Posts by ohiomike

1) Message boards : BOINC client : ATI Detection with Windows 7, Boinc 6.10.17 (Message 28428)
Posted 31 Oct 2009 by ohiomike
Post:
Duh! I'd heard that and it didn't register. Thanks Jorg.
Now if they would just fix it. Or can I start the Boinc Client without starting the GUI? My headless machines (most of them) would like that. I guess I could set them up to do an auto-login, and then put Boinc.exe in the start menu (if Win7 still has one).
2) Message boards : BOINC client : ATI Detection with Windows 7, Boinc 6.10.17 (Message 28426)
Posted 31 Oct 2009 by ohiomike
Post:
Windows 7 Pro, 64 Bit Edition
Boinc 6.10.17 X64,
ATI Radeon HD 5750 not found:

10/31/2009 18:27:16 Starting BOINC client version 6.10.17 for windows_x86_64
10/31/2009 18:27:16 log flags: file_xfer, sched_ops, task
10/31/2009 18:27:16 Libraries: libcurl/7.19.4 OpenSSL/0.9.8k zlib/1.2.3
10/31/2009 18:27:16 Running as a daemon
10/31/2009 18:27:16 Data directory: V:\Boinc
10/31/2009 18:27:16 Running under account boinc_master
10/31/2009 18:27:16 Processor: 4 GenuineIntel Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz [Intel64 Family 6 Model 15 Stepping 11]
10/31/2009 18:27:16 Processor: 4.00 MB cache
10/31/2009 18:27:16 Processor features: fpu tsc pae nx sse sse2 pni
10/31/2009 18:27:16 OS: Microsoft Windows 7: x64 Edition, (06.01.7600.00)
10/31/2009 18:27:16 Memory: 3.19 GB physical, 6.90 GB virtual
10/31/2009 18:27:16 Disk: 4.00 GB total, 3.53 GB free
10/31/2009 18:27:16 Local time is UTC -4 hours
10/31/2009 18:27:16 No usable GPUs found

I would open a Trak ticket, but I can't find the page anymore.
3) Message boards : BOINC client : Boinc Linux/Samba/ntfs-3g Question (Message 14336)
Posted 15 Dec 2007 by ohiomike
Post:
I recently changed my back-up server over from Win2K to Linux. Due to Windows machines on my network, I have some that do back-ups via Samba -> ntfs-3g to a NTFS disk. What I have noticed is that Boinc (and it's client apps) hog the resources in spite of being started with "nice". A 10 GB back-up file takes 2 hours to write with Boinc running as opposed to about 6 minutes with Boinc suspended. Is this an issue with Boinc, Samba, ntfs-3g, or nice?
PS- I've gotten around the problem by having Boinc suspend for an hour while back-ups are scheduled.
4) Message boards : BOINC client : Feature request: limit processor affinity (smp machines) (Message 14202)
Posted 4 Dec 2007 by ohiomike
Post:
I am running two projects; seti@home that uses a modest amount of memory, and World Community grid that uses a large portion of memory. What I would like to be able to do is to set a processor affinity limit to a particular project in order to limit the amount of total memory that boinc is consuming. For example; set seti@home to only use processor #1 and WCG to processor #2. This would drastically improve swapping and memory consumption.


On a typical SMP mchine, where you also have shared-memory design, you won't really gain much from that. The only thing you will gain by 'locing' processes to particular CPUs is that when OS scheduller moves one process over to another processor, it won't have to flush and re-fill the L2 nd L1 caches.

Nothing there about memory consumption and swapping (as every single byte of information is identically accessible by both (all) processors).

Things are different if you start to talk about non-SMP multi-processor beasts (such as some clustered machines or huge mainframes, etc.). But then, there are special process schedullers run which really take care about CPU affinity on those beasts.

The flush/re-load is costly in terms of time! With more an more multiple core machines out there, I think this is something that needs to be looked into.

5) Message boards : BOINC Manager : Boinc Mgr in Fedora 8 (Message 13901)
Posted 16 Nov 2007 by ohiomike
Post:
that is really sad! Guess the 2 million or so users of Fedora are left out!

