But to make sure that it work on all releases a check of the different libm.so.* on a Solaris 10 system would be needed.

What do you need:

/lib$ uname -srpiv
SunOS 5.10 Generic_118833-36 sparc SUNW,Sun-Fire-V490
/lib$ ls -l libm.so*
lrwxrwxrwx   1 root     root           9 Apr 10  2007 libm.so -> libm.so.2
-rwxr-xr-x   1 root     bin        23828 Jan 23  2005 libm.so.1
-rwxr-xr-x   1 root     bin       792916 Jan 23  2005 libm.so.2
/lib$

Thank you very much !

According to your output I think that the astro application has no problem, if it would be linked explicit against the libm.so.1.

Now as this seems to be near a solution.
What about the new problem with the pthreads lib?

Remember I did not have any trouble with verion 1.07, but I have trouble with 1.13

Thanks

BUMP. Any news about this problem ?

Just wanted to add a voice to those who would be interested in running this project on a SPARC running Solaris 8. I applied some updates (hadn't been following patches), and got to a state where everything resolved, but it errors at startup.

~/boinc/BOINC>./boinc --version
5.10.17 sparc-sun-solaris

~/boinc/BOINC>./boinc_cmd --get_host_info
timezone: -18000
domain name: longwood
IP addr: 192.168.0.10
#CPUS: 1
CPU vendor: SUNW,Ultra-5_10
CPU model: sparcv9+vis
CPU FP OPS: 209424083.769634
CPU int OPS: 428183199.086342
CPU mem BW: 1000000000.000000
OS name: SunOS
OS version: 5.8
mem size: 402653184.000000
cache size: 1000000.000000
swap size: 537247744.000000
disk size: 56733159526249472000.000000
disk free: 6148447033722383360.000000

~/boinc/BOINC/projects/milkyway.cs.rpi.edu_milkyway>./astronomy_1.13_sparc-sun-solaris

~/boinc/BOINC/projects/milkyway.cs.rpi.edu_milkyway>cat stderr.txt
Can't open init data file - running in standalone mode
set_worker_timer(): pthread_create(): -1

~/boinc/BOINC/projects/milkyway.cs.rpi.edu_milkyway>ldd astronomy_1.13_sparc-sun-solaris
libCstd.so.1 => /lib/libCstd.so.1
libCrun.so.1 => /lib/libCrun.so.1
libm.so.2 => /lib/libm.so.2
libc.so.1 => /lib/libc.so.1
libdl.so.1 => /usr/lib/libdl.so.1
/usr/lib/cpu/sparcv8plus/libCstd_isa.so.1
/usr/platform/SUNW,Ultra-5_10/lib/libc_psr.so.1

Note that I did cheat on libm.so.2 (symlink to libm.so.1)

My dual processor Solaris 9.05 machine is failing as well. ldd shows all libraries in place. Like everyone else I cheated with the libm.so.2 file but everything should be fine.

Todd

This is the BOINC error log that hopefully can help troubleshooting.

