FreeBSD w. Linux emulation - should it work?
22 Jan 2005 21:19:46 UTC
Topic 187303
(moderation:
Hi all,
I'm new to both Boinc and Einstein@Home. I'm trying to run on a couple of FreeBSD boxes, using the Linux emulation and the latest (4.16) binary client from Berkeley, as pointed to by the E@H dowload page.
The client seems to startup fine, talks to the server and starts computing, but it outputs errors ("RP is an idiot"?!) and doesn't seem to upload anything. (Output included below.) Has anyone tried anything similar? I'm running FreeBSD 4.10-release, with a pretty fresh ports tree - I'm also including some port version info below. Any clues would be greatly appreciated.
Here's the ouput:
[pre]
djaa[software/boinc]% ./boinc_4.16_i686-pc-linux-gnu
2005-01-22 13:33:02 [---] Starting BOINC client version 4.16 for i686-pc-linux-gnu
Enter the URL of the project: http://einstein.phys.uwm.edu
You should have already registered with the project
and received an account key by email.
Paste the account key here: XXXXXXXXXXXXXXXXXXXXXXXXX
2005-01-22 13:33:30 [http://einstein.phys.uwm.edu/] Project prefs: using your defaults
2005-01-22 13:33:30 [http://einstein.phys.uwm.edu/] Host ID not assigned yet
2005-01-22 13:33:30 [---] No general preferences found - using BOINC defaults
2005-01-22 13:33:30 [---] Running CPU benchmarks
2005-01-22 13:34:32 [---] Benchmark results:
2005-01-22 13:34:32 [---] Number of CPUs: 0
2005-01-22 13:34:32 [---] nan double precision MIPS (Whetstone) per CPU
2005-01-22 13:34:32 [---] nan integer MIPS (Dhrystone) per CPU
2005-01-22 13:34:32 [---] Finished CPU benchmarks
2005-01-22 13:34:33 [---] Insufficient work; requesting more
2005-01-22 13:34:33 [---] Insufficient work; requesting more
2005-01-22 13:34:33 [http://einstein.phys.uwm.edu/] Requesting 17280 seconds ofwork
2005-01-22 13:34:33 [http://einstein.phys.uwm.edu/] Sending request to scheduler: http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
2005-01-22 13:34:34 [http://einstein.phys.uwm.edu/] Scheduler RPC to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
2005-01-22 13:34:34 [Einstein@Home] General preferences have been updated
2005-01-22 13:34:34 [---] General prefs: from Einstein@Home (last modified 2005-01-22 13:31:34)
2005-01-22 13:34:34 [---] General prefs: no separate prefs for home; using yourdefaults
2005-01-22 13:34:34 [Einstein@Home] Project prefs: no separate prefs for home; using your defaults
2005-01-22 13:34:34 [Einstein@Home] Started download of einstein_4.73_i686-pc-linux-gnu
2005-01-22 13:34:34 [Einstein@Home] Started download of einstein_4.73_i686-pc-linux-gnu.so
2005-01-22 13:34:39 [Einstein@Home] Finished download of einstein_4.73_i686-pc-linux-gnu.so
2005-01-22 13:34:39 [Einstein@Home] Throughput 168835 bytes/sec
2005-01-22 13:34:39 [Einstein@Home] Started download of earth
2005-01-22 13:34:46 [Einstein@Home] Finished download of einstein_4.73_i686-pc-linux-gnu
2005-01-22 13:34:46 [Einstein@Home] Throughput 138003 bytes/sec
2005-01-22 13:34:46 [Einstein@Home] Started download of sun
2005-01-22 13:34:47 [Einstein@Home] Finished download of sun
2005-01-22 13:34:47 [Einstein@Home] Throughput 242198 bytes/sec
2005-01-22 13:34:47 [Einstein@Home] Started download of Config_Test02
2005-01-22 13:34:47 [Einstein@Home] Finished download of Config_Test02
2005-01-22 13:34:47 [Einstein@Home] Throughput 1352 bytes/sec
2005-01-22 13:34:47 [Einstein@Home] Started download of H1_0069.9
2005-01-22 13:34:55 [Einstein@Home] Finished download of earth
2005-01-22 13:34:55 [Einstein@Home] Throughput 176271 bytes/sec
2005-01-22 13:35:36 [Einstein@Home] Finished download of H1_0069.9
2005-01-22 13:35:36 [Einstein@Home] Throughput 249833 bytes/sec
2005-01-22 13:35:36 [Einstein@Home] Starting result H1_0069.9__0070.2_0.1_T08_Test02_2 using einstein version 4.73
RP is an idiot: the assumption xTemp >= 0 is not generally true!!
