Windows S5R4 App 6.07 available for Beta Test

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4273
Credit: 245184851
RAC: 13891
Topic 194065

A new Windows App is available from our Beta Test Page.

This should fix the checkpointing bug introduced in the previous 6.06 Beta App. Apart from that fix it should have the same properties as that release.

Switching from 6.04 - 6.06 and the 6.07 SSE Power App should be painless with the included app_info.xml.

Please test and report.

BM

BM

Wedge009
Wedge009
Joined: 5 Mar 05
Posts: 117
Credit: 15789955262
RAC: 7314048

Windows S5R4 App 6.07 available for Beta Test

Bloated download. Are you sure you meant to include two copies of the 6.06 release in the archive?

In any case, will let you know if anything goes wrong on my SSE host.

Soli Deo Gloria

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5845
Credit: 109880985942
RAC: 30611139

RE: Bloated download. Are

Message 88781 in response to message 88780

Quote:
Bloated download. Are you sure you meant to include two copies of the 6.06 release in the archive?

Yes, the archive is really bloated and I'm sure Bernd didn't mean to include the 6.06 zip archive and all the expanded 6.06 files inside as well. I guess he'll fix it shortly so in the meantime just discard everything else except the five 6.07 files and the app_info.xml.

I've just installed these six files only on an SSE host that had been running the flawed 6.06 version where the checkpoint couldn't be updated. Before stopping BOINC I forced a checkpoint "refresh" by deleting the .cpt file in the appropriate slots directory. As soon as I saw a new checkpoint file get created, I shut down BOINC and moved the 6.07 files into position. On restarting BOINC, the "refreshed" checkpoint was read and the new app didn't lose any of the previous crunching (~50% completed). New checkpoints are being created at regular intervals so all seems in order.

Thanks very much to Bernd for fixing this over the weekend.

Cheers,
Gary.

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4273
Credit: 245184851
RAC: 13891

The archive should be ok now.

The archive should be ok now. Sorry, was the common "whatever *" instead of "whatever*" typo.

BM

BM

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5845
Credit: 109880989275
RAC: 30611467

RE: The archive should be

Message 88783 in response to message 88782

Quote:

The archive should be ok now. Sorry, was the common "whatever *" instead of "whatever*" typo.

BM

No problems - all looks good now, thanks.

I'm busily converting SSE hosts from 6.06 to 6.07 and all looks good so far.

Cheers,
Gary.

Wedge009
Wedge009
Joined: 5 Mar 05
Posts: 117
Credit: 15789955262
RAC: 7314048

RE: Sorry, was the common

Message 88784 in response to message 88782

Quote:
Sorry, was the common "whatever *" instead of "whatever*" typo.


No apologies necessary. I meant the post as a notification rather than a complaint. :)

I didn't use 6.06 at all, I moved from 6.05 on SSE2-capable hosts and from 6.04 on SSE-only CPUs. Either way, no apparent problems so far.

Soli Deo Gloria

Byron S Goodgame
Byron S Goodgame
Joined: 16 Jan 06
Posts: 187
Credit: 56581
RAC: 0

Ran a couple of the 6.07 on

Ran a couple of the 6.07 on my AMD64 3800+. Times are about the same as the previous app as expected. No issues have sprung up.

I've started them on my Intel E2180 dual core now, so we'll see how that goes. Don't expect any problems there either.

Byron S Goodgame
Byron S Goodgame
Joined: 16 Jan 06
Posts: 187
Credit: 56581
RAC: 0

Had an error on one of the

Had an error on one of the tasks I was runing on the Intel dual core. Had an Exit status -1 (0xffffffffffffffff) which I'm pretty sure had to do with my graphics card having a fit during a video game. I guess I won't be doing that again while running tasks. Seems strange it only affected one of the tasks but I'm not complaining. I noticed it right after it happned when the graphics for the game crashed I checked on Boinc. It ran for a couple of minutes then gave a compute error.

Bernd Machenschalk
Bernd Machenschalk
Moderator
Administrator
Joined: 15 Oct 04
Posts: 4273
Credit: 245184851
RAC: 13891

RE: Had an error on one of

Message 88787 in response to message 88786

Quote:
Had an error on one of the tasks I was runing on the Intel dual core. Had an Exit status -1 (0xffffffffffffffff) which I'm pretty sure had to do with my graphics card having a fit during a video game. I guess I won't be doing that again while running tasks. Seems strange it only affected one of the tasks but I'm not complaining. I noticed it right after it happned when the graphics for the game crashed I checked on Boinc. It ran for a couple of minutes then gave a compute error.

