Posts by steveplanetary

1) Message boards : Questions and problems : dcf_debug message (Message 30682) Posted 25 Jan 2010 by steveplanetary Post: That helps. Thanks for the info. I may have other questions later, however. I have to do some calculations first.
2) Message boards : Questions and problems : dcf_debug message (Message 30680) Posted 25 Jan 2010 by steveplanetary Post: Having solved the problem of getting a working cc_config.xml file (with lots of help), I got my first dcf_debug message following the completion of an Einstein task: 1/24/2010 2:52:07 PM Einstein@Home [dcf] DCF: 7.401496->7.194387, raw_ratio 5.330402, adj_ratio 0.720179 Questions: What are raw_ratio and adj_ratio, and how are they calculated?
3) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 30630) Posted 23 Jan 2010 by steveplanetary Post: Gundolf Jahn, your epiphany was right on. Saving the text as ANSI was the key...I had been using Unicode. Between your last post and yesterday I had given up on having a cc_config.xml file. I never thought I would say I gave up on a computing problem. Some problems can be solved quickly, others take longer, but the reward is great. That's what makes computing so addictive. And thank you, Ageless, for all your help, as well as everybody else for your contributions. After saving the config file to the desktop I dragged and dropped it into the BOINC data directory and read the config file. It now works perfectly: 1/23/2010 6:33:46 AM log flags: file_xfer, sched_ops, task, dcf_debug Thanks again to everybody. Steve
4) Message boards : Questions and problems : CPU Time is less than Elapsed Time (Message 30586) Posted 18 Jan 2010 by steveplanetary Post: Thanks for the information. In posting this reply I'm having a different problem, though. After typing the message and clicking on Preview nothing happens. In other words, I don't get to see the preview. But when I edited the message Preview worked.
5) Message boards : Questions and problems : CPU Time is less than Elapsed Time (Message 30558) Posted 17 Jan 2010 by steveplanetary Post: This is a problem I have only had with Einstein@Home, and I have reported it there, but it could have something to do with the way E@H software runs on BOINC, so I'm reporting it here too. I'm using BOINC v6.10.18. Here are the particulars: 1/17/2010 1:18:09 AM Einstein@Home Restarting task h1_1098.70_S5R4__286_S5R6a_1 using einstein_S5R6 version 301 CPU Time: 10:11:24 Elapsed Time: 09:48:18 Edit: To be precise, I'm using S5R6sse
6) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29436) Posted 14 Dec 2009 by steveplanetary Post: I really appreciate your patience with me, Ageless, as well as the input by others. But to be certain no stone was left unturned, I did run chkdsk c: /f at reboot after reading your last post, although it didn't seem to be necessry, since /x also includes the functionality of /f. I was not made aware of any errors. I don't know how to check that Control Return Line Feeds after each end tag in cc_config.xml is active, but I can ensure you that I pressed Enter after each end tag. In my last copy/print activity I just copied the text provided by Gundolf Jahn on 11 Dedember. My current activity is reinstalling Window from c:\i386, since I know my OS contains corruption, and my install CD was lost in a fire in 2005. Would you know anything about reinstalling the OS from c:\i386? We could PM each other if you feel that topic doesn't belong in this message board. Steve
7) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29381) Posted 11 Dec 2009 by steveplanetary Post: I did exactly what you suggested, even creating cc_config.xml in a stepwise manner, using the cc_config.xml you provided. I then saved it to the BOINC Data file, and ran Read config file, which I present below: 12/11/2009 1:18:59 PM Re-reading cc_config.xml 12/11/2009 1:18:59 PM Missing start tag in cc_config.xml 12/11/2009 1:18:59 PM Re-read config file 12/11/2009 1:18:59 PM log flags: file_xfer, sched_ops, task One nice thing this time is that no line was in red. Steve
8) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29365) Posted 11 Dec 2009 by steveplanetary Post: Here's what I did: I ran chkdsk c: /r /x when I restarted the computer. You will recall the meanings of the optional parameters; /r Locates bad sectors and recovers readable information. The disk must be locked. /x Use with NTFS only. Forces the volume to dismount first, if necessary. All open handles to the drive are invalidated. /x also includes the functionality of /f. /f Fixes errors on the disk. The disk must be locked. If chkdsk cannot lock the drive, a message appears that asks you if you want to check the drive the next time you restart the computer. You will also recall the following example (from Help and Support): If you want to check the disk in drive D and have Windows fix errors, type: chkdsk d: /f If it encounters errors, chkdsk pauses and displays messages. Chkdsk finishes by displaying a report that lists the status of the disk. When I ran chkdsk c: /r /x it never paused and displayed messages. When chkdsk finished, the report it displayed was so fleeting it could not be read. I then booted into Windows, deleted cc_config.xml, and ran a third party app called PC Matic. Among other things, PC Matic performs