server had no record of proth test, test abandoned ???

[looxix]

From my recent log:
[Tue Dec 17 22:53:09 2002] server had no record of proth test, test abandoned
[Tue Dec 17 22:53:09 2002] connecting to server
[Tue Dec 17 22:53:10 2002] logging into server
[Tue Dec 17 22:53:10 2002] requesting a block

What does it mean :confused:

Full(er) log in attachment

[shifted]

We've already discussed this here (http://www.free-dc.org/forum/showthread.php?s=&threadid=2135&highlight=abandoned) and here (http://www.free-dc.org/forum/showthread.php?s=&threadid=2025&highlight=abandoned), but no solution was found. :(

[looxix]

Ooops didn't see it.

Let's wait the next cllient's version.

[Mystwalker]

I would guess that it's a server-side problem - or maybe a strange communication problem.
Hard to say without having the log files. ;)

[jjjjL]

this is definately a client bug. it only effects the linux client and only very infrequently. all the same, it's still a bug.

the client is going though a total rewrite right now. when done, it will correct this flaw.

i'd be very interested to know more about the linux setup this client is being run on. specifically the distro, version, kernal #, and glib info. also, what memory and processor is it and is anything overclocked?


-Louie

[Mystwalker]

I had sort of this error on one of my Windows machines, too. I'll look up the log to determine the exact message once I'm at home again...

[looxix]

Originally posted by jjjjL
this is definately a client bug. it only effects the linux client and only very infrequently. all the same, it's still a bug.

the client is going though a total rewrite right now. when done, it will correct this flaw.

i'd be very interested to know more about the linux setup this client is being run on. specifically the distro, version, kernal #, and glib info. also, what memory and processor is it and is anything overclocked?


-Louie

Humm,

Distro: more or less an hybrid of a slckaware 8.0 and LinuxFromScratch.
kernal: 2.5.51 or 2.5.52
glibc: the linux client is statically linked, so glibc can't have an influence (glibc-2.2.3)
mobo: Asus A7M266D
cpu: 2x Athlon MP 1500+ (not overclocked)
memory: 2x 512MB DRR 2100MHz not registered.

Since one year that I have this machine I never had any problem of any sort.


Hope that it help you.

[jjjjL]

Originally posted by looxix

glibc: the linux client is statically linked, so glibc can't have an influence (glibc-2.2.3)


You know, that's what I thought too. That would make sense. And in a fair and just world, compiling static code would make the glibc irrelivant. Unfortunately, we don't live in that world.

It turns out that even statically compiled binaries use dynamic calls "transparently" though glibc :shocked:


Surfice to say, these "transparent" calls the OS does aren't always so transparent and have caused quite a lot of problems for many project designers... not just me. Hopefully we can find a way to overcome these flaws in the glibc architechture in the next version of the SB client that we are working on.

-Louie

[looxix]

Originally posted by jjjjL
You know, that's what I thought too. That would make sense. And in a fair and just world, compiling static code would make the glibc irrelivant. Unfortunately, we don't live in that world.

It turns out that even statically compiled binaries use dynamic calls "transparently" though glibc :shocked:


Indeed, and this is the resolver part of glibc


Surfice to say, these "transparent" calls the OS does aren't always so transparent and have caused quite a lot of problems for many project designers... not just me. Hopefully we can find a way to overcome these flaws in the glibc architechture in the next version of the SB client that we are working on.

-Louie

You may try a client statically linked with µClibc:
www.uclibc.org (http://www.uclibc.org) ?