Tell Fedora people to make an official package. BOINC devs won't waste their time making packages for every single Linux distro in existence!

Maybe if the developers would just include a list of what dependencies are needed if would help. I am fighting the same problem with ARCH Linux. On the up-side, Boinc installs and runs fine, I just use the Boinc Manager from a remote machine to look at things.
6) Message boards : Web interfaces : Merge Error/Question (Message 13900)
Posted 16 Nov 2007 by ohiomike
Post:
I am having a problem merging a machine back to it's self after a crash/re-build. Cosmology, Boinc Server V 14210M.
Why can't I merge these two?

Oh... hold on, I see it now. The kernel version is different! That will hold you back from merging those computers. The OS has changed!


Thanks for the thought Ageless, that didn't do the trick though. I also re-named the machine in case the server SW is case sensitive- still no go.

Machines:
----------------------------------------------------------------------------------
4720 CES-Mac 2,516.76 316,035 GenuineIntel
Intel(R) Xeon(R) CPU 5150 @ 2.66 GHz [Family 6 Model 15 Stepping 6] Linux
2.6.23-MacPro.smpLL 165 16 Nov 2007 0:45:45 UTC

8596 CES-Mac 29.00 300 GenuineIntel
Intel(R) Xeon(R) CPU 5150 @ 2.66GHz [Family 6 Model 15 Stepping 6] Linux
2.6.23-MacPro.smpLL 101 16 Nov 2007 15:33:53 UTC

Msg on merge attempt:
----------------------------------------------------------------------------------
Sometimes BOINC assigns separate identities to the same computer by mistake. You can correct this by merging old identities with the newest one.


h2 1184664188 1195217985
h2 1184856122 1195220216
h2 1186161494 1195207882
h2 1189874794 1195225867
h2 1190848749 1195226276
h2 1191272284 1195224239
h2 1192294398 1195227007
h2 1193327720 1195227021
No hosts are eligible for merging with this one.
----------------------------------------------------------------------------------

[/quote]
7) Message boards : BOINC client : 5.10.20 - uploading but not reporting (Message 13165)
Posted 20 Oct 2007 by ohiomike
Post:
The send results definitely needs to be an option. I run between 6 and 11 machines full time. When I walk by and check the status with BoincView, the new version makes it look like multiple projects are down. I then had to either check each project by hand or do updates on all to see what was doing on. If I wanted a bunch of software making decisions for me, I would say so. Until SETI, Einstein, and Cosmo manage to stay up 24/7/365, I'm back to version 5.10.13-.
8) Message boards : API : Compile error question- AM_PROG_CC_C_O & XML_PARSER (Message 12702)
Posted 23 Sep 2007 by ohiomike
Post:
How did you download the code? What are the commands you used to compile? (especially the ./configure line)


a) Tried code from:
$ svn co http://boinc.berkeley.edu/svn/tags/boinc_core_release_5_10_8
$ svn co http://boinc.berkeley.edu/svn/tags/boinc_core_release_5_10_20
$ svn co http://boinc.berkeley.edu/svn/trunk/boinc
all 3 failed the same.

b) After $ ./_autosetup, config was run with:

$ ./configure --disable-server --enable-client
$ ./configure --disable-server --enable-client --enable-maintainer-mode
both caused errors by $ make.

rather than make a ridiculously long post, I placed the config & make outputs on the web:
config.log : config.log
make.log : make.log


9) Message boards : API : Compile error question- AM_PROG_CC_C_O & XML_PARSER (Message 12697)
Posted 23 Sep 2007 by ohiomike
Post:
Trying to compile boinc client/client_gui for linux x86_64. I think I must be missing something, but don't know what.
Happens on v 5.10.8 (my initial target), 5.10.20, and current. Any help would be appreciated.

