[lvs-users] Problem with ldirectord Bug?

Ben Cabot ben at loadbalancer.org
Thu Jun 5 11:23:10 BST 2014


I've seen the same thing, It looks like ipvsadm is not displaying the
details from lvs correctly.
However looking at a packet capture you can see it is sending traffic to
all servers.


[root at ip-10-0-1-72]# ipvsadm -A -t 10.0.1.72:808 -s rr
[root at ip-10-0-1-72]# ipvsadm -a -t 10.0.1.72:808 -r 10.0.1.73:808 -m
[root at ip-10-0-1-72]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  10.0.1.72:808 rr
  -> 10.0.1.73:808                Masq    1      0          0
[root at ip-10-0-1-72]# ipvsadm -a -t 10.0.1.72:808 -r 10.0.1.74:808 -m
[root at ip-10-0-1-72]# ipvsadm -Ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  10.0.1.72:808 rr
  -> 10.0.1.73:808                Masq    1      0          0
[root at ip-10-0-1-72]# ipvsadm -v
ipvsadm v1.26 2008/5/15 (compiled with popt and IPVS v1.2.1)

This was using ec2 with amazons kernel 3.10.40-50.136.amzn1.i686.



On 4 June 2014 10:38, Migue Martin <miguegt at hotmail.es> wrote:

> Hi all.
>
> I'm having a problem with ldirectord that I can not solve, I comment my
> situation:
>
>
> The problem is in the ipvsadm only show a one real IP, when i configured
> two now, show  the ldirector.cf:
>
>
> root at ldap1:/home/mamartin# ipvsadm
> IP Virtual Server version 1.2.1 (size=4096)
> Prot LocalAddress:Port Scheduler Flags
>   -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
> TCP  ldap-maqueta.cica.es:ldap wrr
>   -> ldap2.cica.es:ldap            Route   4      0 0
> TCP  [[fc00:1::3]]:ldap wrr
>   -> [[fc00:1::2]]:ldap Route   4      0 0
>
> root at ldap1:/home/mamartin# cat /etc/ldirectord.cf
> checktimeout=10
> checkinterval=2
> autoreload=yes
> logfile="/var/log/ldirectord.log"
> quiescent=yes
>
> #ipv4
> virtual=192.168.1.10:389
>         real=192.168.1.11:389 gate 4
>         real=192.168.1.12:389 gate 4
>         scheduler=wrr
>         protocol=tcp
>      checktype=on
> #ipv6
> virtual6=[[fc00:1::3]]:389
>         real6=[[fc00:1::1]]:389 gate 4
>         real6=[[fc00:1::2]]:389 gate 4
>         scheduler=wrr
>         protocol=tcp
>         checkport=389
>      checktype=on
>
>
> and in the logs I see nothing clear:
>
> root at ldap1:/home/mamartin# ldirectord -d /etc/ldirectord.cf start
> DEBUG2: Running system(/sbin/ipvsadm -a -t 192.168.1.10:389 -r
> 192.168.1.11:389 -g -w 0)
> Running system(/sbin/ipvsadm -a -t 192.168.1.10:389 -r 192.168.1.11:389
> -g -w 0)
> DEBUG2: Quiescent real server: 192.168.1.11:389 (192.168.1.10:389)
> (Weight set to 0)
> Quiescent real server: 192.168.1.11:389 (192.168.1.10:389) (Weight set to
> 0)
> DEBUG2: Disabled real server=on:tcp:192.168.1.11:389:::4:gate:\/:
> (virtual=tcp:192.168.1.10:389)
> DEBUG2: Running system(/sbin/ipvsadm -a -t 192.168.1.10:389 -r
> 192.168.1.12:389 -g -w 0)
> Running system(/sbin/ipvsadm -a -t 192.168.1.10:389 -r 192.168.1.12:389
> -g -w 0)
> DEBUG2: Quiescent real server: 192.168.1.12:389 (192.168.1.10:389)
> (Weight set to 0)
> Quiescent real server: 192.168.1.12:389 (192.168.1.10:389) (Weight set to
> 0)
> DEBUG2: Disabled real server=on:tcp:192.168.1.12:389:::4:gate:\/:
> (virtual=tcp:192.168.1.10:389)
> DEBUG2: Checking on: Real servers are added without any checks
> DEBUG2: Resetting soft failure count: 192.168.1.12:389 (tcp:
> 192.168.1.10:389)
> Resetting soft failure count: 192.168.1.12:389 (tcp:192.168.1.10:389)
> DEBUG2: Running system(/sbin/ipvsadm -a -t 192.168.1.10:389 -r
> 192.168.1.12:389 -g -w 4)
> Running system(/sbin/ipvsadm -a -t 192.168.1.10:389 -r 192.168.1.12:389
> -g -w 4)
> Destination already exists
>
>
>
>
> root at ldap1:/home/mamartin# cat /var/log/ldirectord.log
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Quiescent real server:
> 192.168.1.11:389 (192.168.1.10:389) (Weight set to 0)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Quiescent real server:
> 192.168.1.12:389 (192.168.1.10:389) (Weight set to 0)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Resetting soft failure
> count: 192.168.1.12:389 (tcp:192.168.1.10:389)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] system(/sbin/ipvsadm -a -t
> 192.168.1.10:389 -r 192.168.1.12:389 -g -w 4) failed:
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Added real server:
> 192.168.1.12:389 (192.168.1.10:389) (Weight set to 4)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Resetting soft failure
> count: 192.168.1.11:389 (tcp:192.168.1.10:389)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Restored real server:
> 192.168.1.11:389 (192.168.1.10:389) (Weight set to 4)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Resetting soft failure
> count: [[fc00:1::2]]:389 (tcp:[[fc00:1::3]]:389)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] system(/sbin/ipvsadm -a -t
> [[fc00:1::3]]:389 -r [[fc00:1::2]]:389 -g -w 4) failed:
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Added real server:
> [[fc00:1::2]]:389 ([[fc00:1::3]]:389) (Weight set to 4)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Resetting soft failure
> count: [[fc00:1::1]]:389 (tcp:[[fc00:1::3]]:389)
> [Tue Jun  3 09:39:29 2014|ldirectord.cf|19266] Restored real server:
> [[fc00:1::1]]:389 ([[fc00:1::3]]:389) (Weight set to 4)
>
>
> do not know if this is a bug or a configuration error, can anyone help?
>
> Regards.
>
> _______________________________________________
> Please read the documentation before posting - it's available at:
> http://www.linuxvirtualserver.org/
>
> LinuxVirtualServer.org mailing list - lvs-users at LinuxVirtualServer.org
> Send requests to lvs-users-request at LinuxVirtualServer.org
> or go to http://lists.graemef.net/mailman/listinfo/lvs-users
>



-- 
LOADBALANCER.ORG LTD.
www.loadbalancer.org
support at loadbalancer.org


More information about the lvs-users mailing list