This is Solaris 9 using the BOINC 5.10.17 client

1/27/2008 01:21:33|Milkyway@home|Starting gs_163_1201401068_278582_0
1/27/2008 01:21:33|Milkyway@home|Starting task gs_163_1201401068_278582_0 using astronomy version 113
1/27/2008 01:21:35|Milkyway@home|Deferring communication for 1 min 0 sec
1/27/2008 01:21:35|Milkyway@home|Reason: Unrecoverable error for result gs_163_1201401068_278582_0 (process exited with code 255 (0xff, -1))
1/27/2008 01:21:35|Milkyway@home|Computation for task gs_163_1201401068_278582_0 finished
1/27/2008 01:21:35|Milkyway@home|Output file gs_163_1201401068_278582_0_0 for task gs_163_1201401068_278582_0 absent
1/27/2008 01:21:35|Milkyway@home|Starting gs_163_1201401068_278583_0
1/27/2008 01:21:35|Milkyway@home|Starting task gs_163_1201401068_278583_0 using astronomy version 113
1/27/2008 01:21:35|Milkyway@home|Starting gs_163_1201401068_278584_0
1/27/2008 01:21:35|Milkyway@home|Starting task gs_163_1201401068_278584_0 using astronomy version 113
1/27/2008 01:21:36|Milkyway@home|Deferring communication for 1 min 0 sec
1/27/2008 01:21:36|Milkyway@home|Reason: Unrecoverable error for result gs_163_1201401068_278584_0 (process exited with code 255 (0xff, -1))
1/27/2008 01:21:36|Milkyway@home|Computation for task gs_163_1201401068_278584_0 finished
1/27/2008 01:21:36|Milkyway@home|Output file gs_163_1201401068_278584_0_0 for task gs_163_1201401068_278584_0 absent
1/27/2008 01:21:36|Milkyway@home|Starting gs_161_1201401068_278585_0
1/27/2008 01:21:36|Milkyway@home|Starting task gs_161_1201401068_278585_0 using astronomy version 113
1/27/2008 01:21:37|Milkyway@home|Deferring communication for 1 min 0 sec
1/27/2008 01:21:37|Milkyway@home|Reason: Unrecoverable error for result gs_161_1201401068_278585_0 (process exited with code 255 (0xff, -1))
1/27/2008 01:21:37|Milkyway@home|Computation for task gs_161_1201401068_278585_0 finished
1/27/2008 01:21:37|Milkyway@home|Output file gs_161_1201401068_278585_0_0 for task gs_161_1201401068_278585_0 absent
1/27/2008 01:21:37|Milkyway@home|Starting gs_161_1201401068_278586_0
1/27/2008 01:21:37|Milkyway@home|Starting task gs_161_1201401068_278586_0 using astronomy version 113
1/27/2008 01:21:38|Milkyway@home|Deferring communication for 1 min 0 sec
1/27/2008 01:21:38|Milkyway@home|Reason: Unrecoverable error for result gs_161_1201401068_278586_0 (process exited with code 255 (0xff, -1))
1/27/2008 01:21:38|Milkyway@home|Computation for task gs_161_1201401068_278586_0 finished
1/27/2008 01:21:39|Milkyway@home|Deferring communication for 1 min 29 sec
1/27/2008 01:21:39|Milkyway@home|Reason: Unrecoverable error for result gs_163_1201401068_278583_0 (process exited with code 255 (0xff, -1))
1/27/2008 01:21:39|Milkyway@home|Computation for task gs_163_1201401068_278583_0 finished
1/27/2008 01:21:39|Milkyway@home|Output file gs_163_1201401068_278583_0_0 for task gs_163_1201401068_278583_0 absent

This is the BOINC error log that hopefully can help troubleshooting.

This is Solaris 9 using the BOINC 5.10.17 client

1/27/2008 01:21:33|Milkyway@home|Starting gs_163_1201401068_278582_0
1/27/2008 01:21:33|Milkyway@home|Starting task gs_163_1201401068_278582_0 using astronomy version 113
1/27/2008 01:21:35|Milkyway@home|Deferring communication for 1 min 0 sec
1/27/2008 01:21:35|Milkyway@home|Reason: Unrecoverable error for result gs_163_1201401068_278582_0 (process exited with code 255 (0xff, -1))

I've found the same problem downgrading a sunfirev240 from Solaris 10 to 9.
I've backuped in a tar file the working boinc (5.10.17) user, and as usual move to the new env.

Without the libm.so.2 present the task continue to restart:

Project Date Message
Milkyway@home 23/04/2008 23.20.39 Restarting task gs_571_1208988172_92822_0 using astronomy version 121

after coping the libm.so.2 from a Solaris 10 server, (ldd => ok) I see a WUs waste with continue computation error.

Have you found the solution ?

Regards,

Alessandro

Project Date Message
Milkyway@home 23/04/2008 23.20.39 Restarting task gs_571_1208988172_92822_0 using astronomy version 121

Sorry, my complete log was:

Project Date Message
Milkyway@home 23/04/2008 23.20.41 Restarting task gs_571_1208988172_92822_0 using astronomy version 121
Milkyway@home 23/04/2008 23.20.42 Task gs_571_1208988172_92822_0 exited with zero status but no 'finished' file
Milkyway@home 23/04/2008 23.20.42 If this happens repeatedly you may need to reset the project.