defragmentation. Defragmentation involves both moving and consolidating files. I then copied cc_config.xml to the desktop using Notepad, then copied it to the BOINC Data directory. When I restarted BOINC the same error message was written. Given the lack of any error messages when I ran chkdsk, I have to assume there were no errors. And given the file moving and consolidation that occurs during defragmentation, I think the probability that cc_config.xml was written to the exact sector as before was minimized. Actually, if I understand how the OS works, saving the file to another directory, then copying it to the BOINC data directory didn't actually move the file, it just changed the file's starting and ending location in the Master File Table (MFT). After doing all this I ran chkdsk c: from the command line in Windows, and got the following report: CHKDSK is verifying files (stage 1 of 3)... File verification completed. CHKDSK is verifying indexes (stage 2 of 3)... Index verification completed. CHKDSK is recovering lost files. CHKDSK is verifying security descriptors (stage 3 of 3)... Security descriptor verification completed. CHKDSK is verifying Usn Journal... Usn Journal verification completed. CHKDSK discovered free space marked as allocated in the master file table (MFT) bitmap. CHKDSK discovered free space marked as allocated in the volume bitmap. Windows found problems with the file system. Run CHKDSK with the /F (fix) option to correct these. 39062015 KB total disk space. 21612508 KB in 92518 files. 24488 KB in 6372 indexes. 0 KB in bad sectors. 151979 KB in use by the system. 4096 KB occupied by the log file. 17273040 KB available on disk. 4096 bytes in each allocation unit. 9765503 total allocation units on disk. 4318260 allocation units available on disk. But I have to take into consideration the following information from Help and Support: If you run chkdsk without the /f command-line option on an active partition, it might report spurious errors because it cannot lock the drive. I am truely grateful for the help individuals have extended to me on this message board, but I have to ask for feedback on what I have reported in this post. Steve
9) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29353) Posted 10 Dec 2009 by steveplanetary Post: One simple question: After I copy your cc_config.xml to Notepad and save it to the desktop, should I copy cc_config.xml from the desktop and paste it to the BOINC Data directory, or should I drag and drop it into the directory?
10) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29338) Posted 10 Dec 2009 by steveplanetary Post: I really appreciate your time and effort, Ageless, but I am confused. When I created cc_config.xml, I did so using Notepad. I then saved it to the BOINC data directory. I thought that was where it was supposed to go. So I don't know what you mean by Save it wherever you want to, just not in your BOINC Data directory And I never copied or saved any text from IE. Sometimes I use IE just to examine the contents of a file. When finished doing that I just close IE. Every time I have created and/or edited cc_config.xml I have immediately saved it to the BOINC data directory. And I don't understand what you mean by Just copy this file to over there, overwriting the one that is already there. I could overwrite or delete cc_config.xml from it's current location and paste it somewhere else, but you'll have to tell me where.
11) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29329) Posted 9 Dec 2009 by steveplanetary Post: I did what you said. In Notepad there were no blank lines in front of the text, and no hyphens. I saved the file in the BOINC data directory, then opened cc_config.xml. There is a blank line above the text, and hyphens before the first two lines of the text, but these are just for expanding and contracting the lines that come after them. If I click on the hyphen before <cc_config> the hyphen turns into a +, and all the other lines of text disappear, leaving +<cc_config>. Even after I clicked on the bar at the bottom of the IE header to allow scripts and activex controls to run (I don't normally allow it) and again clicked on Read config file, I got the same result.
12) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29319) Posted 9 Dec 2009 by steveplanetary Post: My brain must have taken a short vacation. Here is the 1st line of my messages tab after I created cc_config.xml, and restarted BOINC: 12/8/2009 4:08:42 PM Missing start tag in cc_config.xml I do know where the BOINC data directory is, though (C:\Documents and Settings\All Users\Application Data\BOINC), and I don't have any hidden directories. I just can't explain this error message. Here is my cc_config.xml file. <cc_config> <log_flags> <sched_op_debug>0</sched_op_debug> <dcf_debug>1</dcf_debug> </log_flags> </cc_config>
13) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29313) Posted 8 Dec 2009 by steveplanetary Post: Thanks, everybody, for the explanations. I'll try it and see how it goes. But I still have a question (and obviously a lot to learn): Where do I read these log files? Steve
14) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29256) Posted 5 Dec 2009 by steveplanetary Post: Pardon my ignorance, but I can't find cc_config.xml, and I don't know what you mean by 'switch on /'.
