LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

Re: SNAT Confusion

To: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Subject: Re: SNAT Confusion
Cc: jkrzyszt@xxxxxxxxxxxx
From: "Rodre Ghorashi-Zadeh" <rodrico7@xxxxxxxxxxx>
Date: Sun, 18 Mar 2007 16:08:45 -0700
Hello,


no sorry. I assume you've read the sections in the HOWTO on clients on realservers?

I read a few sections in the howto regarding this, but I didn't feel any of them were taking into consideration if the realserver was to loop back to a service running on itself via the director:

realserver1 -> LVS-DR -> realserver1

Could you please send me a link to what you think is the relevent section? Thanks.

~Rodre


From: Joseph Mack NA3T <jmack@xxxxxxxx>
Reply-To: "LinuxVirtualServer.org users mailing list." <lvs-users@xxxxxxxxxxxxxxxxxxxxxx> To: "LinuxVirtualServer.org users mailing list." <lvs-users@xxxxxxxxxxxxxxxxxxxxxx>
CC: jkrzyszt@xxxxxxxxxxxx
Subject: Re: SNAT Confusion
Date: Sun, 18 Mar 2007 14:19:12 -0700 (PDT)

On Sat, 17 Mar 2007, Rodre Ghorashi-Zadeh wrote:

Hello,


So if the realserver is dead, it can't ask the 2nd request?


I think the fear is more along the lines of what if the service is dead, and perhaps being able to perform rolling maintenance. Also the app that I am trying to get load balanced, Oracle OCS, actually mentions the f5 load balancers SNAT mode, so I believe it is a pretty deep rooted requirement.

yes a few people are starting to ask about the F5 SNAT mode. We don't have it unfortunately.



LVS could be pure netfilter, but it would be really slow.

Enough said. I knew there had to be a reason, now I understand why. Out of curiosity do you think that this still holds true with todays hardware, gig nics, dual/quad core CPUs, etc?

I expect so. I can't measure the increase in latency caused by LVS over just straight routing. I haven't done any measurements with netfilter, but everyone seems to know when it's on. The clock rate for everything increases at about the same speed - the same technology is being used for all new hardware - so while the absolute speed is increasing, the ratio of speeds remains the same. The step that was the bottleneck 10yrs ago is still the bottleneck today.


In regards to my problem I still can't get the reply packets, once SNAT-ed, sent to the realserver, and sent back to the director to be accepted by the director and sent back to the client. I am thinking it might have some thing to do with some of the the /proc/sys/net/ipv4 params, anyone have any ideas?

no sorry. I assume you've read the sections in the HOWTO on clients on realservers?

Joe
--
Joseph Mack NA3T EME(B,D), FM05lw North Carolina
jmack (at) wm7d (dot) net - azimuthal equidistant map
generator at http://www.wm7d.net/azproj.shtml
Homepage http://www.austintek.com/ It's GNU/Linux!

_________________________________________________________________
Have Some Fresh Air Fun This March Break http://local.live.com/?mkt=en-ca/?v=2&cid=A6D6BDB4586E357F!147


<Prev in Thread] Current Thread [Next in Thread>