Until we got the newer WU types, wasn't able to report on any problems with them obviously, and could only report on the older types... Sorry us testers weren't able to catch this problem before it started rolling out...

I disagree of the quoted above

I started the thread at ralph@home announcing the new version 4.97 to test
on 7 April at 09:37 UTC

by 12:52 UTC 7 April, I have already reported this error on Windows 4.97

by 00:24 UTC 8 April, Son Goku posted that 4.97 was working fine

After that time ... 8 April, is that 4.97 go to rosetta@home ...


I wonder why 4.95 that was working very well ... fixed several problems.
is not what was placed into rosetta@home instead -:(
http://www.fadbeens.co.uk/phpBB2/viewtopic.php?t=53&start=165

Now, that was rolled back to 4.83, I know why I crunched all the day
w/o completing only one WU of 4.98 into two of my pcs

-> see my signature ... my rac is failing down on rosetta -:(

So, I will STOP crunching to rosetta again, until a new version that
checkpoint enough to allow swapping apps removing from ram, comes in.

4.95 was that version !!! 4.96 too

In fact the information that I have is that Rosetta version 4.97 WAS in fact RALPH version 4.95. The version number was all that changed when it was implemented for Rosetta. What was not known at the time was how the newer WUs would react in the production environment. What is interesting here is that the RALPH testers are usually running BOINC version 5.2.32 and most Rosetta users are running BOINC 5.2.13. This may be part of the issue.

In any case you are wrong about what was implemented in Rosetta. While the version number for the Rosetta application is different, it is the same application that was working well in RALPH. RALPH version 4.97 is not what was deployed in Rosetta.

The workunit testing in RALPH did not show any problems with the newer workunits, however RALPH is a VERY limited subset of the types of systems and configurations running in Rosetta. Because of this fact it is not possible to test every possible issue before new work unit types are deployed.

As has been pointed out on may occasions, a number of your systems are below the minimum memory requirements for the project. The ones that are not, are reporting a significant portion of the memory as not available for Rosetta to use. This single fact has been and will continue to be the largest problem facing you in running Rosetta or Ralph. There are almost no problems reported for systems running with more memory unless a batch of bad work units comes along, and that will happen from time to time.

Click signature for global team stats

I guess I need to turn on the lights when I type. I meant to say BOINC 5.2.28. Rom had asked all the RALPH testers top upgrade to this version for improved error reporting. I think almost all of them performed the upgrade.

In any case it would be interesting to see if this had any impact on what has happened over the last day or so.

[b]"I'm trying to maintain a shred of dignity in this world." - Me[/b]

I guess you mean the 5.3.28 version? :-) Maybe some more light is needed? ;-)

That was the one Rom asked us to upgrade to. And it is pretty stable, runs fine on my computer.

It seems like BOINC v5.3.31 is latest, to see the BETA versions one can use the URL

http://boinc.berkeley.edu/download.php?dev=1

whereas the "official" stable versions are at

http://boinc.berkeley.edu/download.php

I guess I need to turn on the lights when I type. I meant to say BOINC 5.2.28. Rom had asked all the RALPH testers top upgrade to this version for improved error reporting. I think almost all of them performed the upgrade.

In any case it would be interesting to see if this had any impact on what has happened over the last day or so.

I did run some Ralph WU's on a 5.2.13 Boinc Client and they finished fine.