15) Message boards : Questions and problems : Transitory Nature of DCF Changes (Message 29206) Posted 4 Dec 2009 by steveplanetary Post: Formerly, when I found the duration correction factor (DCF) for a project to be too large, which resulted in the initial time to completion (TTCi) for the just completed WU being greater than the final CPU time (CPUf), I would do the following: I would exit BOINC, open client_state.xml using Notepad, make note of DCF for that project, and multiply that DCF (DCF0) by the ratio of CPUf/TTCi, giving a new DCF (DCF1), with which I would edit client_state.xml, and which I assumed would be used when I allowed a new WU to download. What I found, however, if I immediately exited BOINC following download and opened client_state.xml, was that DCF was greater than DCF1, and in fact was closer to DCF0. This resulted in TTCi for the new WU to be too large, and not reflective of the CPUf for that WU. So, apparently, BOINC ignored DCF1 and calculated it's own DCF, which was a much smaller change from DCF0. This recalculation of DCF by BOINC appears to occur throughout the processing of the WU, as can be seen by exiting BOINC and opening client_state.xml at various times during the processing of the WU. At present, I am not making any changes to DCF, since it is futile. I should have known, since "Resistance is futile", but the scientist and frustrated mathematician in me longs for an accurate TTCi. Can anybody tell me how to conquer this beast? Steve Edit: I didn't mean to post this on Questions and Problems, so I have also posted it on Projects.
16) Message boards : Questions and problems : Editing client_state.xml/duration_correction_factor (Message 27093) Posted 5 Sep 2009 by steveplanetary Post: Just one more question... I still have the ORIGINAL 41 page .txt version of the client_state.xml file that I copied and pasted from the BOINC directory into Notepad. Is there any way to convert this back into .xml, thereby returning to the beginning of this mess? Steve
17) Message boards : Questions and problems : Editing client_state.xml/duration_correction_factor (Message 27089) Posted 5 Sep 2009 by steveplanetary Post: Correction: I just noticed that a tab in IE is a reproduction of client_state_prev.xml that is 15 pages in length. It has a DCF of 2.553912, as opposed to 7.969075 in the original client_state.xml file. Regards, Steve
18) Message boards : Questions and problems : Editing client_state.xml/duration_correction_factor (Message 27088) Posted 5 Sep 2009 by steveplanetary Post: I have backed up c"\documents and settings\all users\application data\boinc\ to c:\BOINC Backup 2009-09-05. I assume that's what you meant by "your whole BOINC data directory", as opposed to c"\program files\boinc. But I still have the orginal client_state.xml that I copied to Notepad. It is 41 pages in length. I haven't made a line-by-line comparison of the current client_state.xml with client_state_prev.xml, but they are both just 11 pages. It would be nice if I could reproduce the original client_state.xml file from the Notepad .txt copy, but I don't know of any way to do that. So I don't know what the best course of action would be now. Therefor, I still need help. Yours in anguish, Steve
19) Message boards : Questions and problems : Editing client_state.xml/duration_correction_factor (Message 27086) Posted 5 Sep 2009 by steveplanetary Post: The only way I was able to change the DCF was to open client_state.xml, highlight the entire file, copy it, and paste it to Notepad. I then modified DCF and did Save As..., saving it to c:\documents and settings\application data\boinc\client_state.xml. When I performed Save As..., I selected Save as type: All Files, Encoding: ANSI. I had exited BOINC and closed the browser that had displayed the orginal file client_state.xml. When I then navigated to c:\documents and settings\application data\boinc\ via My Computer and clicked on client_state.xml IE launched and displayed the following message: The XML page cannot be displayed Cannot view XML input using style sheet. Please correct the error and then click the Refresh button, or try again later. -------------------------------------------------------------------------------- Invalid at the top level of the document. Error processing resource 'file:///C:/Documents and Settings/All Users/Applicatio... - <client_state> ^ Please advise. Thank you, Steve
20) Message boards : Questions and problems : Editing client_state.xml/duration_correction_factor (Message 27082) Posted 5 Sep 2009 by steveplanetary Post: 1) The Time to Completion of newly downloaded cosmology@home WUs is always way too high. After a relatively short computation period this gets downgraded to a more reasonable time. If I changed the DCF would the initial TTC be more accorate? More importantly, would modifying the DCF result in susequent client recalculations of TTC to be underreported? Also, I doen't know how to edit client_state.xml for any reason, since Application Data in the path C:\Documents and Settings\All Users\Application Data\BOINC\ has Read-only not only checked but greyed out. 2) How do I get email notification of replies to my posts? Any help would be very much appreciated. Steve

Next 20

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.