page in a week and see if we are over 5.03, and that will mean the fix is in place.

You may also want to consider a longer runtime for each work unit. This means you crunch more models against a single protein before you return the results, and so less upload/download bandwidth. You do this in your Rosetta preferences. Do it before you have a large cache of work, or ratchet it up only a little at a time so your cache of 20 x 2hr work doesn't become 20 x 24hr work instantly.

P.S. thanks for not leaving "in flames" the way some do. It just shows you how messed up some people are. Hope to see you back crunching in a week.

---

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/

P.S. thanks for not leaving "in flames" the way some do. It just shows you how messed up some people are. Hope to see you back crunching in a week.

Although I agree, I think it's easy to leave with a smile when you're just here for a week. People who you are blaiming for being "messed up" have a little longer history here... so breaking up is a little harder for them (just like in the real world).

---

Also, it's easy to crunch on just 2-4 PCs within your immediate reach (like I do) and a COMPLETELY different experience for people who operate large networks of dozens of PCs with BOINC running as a service, spread across different locations not in their physical reach, in which case locating and aborting stuck WUs can become really frustrating.

And it's the latter group which brings the big TeraFLOPS.

---

Best UFO Resources
Wikipedia R@h
How-To: Join Distributed Computing projects that benefit humanity

Just too many problems for me to deal with. I am not qualified to comment on this very fine project and will continue to support this BOINC project in the future when the problems are put to bed. In the mean while I will support a number of other projects.

It looks like you had several very long "hung work units" (they ran for 3 & 4 days, vs his 2hr preference, before he found and aborted them).

I know software guys are always promising "the next release". But they are presently testing changes on Ralph (I've seen them) which should be available on Rosetta in the next couple of days that will prevent these "hung work units" problems from, well, getting you hung up.

With the next release, these will work their way through the system normally, reporting what models they did complete successfully, and the fact that it seemed to run for an extended period (an hour) of time without progress and so it was ended.

This will help the project resolve any future problems with the application or the work units, and help participants like yourself not to lose 4 days of crunch time, which further helps the project get more valid models crunched and you to get more credits.

Please check the applications page in a week and see if we are over 5.03, and that will mean the fix is in place.

You may also want to consider a longer runtime for each work unit. This means you crunch more models against a single protein before you return the results, and so less upload/download bandwidth. You do this in your Rosetta preferences. Do it before you have a large cache of work, or ratchet it up only a little at a time so your cache of 20 x 2hr work doesn't become 20 x 24hr work instantly.

P.S. thanks for not leaving "in flames" the way some do. It just shows you how messed up some people are. Hope to see you back crunching in a week.

Ok, I'll be looking for that 5.03 upgrade.

---

Btw you will get credit for your stuck WUs after a week or so.

Will the same thing happen with all the jobs that complete successfully but finish with 0 CPU time ? Not the most important thing, but it would be nice.

That's a problem that cropped up with Win98 and WinME - which I believe Rom was looking into. Ask for an update in the "Give credit where credit is due" thread.

---

Btw you will get credit for your stuck WUs after a week or so.

Will the same thing happen with all the jobs that complete successfully but finish with 0 CPU time ? Not the most important thing, but it would be nice.

That's a problem that cropped up with Win98 and WinME - which I believe Rom was looking into. Ask for an update in the "Give credit where credit is due" thread.

Thanks for the reply Benny.

I just read that thread and I thought that a script would have been being run by now ?

Its not worth the hassle and cost. I just wont bother running Rosetta on those machines anymore (About 5.5 Ghz) :shrug icon here:

---

Thanks for the reply Benny.

I just read that thread and I thought that a script would have been being run by now ?

Its not worth the hassle and cost. I just wont bother running Rosetta on those machines anymore (About 5.5 Ghz) :shrug icon here:

They do run the script... periodicly. (I think once a week, because the 0 cpu time claims from my Win98 pc have been corrected (granted 20 credits) twice the last two weeks).

So if the average score of your pc's would be about 20 credits it is worth to keep them crunching.

---

Yep, Thanks. The script was run later on. :thumbsup emotie here:

---

We are running the credit granting script every 24 hours now.

Good to know. Thanks David.

---