Q1) ./autosetup returns:
sched/Makefile.am:177: compiling `md5.c' with per-target flags requires `AM_PROG_CC_C_O' in `configure.ac'

Q2) make returns:
boinc_client-acct_mgr.o: In function `ACCT_MGR_INFO::parse_login_file(_IO_FILE*)':
/MC/src/boinc/client/acct_mgr.C:611: undefined reference to `XML_PARSER::skip_unexpected(char const*, bool, char const*)'
boinc_client-acct_mgr.o: In function `ACCT_MGR_INFO::init()':
/MC/src/boinc/client/acct_mgr.C:649: undefined reference to `XML_PARSER::skip_unexpected(char const*, bool, char const*)'
boinc_client-acct_mgr.o: In function `AM_ACCOUNT::parse(XML_PARSER&)':
/MC/src/boinc/client/acct_mgr.C:246: undefined reference to `XML_PARSER::skip_unexpected(char const*, bool, char const*)'
boinc_client-acct_mgr.o: In function `ACCT_MGR_OP::parse(_IO_FILE*)':
/MC/src/boinc/client/acct_mgr.C:323: undefined reference to `XML_PARSER::skip_unexpected(char const*, bool, char const*)'
boinc_client-acct_setup.o: In function `CREATE_ACCOUNT_OP::handle_reply(int)':
.. then errors out
10) Message boards : BOINC Manager : linux boincmgr: No such file (Message 12663)
Posted 20 Sep 2007 by ohiomike
Post:


ls -al /MC/BOINC/boincmgr
-rwxr-xr-x 1 root root 8792428 2007-09-18 05:10 /MC/BOINC/boincmgr


Why is root the owner?

Did you install BOINC from the BOINC download pages or is it a package provided by your distro?

The other thing that catches my attention is the file size 8792428 whereas my self-compiled 6.1.0 only comes in at 3484720. The size difference could be explained by your version being statically compiled - ie ldd will report that it is not a dynamically linked executable.

run
 ls -al /MC/BOINC/ 
and see if root is the owner of all BOINC files - this may just be a simple permissions problem.


Ya, I own all the files.
There is no real user but root on my test machines, the su this and su that gets old real quick. They are all locked down tight so I don't worry about it.

I should have given some background on this problem:
a) Boinc 5.10.8, x86_64.
b) Same files worked when I was running Fedora Core 6 & 7. It has just been since the change to ARCH Linux that I've had the problems.
c) All patches current as of "now".
d) The i686 version runs fine on ARCH Linux i686, it is just the 64 bit version fighting with me.
It's like it is missing something, and refuses to tell me. The boinc client will always report what it is missing, I wish the manager would too.

I think I'll compile a version of 5.10.8 here, and see what I get.
Anybody have a link to the source that that was built on, I think the newest source I have is v 5.10.7, and the new stuff is ??
11) Message boards : BOINC Manager : linux boincmgr: No such file (Message 12642)
Posted 20 Sep 2007 by ohiomike
Post:
<snip>

G'day ohiomike

What's the output of
ls -al boincmgr 
?


ls -al /MC/BOINC/boincmgr
-rwxr-xr-x 1 root root 8792428 2007-09-18 05:10 /MC/BOINC/boincmgr
12) Message boards : BOINC Manager : linux boincmgr: No such file (Message 12590)
Posted 19 Sep 2007 by ohiomike
Post:
<snip>

The last stance is really weird. I can imagine a couple of reasons for it:

  • this particular executable somehow got invalid. As it works fine on your other linux boxen, you might try to copy it over
  • dynamic loader on your Arch Linux has some problems either with executable or some library that is required by boincmgr. Not much that you can do about it unless there are updates/upgrades available for your linux distribution
  • file somehow lost execute permissions. You can set them by executing command chmod 755 boincmgr



HTH.


1) I have tried a new copy from a working machine- no go.
2) I am trying to find out what library is causing the problem, I thought ldd would have found it, no go. I am going to try doing a static link of boincmgr and see if that will run.
3) First this I tried was chmod, no go.
13) Message boards : BOINC Manager : linux boincmgr: No such file (Message 12574)
Posted 18 Sep 2007 by ohiomike
Post:
I just installed Boinc x64, 5.10.8 on a Linux machine (Arch Linux). The client runs fine, and I can look at the tasks remotely, but when I try to look at things locally, I get:

boincmgr: No such file or directory

Logged in as root, boincmgr does exist, and permissions are set to 'Is executable'. ldd shows:

ldd: warning: you do not have execution permission for boincmgr, not a dynamic executable.

