Re: ultramonkey L7

["Hideaki Kondo" <toreno4257@xxxxxxxxx>]

Hello,

I'm one of the parties concerned of UltraMonkey-L7 project.
I'm sorry for being late to reply to this matter
and my poor english.

----- Original Message ----- 
From: "Neil" <prockter@xxxxxxxxx>
To: "LinuxVirtualServer.org users mailing list." 
<lvs-users@xxxxxxxxxxxxxxxxxxxxxx>
Sent: Saturday, April 22, 2006 3:43 AM
Subject: ultramonkey L7


> Hello
>
> Anyone else here had a chance to play with ultramonkey L7?
> http://ultramonkey-l7.sourceforge.jp/index_en.html
>
> Be good just to know someone had had a look and what they thought
>
> I've not noticed any talk of it here and google returns mostly japanese
Very sorry.
To be sure, there aren't any english manuals etc at present
execpt for the above web page.
We think we'd like to translate the manuals from Japanese into
English by the end of this month or in the near future.

> Also I've got a spot of difficulty with compiling kssl using rhel3-as U7
> with all the latest vanessa libraries
>
> make
> ..
> ..
> make[1]: Entering directory `/opt/monkeyL7/kssl-0.2.0-1/kssld'
> gcc -D__KERNEL__ -DMODULE -DEXPORT_SYMTAB -DMODVERSIONS
> -DCONFIG_CRYPTO_HMAC -DCONFIG_KMOD -DVERSION=\"0.1.2.cvs.20050701\"  -O2
> -Wall -Wstrict-prototypes -I..//include -I/usr/src/linux/include
> -I/crypto -include /usr/src/linux/include/linux/modversions.h
> -I../include/kssl -I../asym -DCONFIG_KSSL_DEBUG    -c -o core.o core.c
> core.c: In function `kssl_thread':
> core.c:270: structure has no member named `sigmask_lock'
> core.c:273: too many arguments to function `recalc_sigpending_Rfb6af58d'
> make[1]: *** [core.o] Error 1
> make[1]: Leaving directory `/opt/monkeyL7/kssl-0.2.0-1/kssld'
> make: *** [all] Error 1
>
> anyone come across or even better over that? My only thoughts so far is
> possibly I am missing some kernel patches
When I complied kssl-0.2.0-1 using rhel3-as U4(kernel-2.4.21-27.EL)
and "make KERNELSOURCE=/usr/src/linux-2.4.21-27.EL",
Same compile error occured.

core.c:269-274
------------------------------------------------------
// spin_lock_irq(&current->sighand->siglock);
spin_lock_irq(&current->sigmask_lock);
siginitsetinv(&current->blocked, 0);
// recalc_sigpending();
recalc_sigpending(current);
spin_unlock_irq(&current->sigmask_lock);
------------------------------------------------------
　　　　　　　　　　　　　↓
------------------------------------------------------
spin_lock_irq(&current->sighand->siglock);
// spin_lock_irq(&current->sigmask_lock);
siginitsetinv(&current->blocked, 0);
recalc_sigpending();
// recalc_sigpending(current);
// spin_unlock_irq(&current->sigmask_lock);
spin_unlock_irq(&current->sighand->siglock);
------------------------------------------------------

If you change to the way above, the compile error will not occur.
However, there is no guarantee that kssl operates correctly. 

In fact, kssl was developed and tested using vanilla kernel 2.4.26 only.
So kssl doesn't support redhat kernel at present.
Also you have to apply the kernel patch for kssl (kssl-0.2.0-1/patch/
linux-2.4.26-pre6-crypto_cpy.patch) to vanilla kernel-2.4.26 source code.

I don't recommend that you use kssl.
Because kssl needs the vanilla kernel(2.4.26 or maybe later ) and
the kernel patch, and a redhat support isn't received to you.
Besides, kssl needs the specific SSL accelerrator(AEP1000L) for high 
performance.
Sadly, AEP1000L became manufacturing discontinuance. 

The main functionality of kssl is reverse proxy with SSL support.
So you can substitute a revese proxy software for kssl.
For example, Apache(mod_proxy+mod_ssl), Squid, Delegate, Pound etc.

> Thanks for any comment
>
> Neil
>
> p.s. fell free to enlighten me if there is a english ultramonkey/
> ultramonkey l7 mailing list I should use other than this, I'll apreciate
> that as much as a straight answer
P.S. Kssl was developed by NTT COMWARE through Simon Horman's large support.

Best regards,

--
Hideaki Kondo