2005-01-22 13:36:43 [Einstein@Home] Unrecoverable error for result H1_0069.9__0070.2_0.1_T08_Test02_2 (process exited with code 255 (0xff))
2005-01-22 13:36:43 [Einstein@Home] Unrecoverable error for result H1_0069.9__0070.2_0.1_T08_Test02_2 (process exited with code 255 (0xff))
2005-01-22 13:36:43 [Einstein@Home] Deferring communication with project for 1 minutes and 0 seconds
2005-01-22 13:36:43 [Einstein@Home] Deferring communication with project for 1 minutes and 0 seconds
2005-01-22 13:36:43 [Einstein@Home] Computation for result H1_0069.9__0070.2_0.1_T08_Test02 finished
linux: 'ipc' typ=258 not implemented
2005-01-22 13:37:44 [---] Insufficient work; requesting more
2005-01-22 13:37:44 [Einstein@Home] Requesting 172800 seconds of work
2005-01-22 13:37:44 [Einstein@Home] Sending request to scheduler: http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi
2005-01-22 13:37:45 [Einstein@Home] Scheduler RPC to http://einstein.phys.uwm.edu/EinsteinAtHome_cgi/cgi succeeded
2005-01-22 13:37:45 [Einstein@Home] Starting result H1_0069.9__0070.1_0.1_T09_Test02_3 using einstein version 4.73
RP is an idiot: the assumption xTemp >= 0 is not generally true!!
2005-01-22 13:49:17 [Einstein@Home] Unrecoverable error for result H1_0069.9__0070.1_0.1_T09_Test02_3 (process exited with code 255 (0xff))
2005-01-22 13:49:17 [Einstein@Home] Unrecoverable error for result H1_0069.9__0070.1_0.1_T09_Test02_3 (process exited with code 255 (0xff))
2005-01-22 13:49:17 [Einstein@Home] Deferring communication with project for 1 minutes and 0 seconds
2005-01-22 13:49:17 [Einstein@Home] Deferring communication with project for 1 minutes and 0 seconds
2005-01-22 13:49:17 [Einstein@Home] Computation for result H1_0069.9__0070.1_0.1_T09_Test02 finished
linux: 'ipc' typ=258 not implemented
2005-01-22 13:49:17 [Einstein@Home] Starting result H1_0069.9__0070.2_0.1_T09_Test02_2 using einstein version 4.73
RP is an idiot: the assumption xTemp >= 0 is not generally true!!
2005-01-22 13:57:25 [Einstein@Home] Unrecoverable error for result H1_0069.9__0070.2_0.1_T09_Test02_2 (process exited with code 255 (0xff))
2005-01-22 13:57:25 [Einstein@Home] Unrecoverable error for result H1_0069.9__0070.2_0.1_T09_Test02_2 (process exited with code 255 (0xff))
2005-01-22 13:57:25 [Einstein@Home] Deferring communication with project for 1 minutes and 0 seconds
2005-01-22 13:57:25 [Einstein@Home] Deferring communication with project for 1 minutes and 0 seconds
2005-01-22 13:57:25 [Einstein@Home] Computation for result H1_0069.9__0070.2_0.1_T09_Test02 finished
linux: 'ipc' typ=258 not implemented
2005-01-22 13:57:25 [Einstein@Home] Starting result H1_0069.9__0070.3_0.1_T09_Test02_2 using einstein version 4.73
zsh: killed ./boinc_4.16_i686-pc-linux-gnu
djaa[software/boinc]% pkg_info | grep -i linux
linux-XFree86-libs-4.3.99.902_1 XFree86 libraries, Linux binary
linux-expat-1.95.5_2 Linux/i386 binary port of Expat XML-parsing library
linux-fontconfig-2.1_2 Linux/i386 binary of Fontconfig
linux-sun-jdk-1.4.2.06_1 Sun Java Development Kit 1.4 for Linux
linux_base-8-8.0_4 Base set of packages needed in Linux mode (only for i386)
linuxdoc-1.1_1 The Linuxdoc SGML DTD
sgmlformat-1.7_2 Generates groff and HTML from linuxdoc and docbook SGML doc
djaa[software/boinc]% uname -a
FreeBSD djaa.gaudette.org 4.10-RELEASE FreeBSD 4.10-RELEASE #8: Wed Jan 5 21:37:09 EST 2005 :/usr/src/sys/compile/DJAA i386
djaa[software/boinc]%
[/pre]
Acajou
FreeBSD w. Linux emulation - should it work?
