Advanced search

Message boards : News : [dna] updated applications

Author Message
Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 4548 - Posted: 27 Sep 2014, 20:14:49 UTC

I noticed something was going wrong with the DNA@Home applications where they were not using the input file correctly to determine their starting values. I've upgraded all the applications to 0.44 to fix this.

This might cause some work units calculated with application versions 0.42 or 0.43 to not validate against 0.44 applications, so feel free to cancel any tasks running on version 0.42 and 0.43.

While I was at it, I also updated the 64 bit linux application with a statically compiled (and somewhat optimized) version of BOINC. So this should hopefully speed it's performance up a bit (as some users were noticing the 32 bit linux app was faster than the 64 bit app).

Let me know if you are having any issues with the new binaries!

mugnaio [TNAA]
Send message
Joined: 11 Aug 14
Posts: 8
Combined Credit: 721,094
DNA@Home: 105,856
SubsetSum@Home: 28,660
Wildlife@Home: 586,578
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 0

      
Message 4553 - Posted: 28 Sep 2014, 8:27:40 UTC - in response to Message 4548.

Let me know if you are having any issues with the new binaries!


Hi Travis,
the new DNA 0.44 application fails almost immediately on my linux box (Oracle linux based on RH 6.5); looking into stderr there is a "version `GLIBC_2.14' not found" message.

Andrea

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 4558 - Posted: 28 Sep 2014, 15:20:40 UTC - in response to Message 4553.

Let me know if you are having any issues with the new binaries!


Hi Travis,
the new DNA 0.44 application fails almost immediately on my linux box (Oracle linux based on RH 6.5); looking into stderr there is a "version `GLIBC_2.14' not found" message.

Andrea


Is this a 32 or 64 bit system?

mugnaio [TNAA]
Send message
Joined: 11 Aug 14
Posts: 8
Combined Credit: 721,094
DNA@Home: 105,856
SubsetSum@Home: 28,660
Wildlife@Home: 586,578
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 0

      
Message 4560 - Posted: 28 Sep 2014, 15:47:21 UTC - in response to Message 4558.

Is this a 32 or 64 bit system?


64 bit

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 4561 - Posted: 28 Sep 2014, 15:49:25 UTC - in response to Message 4560.

Is this a 32 or 64 bit system?


64 bit


Working on a fix, hopefully have something soon.

mugnaio [TNAA]
Send message
Joined: 11 Aug 14
Posts: 8
Combined Credit: 721,094
DNA@Home: 105,856
SubsetSum@Home: 28,660
Wildlife@Home: 586,578
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 0

      
Message 4598 - Posted: 29 Sep 2014, 17:03:15 UTC - in response to Message 4561.

Working on a fix, hopefully have something soon.


Hi Travis, I got some tasks and they are running fine.
I followed the other threads and asked myself why statically linked executables should have problems with shared libraries, so I downloaded some of the versions you build in these two days.
As you can see below, version 0.44 for linux 64 bit isn't a statically linked executable (unless of course you did change it after the problems I reported); if I run it on the command line I get the same error I reported. This does not happen with version 0.45 which in fact is a statically linked executable.
So my guess is that version 0.45 could have been a good one (except for the errors in accessing them reported by GlennG in message 4573. Since the idea of speeding up things is a Good Thing, feel free to ask me anything that can help you to achieve that.

[root@antares dna]# file Gibbs_0.44_i686-pc-linux-gnu
Gibbs_0.44_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (GNU/Linux), statically linked, for GNU/Linux 2.6.24, not stripped
[root@antares dna]# file Gibbs_0.44_i686-pc-linux-gnu
Gibbs_0.44_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (GNU/Linux), statically linked, for GNU/Linux 2.6.24, not stripped
[root@antares dna]# file Gibbs_0.44_x86_64-pc-linux-gnu
Gibbs_0.44_x86_64-pc-linux-gnu: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.24, not stripped
[root@antares dna]# file Gibbs_0.45_x86_64-pc-linux-gnu
Gibbs_0.45_x86_64-pc-linux-gnu: ELF 64-bit LSB executable, x86-64, version 1 (GNU/Linux), statically linked, for GNU/Linux 2.6.24, not stripped
[root@antares dna]# ./Gibbs_0.44_x86_64-pc-linux-gnu
./Gibbs_0.44_x86_64-pc-linux-gnu: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by ./Gibbs_0.44_x86_64-pc-linux-gnu)
[root@antares dna]# ./Gibbs_0.45_x86_64-pc-linux-gnu
[root@antares dna]#

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 4600 - Posted: 29 Sep 2014, 17:16:26 UTC - in response to Message 4598.

Working on a fix, hopefully have something soon.


Hi Travis, I got some tasks and they are running fine.
I followed the other threads and asked myself why statically linked executables should have problems with shared libraries, so I downloaded some of the versions you build in these two days.
As you can see below, version 0.44 for linux 64 bit isn't a statically linked executable (unless of course you did change it after the problems I reported); if I run it on the command line I get the same error I reported. This does not happen with version 0.45 which in fact is a statically linked executable.
So my guess is that version 0.45 could have been a good one (except for the errors in accessing them reported by GlennG in message 4573. Since the idea of speeding up things is a Good Thing, feel free to ask me anything that can help you to achieve that.

