|
Hello,
Yes, the http://www.howtoforge.com/loadbalanced_mysql_cluster_debian
is the correct link of the tutorial.
I have configured the lo:0 interfaces with the loadbalancer ip as described in
page 7.
on node sql1 i have
ip a s lo:0
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet 192.168.201.100/32 scope global lo:0
on node sql2
ip a s lo:0
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet 192.168.201.100/32 scope global lo:0
I have added another host to the VPN network with the IP 192.168.201.14.
>From it, issuing arping gives off the following:
arping 192.168.201.100
ARPING 192.168.201.100
42 bytes from fe:ff:ff:ff:ff:ff (192.168.201.100): index=0 time=1.205 msec
42 bytes from fe:ff:ff:ff:ff:ff (192.168.201.100): index=1 time=1.467 msec
42 bytes from fe:ff:ff:ff:ff:ff (192.168.201.100): index=2 time=1.460 msec
42 bytes from fe:ff:ff:ff:ff:ff (192.168.201.100): index=3 time=1.481 msec
I also ran tcpdump on the load balancer host and i have noticed a strange thing.
When i do tcpdump on the tun0 interface, i get the chatter between the
management node and the sql nodes but no traffic directed to the VIP.
tcpdump -i tun0 port 3306
tcpdump: WARNING: arptype 65534 not supported by libpcap - falling back to
cooked socket
tcpdump: verbose output suppressed, use -v or -vv for full protocol
decode
listening on tun0, link-type LINUX_SLL (Linux cooked), capture size 96
bytes
09:21:26.929772 IP ip-192-168-201-1.ec2.internal.57553 >
ip-192-168-201-6.ec2.internal.mysql: S 3517636247:3517636247(0) win 5840 <mss
1460,sackOK,timestamp 13004102 0,nop,wscale 7>
09:21:26.935957 IP ip-192-168-201-6.ec2.internal.mysql >
ip-192-168-201-1.ec2.internal.57553: S 1616136901:1616136901(0) ack 3517636248
win 5792 <mss 1366,sackOK,timestamp 171829194 13004102,nop,wscale
7>
09:21:26.935975 IP ip-192-168-201-1.ec2.internal.57553 >
ip-192-168-201-6.ec2.internal.mysql: . ack 1 win 46 <nop,nop,timestamp 13004103
171829194>
09:21:26.936123 IP ip-192-168-201-1.ec2.internal.57553 >
ip-192-168-201-6.ec2.internal.mysql: F 1:1(0) ack 1 win 46 <nop,nop,timestamp
13004103 171829194>
09:21:26.936480 IP ip-192-168-201-1.ec2.internal.34946 >
ip-192-168-201-10.ec2.internal.mysql: S 3505041773:3505041773(0) win 5840 <mss
1460,sackOK,timestamp 13004103 0,nop,wscale 7>
09:21:26.936710 IP ip-192-168-201-6.ec2.internal.mysql >
ip-192-168-201-1.ec2.internal.57553: P 1:128(127) ack 1 win 46
<nop,nop,timestamp 171829199 13004103>
09:21:26.936723 IP ip-192-168-201-1.ec2.internal.57553 >
ip-192-168-201-6.ec2.internal.mysql: R 3517636248:3517636248(0) win
0
09:21:26.937079 IP ip-192-168-201-6.ec2.internal.mysql >
ip-192-168-201-1.ec2.internal.57553: F 128:128(0) ack 1 win 46
<nop,nop,timestamp 171829199 13004103>
On the other hand, if i run tcpdump on the lo interface of the loadbalancer and
from another session on the loadbalancer run a "telnet 192.168.201.100 3306" i
get:
tcpdump -i lo port 3306
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on lo, link-type EN10MB (Ethernet), capture size 96 bytes
09:22:08.367299 IP ip-192-168-201-100.ec2.internal.43644 >
ip-192-168-201-100.ec2.internal.mysql: S 3556049393:3556049393(0) win 32792
<mss 16396,sackOK,timestamp 13014460 0,nop,wscale 7>
09:22:08.369408 IP ip-192-168-201-100.ec2.internal.mysql >
ip-192-168-201-100.ec2.internal.43644: R 0:0(0) ack 3556049394 win 0
So the traffic appears to be directed at lo instead of tun0.