I don't think it had anything to do with the actual graphics. The relevant line of stderr is
2008-12-01 05:25:36.6964 [CRITICAL]: Checksum error: -2681955
i.e. a checksum error in the checkpoint. Looks like a filesystem problem that might have to do something with the game. I'd run a filesystem check on the disk ASAP just to be sure.

Another aspect that I currently consider as a (minor) bug is that the app returned with exit code -1. Actually it _should_ in this case exit with the return value of MAIN, which in this case would be 10.

BM

BM

Byron S Goodgame
Byron S Goodgame
Joined: 16 Jan 06
Posts: 187
Credit: 56581
RAC: 0

RE: I don't think it had

Message 88788 in response to message 88787

Quote:

I don't think it had anything to do with the actual graphics. The relevant line of stderr is
2008-12-01 05:25:36.6964 [CRITICAL]: Checksum error: -2681955
i.e. a checksum error in the checkpoint. Looks like a filesystem problem that might have to do something with the game. I'd run a filesystem check on the disk ASAP just to be sure.

Another aspect that I currently consider as a (minor) bug is that the app returned with exit code -1. Actually it _should_ in this case exit with the return value of MAIN, which in this case would be 10.

BM


Well since I don't have a Vista OS disk and only a Recovery disk for this system I'm not sure how much I can accomplish running it. but I ran sfc /scannow and this came up a couple of times.

Quote:
2008-12-01 06:23:39, Info CSI 0000012d [SR] Cannot repair member file [l:24{12}]"settings.ini" of Microsoft-Windows-Sidebar, Version = 6.0.6001.18000, pA = PROCESSOR_ARCHITECTURE_INTEL (0), Culture neutral, VersionScope = 1 nonSxS, PublicKeyToken = {l:8 b:31bf3856ad364e35}, Type neutral, TypeName neutral, PublicKey neutral in the store, hash mismatch
2008-12-01 06:23:43, Info CSI 0000012f [SR] Cannot repair member file [l:24{12}]"settings.ini" of Microsoft-Windows-Sidebar, Version = 6.0.6001.18000, pA = PROCESSOR_ARCHITECTURE_INTEL (0), Culture neutral, VersionScope = 1 nonSxS, PublicKeyToken = {l:8 b:31bf3856ad364e35}, Type neutral, TypeName neutral, PublicKey neutral in the store, hash mismatch
2008-12-01 06:23:43, Info CSI 00000132 [SR] Could not reproject corrupted file [ml:520{260},l:72{36}]"\??\C:\Program Files\Windows Sidebar"\[l:24{12}]"settings.ini"; source file in store is also corrupted
2008-12-01 06:23:45, Info CSI 00000134 [SR] Verify complete

and the final message was

Quote:
2008-12-01 06:26:21, Info CSI 00000192 [SR] Verify and Repair Transaction completed. All files and registry keys listed in this transaction have been successfully repaired


Here's a copy of the whole thing if anyone feels like looking at it sfcdetails
I saw the mention of the sidebar which I don't use, so I brought it up and it seems to work fine. I then ran sfc /scannow again and the same error came up again. I was trying to see if there was another repair option but I'm never asked for a cd or dvd during the process, which I thought would happen if needed.

This was the only message referring to an error/problem all else checks out.
Nor sure what I can do about this or if a Vista disk is even needed to repair it.

Edit: One other thing I guess I should mention, right after I first checked on Boinc after the video crash when the task was still running, I opened up Boinclogx (I've just started using it recently) and when it came up before the error happend, it showed the task as inactive. When the task had the error I was in the middle of closing it down thinking it needed to be restarted. I'm wondering it it's possible BoinclogX could have something to do with the error too. After running the scan and just before this edit I was bringing it up again and got the same inactive message but closed it right after seeing it and the tasks are still running. I guess it's a long shot but I hadn't noticed it doing that before, and it was running as active before I turned it and Boinc off to run the scan.

paul milton
paul milton
Joined: 16 Sep 05
Posts: 329
Credit: 35825044
RAC: 0

byron, i have an hp pavillion

Message 88789 in response to message 88788

byron, i have an hp pavillion with only recovery cd's does your system have a "recovery partition" i.e. C: = file system D:= recovery. if so then the system should be set up to look in D: as if it was the OS cd.

seeing without seeing is something the blind learn to do, and seeing beyond vision can be a gift.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.