SB v1.2.5 Linux and FreeBSD clients

[Alien88]

On this glorious leapyear day, I bring you a nice, stable, tested client for both FreeBSD and Linux. They are available on the download page (http://www.seventeenorbust.com/download/).

Please Note: team config has been removed, so delete the line from your sclient.conf file or use the new config and re-edit it.

As always, should there be any problems post them here.

v1.2.2 -> v1.2.5
- LINUX:
- no more memory leak
- fix intermediate block problem
- fixed an invalid write of size 1 after getting a new block causing segfaults
- fixed another invalid write causing a segfault on reporting complete blocks
- fixed an unitialzed value in a loop on first block
- fixed network transmit hang bug
- fixed network transmit bug on 2.2 kernels due to lack of socket options
- removed config since teams are done via the website and updated sclient.conf to reflect this
- fix weird strcpy in cpudetect code

* thanks to scrap, Frodo42, FBCrack, hc_grove, eatmadustch and meremortal for their testing and bug reports

--
Mike Garrison

[Joh14vers6]

@Alien88:

Have you contacted Louie for the windowsversion of the 1.25 client?

[Alien88]

Originally posted by Joh14vers6
@Alien88:

Have you contacted Louie for the windowsversion of the 1.25 client?

Louie is kind of busy working on something else right now, and the 1.2.0 client is fairly stable so it's not on the list of majorly important things to do right now.

[MathGuy]

Originally posted by Joh14vers6

Have you contacted Louie for the windowsversion of the 1.25 client?

What problems are you experiencing with the Windows client? All the folks I have been in contact with have complete stability with 1.2.0...maybe I just haven't been in contact with the right folks...

[Joh14vers6]

Originally posted by MathGuy
What problems are you experiencing with the Windows client? All the folks I have been in contact with have complete stability with 1.2.0...maybe I just haven't been in contact with the right folks...

See here my problems: http://www.free-dc.org/forum/showthread.php?threadid=5340

[MathGuy]

e-mail me at kerryj@nospam.math.bsu.edu (remove the first word of the host name) with your configuration info (OS, logs, anything else that seems relevant) and I'll help you work through getting this cleared up. Enough people are making this version work consistently that I'm confident we can get you going as well.

[Joh14vers6]

Originally posted by MathGuy
e-mail me at kerryj@nospam.math.bsu.edu (remove the first word of the host name) with your configuration info (OS, logs, anything else that seems relevant) and I'll help you work through getting this cleared up. Enough people are making this version work consistently that I'm confident we can get you going as well.

Which logs do you want? PC that gives problems is a P4 (see posting here (http://www.free-dc.org/forum/showthread.php?action=newreply&postid=51171)) with OS XP prof, the other one gives no problems.

[MathGuy]

From the "\program files\sb" directory, sb.log and sobsvc.log (if you're using the service handler)

[maltek]

Could you please provide the source code of the sb client for Linux ?

[Alien88]

Originally posted by maltek
Could you please provide the source code of the sb client for Linux ?

The source code for v1 of the client is not available to the general public. Please see this thread (http://www.free-dc.org/forum/showthread.php?s=&threadid=1940) for some of the explanation why.

[Alien88]

I know there is still an issue with linux 2.2.x boxes and it hanging. It's not as bad as before, but my fix doesnt quite work. I, however, have found a better way to handle it and it works fine. Should anyone use sb on a box running 2.2.x contact me and I will compile a test client for you that you can use until the next linux client has been released.

The next linux client I plan on having it save its progress when it is ctrl-c'ed or killed.

[phooey]

:bouncy:

Thanks for all your work on this Alien88! I've been running the 1.2.5 client on a Redhat 9 dual AthlonXP machine for 5 days with no segfaults.