Thanks.
--- On Thu, 4/15/10, Bacterie <eu_rau@xxxxxxxxx> wrote:
From: Bacterie <eu_rau@xxxxxxxxx>
Subject: [lvs-users] Mysql cluster ldirector IP Can't connect to MySQL
To: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Date: Thursday, April 15, 2010, 8:17 PM
Hello,
First, let me thank you for the great work that you are doing.
I'm trying to set up a Mysql Cluster over AWS instances, but due to the
limitations of Amazon(single ip /instance) i cannot use a virtual IP.
So, i decided to setup a VPN network first so i can have all the IP that i need.
I set up the VPN, all works fine regarding it.
I followed the instructions in
http://www.howtoforge.com/loadbalanc...cluster_debian in order to set up the
cluster.
I used debian lenny for the setup.
The following IP addresses are used:
192.168.201.1 -load balancer
192.168.201.6 -sql1
192.168.201.10-sql2
192.168.201.100- Virtual IP
On the load balacer node (only one, which is also the OpenVPN server ) i have
the following:
ha.cf
Quote:
logfacility local0
bcast tun0
mcast tun0 225.0.0.1 694 1 0
auto_failback off
node <name provided by uname -n for this amazon instance>
node loadb2
respawn hacluster /usr/lib/heartbeat/ipfail
apiauth ipfail gid=haclient uid=hacluster
haresources
Quote:
<name provided by uname -n for this amazon instance> \
ldirectord::ldirectord.cf \
LVSSyncDaemonSwap::master \
IPaddr2::192.168.201.100/24/tun0/192.168.201.255
ldirectord.cf
Quote:
# Global Directives
checktimeout=10
checkinterval=2
autoreload=no
logfile="local0"
quiescent=yes
virtual = 192.168.201.100:3306
service = mysql
real = 192.168.201.6:3306 gate
real = 192.168.201.10:3306 gate
checktype = negotiate
login = "ldirector"
passwd = "ldirectorpassword121"
database = "ldirectordb"
request = "SELECT * FROM connectioncheck"
scheduler = wrr
A few checks reveal:
ip addr sh tun0
Quote:
3: tun0: <POINTOPOINT,MULTICAST,NOARP,UP,LOWER_UP> mtu 1500
qdisc pfifo_fast state UNKNOWN qlen 100
link/[65534]
inet 192.168.201.1 peer 192.168.201.2/32 scope global tun0
inet 192.168.201.100/24 brd 192.168.201.255 scope global tun0
ipvsadm -L -n
Quote:
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 192.168.201.100:3306 wrr
-> 192.168.201.6:3306 Route 1 0 0
-> 192.168.201.10:3306 Route 1 0 0
/etc/ha.d/resource.d/LVSSyncDaemonSwap master status
master running
(ipvs_syncmaster pid: 1474)
ldirectord ldirectord.cf status
ldirectord for /etc/ha.d/ldirectord.cf is running with pid: 1391
However, when i try to connect to the virtual ip i get the following
Quote:
mysql -h 192.168.201.100 -u ldirector -p
Enter password:
ERROR 2003 (HY000): Can't connect to MySQL server on '192.168.201.100' (111)
What am i doing wrong? Is The fact that i'm running it over VPN?
Thanks,
Catalin
_______________________________________________
Please read the documentation before posting - it's available at:
http://www.linuxvirtualserver.org/
LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
or go to http://lists.graemef.net/mailman/listinfo/lvs-users
_______________________________________________
Please read the documentation before posting - it's available at:
http://www.linuxvirtualserver.org/
LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
or go to http://lists.graemef.net/mailman/listinfo/lvs-users
|