Anyone have any ideas on this?
PS- I running 5.10.8 on 3 other Linux machines (Fedora Core) and things are fine on those. It is just the Arch Linux installation that has the problem.
14) Message boards : BOINC client : What Happened to my CPU Type? (Message 8552)
Posted 5 Mar 2007 by ohiomike
Post:
It seems that sometime around V5.8.11 or 12 all my Linux based AMD machines changed from "AuthenticAMD AMD Athlon(tm) 64 X2 Dual Core Processor ..." to just "AuthenticAMD". The ClientState.xml files still show the correct thing, as does BoincManager and BoincView, but SETI and Einstien show just "AuthenticAMD".
15) Message boards : BOINC client : v5.9.0 suspend issue (Message 7672)
Posted 21 Jan 2007 by ohiomike
Post:
Thank you, I didn't realize that that forum existed.
16) Message boards : BOINC client : v5.9.0 suspend issue (Message 7669)
Posted 21 Jan 2007 by ohiomike
Post:
I'm aware of that. I'm playing around with a "fair benchmark" for Linux. I've been using the most recent code to make sure I'm not getting a "hacked" version of the source.
My main concern was whether the other "test" versions (ie 5.8.x) are seeing this also.
So far my tests have been encouraging (I split the benchmark code into stand-alone tasks, which cpu_benchmarks() now execute via a 'system' call rather than a 'fork'. By executing the tasks as stand-alone it eliminates the priority issues which are causing false (low) Linux benchmarks. The windows version of fairBench matches the Boinc benchmark almost exactly, while the Linux benchmark is about 40-50% higher then the Boinc benchmark (which is the true performance of the CPU).
(Note- isn't that the beauty of open source code?)
17) Message boards : BOINC client : v5.9.0 suspend issue (Message 7667)
Posted 21 Jan 2007 by ohiomike
Post:
Downloaded via cvs. OS=Linux 2.6.19. Results the same on 3 different computers (1 AMD X2, 1 P4C, 1 P4). Results the same compiled with gcc or icc.
18) Message boards : BOINC client : v5.9.0 suspend issue (Message 7651)
Posted 21 Jan 2007 by ohiomike
Post:
Has anyone else noticed a problem with version 5.9.0 when a task is suspended? When a task(Einstein) is suspended, it marks the task as complete and uploads it (even though it was only at x%, it jumps to 100% and uploads). This happens only on a suspend, the boinc client can be started and stopped without issue. (This also appears to happen with Leiden, so I don't think it is a WU issue).
19) Message boards : BOINC client : Benchmarks. Linux v Windows (Message 7547)
Posted 17 Jan 2007 by ohiomike
Post:
If anyone is interested, I have posted the code used on the web. I have a stand-alone version of the Boinc benchmark code that can be compiled under Linux or Windows. Suprising enough, the Windows version is real close to the Boinc benchmarks, the Linux is way off.
http://members.cox.net/castleengineering/
Click on the C Programming/Boinc link to grab the tar containing all files.
20) Message boards : BOINC client : Benchmarks. Linux v Windows (Message 7528)
Posted 16 Jan 2007 by ohiomike
Post:
Just as an added note. I just took the Whetstone code from Boinc V5.9.0 and converted it to stand-alone code. Test machine is a Linux machine with an AMD x2 4200+ 2.2 gHz, running at 2.75gHz. Compiled using gcc (standard Linux) and icc (Intel compiler with math pack).
Stock Boinc v5.9.0 compiled with gcc benchmarks at 3.521 mFP.
Stock Boinc v5.9.0 compiled with icc benchmarks at 4.069 mFP.
Stand-alone Whetstone compiled with gcc benchmarked at 4.499.
Stand-alone Whetstone compiled with icc benchmarked at 7.298.
So which one do we use? How accurate of a predictor it is depends on the app, the optimized SETI app I run is probably close to the 7.298. Einstein maybe also. QMC and others I don't know about.
???


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.