)
> FreeBSD w. Linux emulation - should it work?
Well, it would be nice if it would work - but we currently have our hands full supporting the three "official" platforms.
BM
BM
The host data that the Linux
)
The host data that the Linux client reports look rather weird:
Number of CPUs 0
CPU AuthenticAMD
FP ops/sec 1000000000
Int ops/sec 0
memory bandwidth 0
Operating System Linux 2.4.2
I don't think that this will work...
BM
BM
Hi, Thanks for the
)
Hi,
Thanks for the response. Just to be clear, I'm not expecting any kind of official support. I just thought there might be other people with a simillar config that might be interested in sharing experiences and troubleshooting. BTW, FreeBSD's Linux "emulation" has an excellent reputation and will run things as complex as Mathematica.
When I ran the E@H Linux client, the CPU benchmarks didn't appear to work at all as I saw no intensive CPU usage at that moment. But then it successfully authenticated to the servers, downloaded work and started crunching so I was hopeful. I'm guessing that the "unrecoverable error" messages make it seem rather unlikely to work. I'd try the Boinc client in the ports tree, but it's at 4.13 whereas E@H requires 4.16. I was hoping to contribute to E@H, but I guess that's not going to happen now. Bummer...
Acajou
Are you using a FreeBSD BOINC
)
Are you using a FreeBSD BOINC client with the Linux einstein science app? Or are you using a BOINC client compiled for Linux as well?
Maybe it's possible to get BOINC compiled for FreeBSD while identifying itself as LINUX to the servers?
- Marco
Team Canada
I am using the official boinc
)
I am using the official boinc client binary compiled for Linux. I'm testing 4.17 right now to see if it does better than 4.16, but looking at the port tree for version 4.13 I see some patches that look like they address issues I'm seeing. For instance, 4.17 just printed out
2005-01-24 08:19:33 [---] GUI RPC bind failed: -1
And the port has a file named "patch-client::gui_rpc_server.C"...
OK, so I'm willing to try to compile the client myself and see if I can apply those 4.13 patches to the most recent version of the source. Is it possible to get source for an official version past 4.11? Must be, but if I look at Berkeley's source distribution site, I only see versions up to 4.11 and nightly CVS snapshots. How do I get a 4.16 or 4.17 source tarball? I'd much rather work with an official release version's source than a nightly CVS snapshot...
Acajou
Hi, I've not been able to
)
Hi,
I've not been able to reproduce this with the exact same WU on either linux or Mac.
In order to help us track down the origin of this bug, you could try the following experiment:
1) download the test-app : http://www.aei.mpg.de/~repr/EinsteinAtHome_binaries/einstein_test_i686-pc-linux-gnu.tar.gz
2) put it into your projects/einstein.phys.uwm.edu/ directory and unpack it there
3) create symlinks: 'ln -s H1_0069.9 H1.sft' and 'ln -s Config_Test02 conf'
4) re-run your commandline:
./einstein_test_i686-pc-linux-gnu @conf -f70.198 -o .Ha --startTime 756889648 --endTime 756931340 // @conf -f70.198 -o .Hb --startTime 756784738 --endTime 756825508 // -1 Fstats.Ha -2 Fstats.Hb -f 0.001 -a 0.02 -d 0.02 -s 70.200 -e 70.300 -b -o polka.out
It would be very valuable for us to know if a) you get the same error-again and b) send us the complete debug-output that you get in that case.
Thanks a lot.
Cheers,
Reinhard.