Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[lvs\-users\]\s+\"wrong\"\s+redirecting\s+by\s+the\s+director\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[lvs\-users\]\s+\"wrong\"\s+redirecting\s+by\s+the\s+director\s*$/: 8 ]

Total 8 documents matching your query.

1. Re: [lvs-users] "wrong" redirecting by the director (score: 1): Author: "Peter Stimpel" <peter@xxxxxxxxxxxxx>; Date: Tue, 17 Jan 2012 16:29:09 +0100; I gave it a try, using opensuse 12.1 64b, kernel is 3.1.0-1.2. It seems there are slight changes, it is becoming better - a little bit. -> 192.168.1.1:7501 Masq 11 72 0 -> 192.168.1.2:7501 Masq 11 62; /html/lvs-users/2012-01/msg00038.html (10,588 bytes)
2. Re: [lvs-users] "wrong" redirecting by the director (score: 1): Author: "Peter Stimpel" <peter@xxxxxxxxxxxxx>; Date: Tue, 17 Jan 2012 13:13:15 +0100; Might be, I will give it a shot. 2.6.22.5-31 Thank you for the idea! _______________________________________________ Please read the documentation before posting - it's available at: http://www.linu; /html/lvs-users/2012-01/msg00035.html (10,261 bytes)
3. Re: [lvs-users] "wrong" redirecting by the director (score: 1): Author: Graeme Fowler <graeme@xxxxxxxxxxx>; Date: Tue, 17 Jan 2012 11:00:51 +0000; I just noted that you're running OpenSUSE 10.3. That's very old. What kernel are you running? I ask because inactive connections are included in the calculation but I'm not 100% sure how long the sch; /html/lvs-users/2012-01/msg00034.html (9,635 bytes)
4. Re: [lvs-users] "wrong" redirecting by the director (score: 1): Author: "Peter Stimpel" <peter@xxxxxxxxxxxxx>; Date: Tue, 17 Jan 2012 11:46:04 +0100; would That idea was our first one, and RR has some downsides in our usecase. The biggest issue, it is not really a smart connection distribution. This means on restarts of a single realserver during; /html/lvs-users/2012-01/msg00033.html (10,073 bytes)
5. Re: [lvs-users] "wrong" redirecting by the director (score: 1): Author: Graeme Fowler <graeme@xxxxxxxxxxx>; Date: Tue, 17 Jan 2012 10:39:37 +0000; Good :) I would try changing the scheduler to either Round Robin (rr) or Weighted Round Robin (wrr) as you're being given the default, Weighted Least Connections (wlc). The basis of rr is exactly as; /html/lvs-users/2012-01/msg00032.html (10,543 bytes)
6. Re: [lvs-users] "wrong" redirecting by the director (score: 1): Author: "Peter Stimpel" <peter@xxxxxxxxxxxxx>; Date: Tue, 17 Jan 2012 11:29:14 +0100; behind the is, the There is no NAT involved between Loadbalancer and Clients. The Loadbalancer sees the clients IP addresses. I should add: this infrastructure is part of a VPN we are running with t; /html/lvs-users/2012-01/msg00031.html (10,656 bytes)
7. Re: [lvs-users] "wrong" redirecting by the director (score: 1): Author: Graeme Fowler <graeme@xxxxxxxxxxx>; Date: Tue, 17 Jan 2012 10:13:01 +0000; Is it possible that all the devices in use for that customer are behind some form of NAT device, so they all appear to be coming from the same IP and port? Alternatively, it may be that the sudden in; /html/lvs-users/2012-01/msg00030.html (10,255 bytes)
8. [lvs-users] "wrong" redirecting by the director (score: 1): Author: "Peter Stimpel" <peter@xxxxxxxxxxxxx>; Date: Tue, 17 Jan 2012 10:58:03 +0100; Hi all, I am using ipvs since years, and I am really happy with it - so first of all a big thank you for that stuff! I have a host running ipvs, using 1.2.3.4 as external address, and using 192.168.1; /html/lvs-users/2012-01/msg00029.html (12,980 bytes)

Current List: 1 - 8
Page: [1]