[root@antares dna]# file Gibbs_0.44_i686-pc-linux-gnu
Gibbs_0.44_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (GNU/Linux), statically linked, for GNU/Linux 2.6.24, not stripped
[root@antares dna]# file Gibbs_0.44_i686-pc-linux-gnu
Gibbs_0.44_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (GNU/Linux), statically linked, for GNU/Linux 2.6.24, not stripped
[root@antares dna]# file Gibbs_0.44_x86_64-pc-linux-gnu
Gibbs_0.44_x86_64-pc-linux-gnu: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.24, not stripped
[root@antares dna]# file Gibbs_0.45_x86_64-pc-linux-gnu
Gibbs_0.45_x86_64-pc-linux-gnu: ELF 64-bit LSB executable, x86-64, version 1 (GNU/Linux), statically linked, for GNU/Linux 2.6.24, not stripped
[root@antares dna]# ./Gibbs_0.44_x86_64-pc-linux-gnu
./Gibbs_0.44_x86_64-pc-linux-gnu: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by ./Gibbs_0.44_x86_64-pc-linux-gnu)
[root@antares dna]# ./Gibbs_0.45_x86_64-pc-linux-gnu
[root@antares dna]#


It would be good to have everyone running the faster one. v0.46 should have compiled libgcc and libstdc++ statically, so I'm confused as to why it would have been crashing for people. v0.45 had libstdc++ statically compiled but libgcc was dynamically compiled (which I think is why people were getting the GLIBC not found error).

If there are other compilation options I should try to get everything compiled statically so people don't have any issues I'm open to trying them.

mugnaio [TNAA]
Send message
Joined: 11 Aug 14
Posts: 8
Combined Credit: 721,094
DNA@Home: 105,856
SubsetSum@Home: 28,660
Wildlife@Home: 586,578
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 0

      
Message 4601 - Posted: 29 Sep 2014, 17:39:22 UTC - in response to Message 4600.

It would be good to have everyone running the faster one. v0.46 should have compiled libgcc and libstdc++ statically, so I'm confused as to why it would have been crashing for people. v0.45 had libstdc++ statically compiled but libgcc was dynamically compiled (which I think is why people were getting the GLIBC not found error).

If there are other compilation options I should try to get everything compiled statically so people don't have any issues I'm open to trying them.


I cannot find v0.46 in the /download/ directory; did you remove it ?
I got GLIBC errors with v0.44, v0.45 seems fine to me and I did not find messages of people getting error with it.
Where am I wrong ?

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 4608 - Posted: 30 Sep 2014, 0:29:54 UTC - in response to Message 4601.

It would be good to have everyone running the faster one. v0.46 should have compiled libgcc and libstdc++ statically, so I'm confused as to why it would have been crashing for people. v0.45 had libstdc++ statically compiled but libgcc was dynamically compiled (which I think is why people were getting the GLIBC not found error).

If there are other compilation options I should try to get everything compiled statically so people don't have any issues I'm open to trying them.


I cannot find v0.46 in the /download/ directory; did you remove it ?
I got GLIBC errors with v0.44, v0.45 seems fine to me and I did not find messages of people getting error with it.
Where am I wrong ?


Oh whoops, you're right. v0.46 had problems because of errors in the update_versions script. so v0.44 was the one without libgcc and v0.45 is the one with libgcc.

mugnaio [TNAA]
Send message
Joined: 11 Aug 14
Posts: 8
Combined Credit: 721,094
DNA@Home: 105,856
SubsetSum@Home: 28,660
Wildlife@Home: 586,578
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 0

      
Message 4620 - Posted: 30 Sep 2014, 12:32:19 UTC - in response to Message 4608.

Oh whoops, you're right. v0.46 had problems because of errors in the update_versions script. so v0.44 was the one without libgcc and v0.45 is the one with libgcc.


If it helps i can try to use v0.46 as a custom application; it is almost rocket science to me but with some time i should be able to get a running config.

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 4626 - Posted: 1 Oct 2014, 1:19:23 UTC - in response to Message 4620.

Oh whoops, you're right. v0.46 had problems because of errors in the update_versions script. so v0.44 was the one without libgcc and v0.45 is the one with libgcc.


If it helps i can try to use v0.46 as a custom application; it is almost rocket science to me but with some time i should be able to get a running config.


We're still fairly alpha with the development of DNA@Home so running a custom app would probably be more trouble than it's worth. Because whenever I update the application you'll have to redo everything. I'd hold off on that, and hopefully I can come up with a better solution to let people use the fast version.

[VENETO] boboviz
Send message
Joined: 10 Aug 14
Posts: 98
Combined Credit: 387,515
DNA@Home: 184,990
SubsetSum@Home: 1,636
Wildlife@Home: 200,890
Wildlife@Home Watched: 0s
Wildlife@Home Events: 0
Climate Tweets: 0
Images Observed: 0

    
Message 4640 - Posted: 2 Oct 2014, 10:23:01 UTC - in response to Message 4626.

We're still fairly alpha with the development of DNA@Home so running a custom app would probably be more trouble than it's worth.


Uh. Application seems to be very stable (and, i hope, "productive") even if it is a "alpha"....

Travis Desell
Volunteer moderator
Project administrator
Project developer
Project scientist
Send message
Joined: 16 Jan 12
Posts: 1813
Combined Credit: 23,514,257
DNA@Home: 293,563
SubsetSum@Home: 349,212
Wildlife@Home: 22,871,482
Wildlife@Home Watched: 212,926s
Wildlife@Home Events: 51
Climate Tweets: 21
Images Observed: 774

              
Message 4641 - Posted: 2 Oct 2014, 15:44:48 UTC - in response to Message 4640.

We're still fairly alpha with the development of DNA@Home so running a custom app would probably be more trouble than it's worth.


Uh. Application seems to be very stable (and, i hope, "productive") even if it is a "alpha"....


Well yes, but that doesn't mean there aren't extensions to it that will involve more updates.


Post to thread

Message boards : News : [dna] updated applications