|
Joseph Mack wrote:
>
> Jon Molin wrote:
>
> >
> > Now i want to combine this with linux-ha. So I downloaded the config
> > script and made this config file:
>
> I've written the configure script with HA in mind, but haven't actually
> used it that way myself. Other people have setup HA with it, but I couldn't
> get them to tell what they'd done to set it up. The configure
> script is run on the director and the realservers and doesn't the "right
> thing"
> whereever it finds itself.
>
> The best thing (probably) is to setup an LVS with the first director
> with the other director off-line, then switch the IP's to the backup
> director and run the rc.lvs script just on the backup director.
yeah, but that requires two rc.lvs scripts, right? One for the
director/server being director and one for it being only server.
I'm very new to this but it feels dangerous if heartbeat does something
wrong and sets both as directors. Or is this easily avoided with the
right(tm) configuration?
>
> > And when i run this on the first localnode, all seems fine. ipvsadm -L
> > shows nice values. Then i scped the rc.lvs_dr over to the other node and
> > it started screaming.
>
> what did it do?
well it didn't litterarly scream :) but i lost conntact with it (i ssh
to it).
here's where things started to go wrong:
LVS realserver type vs-dr
adding route to real-server network 255.255.255.224
route: bogus netmask 212.75.72.63
i've no clue where it gets that netmask...it's not in the conf file...
hmm, i add the output as attachment
/Jon
>
> Joe
>
> --
> Joseph Mack PhD, Senior Systems Engineer, Lockheed Martin
> contractor to the National Environmental Supercomputer Center,
> mailto:mack.joseph@xxxxxxx ph# 919-541-0007, RTP, NC, USA
>
> _______________________________________________
> LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://www.in-addr.de/mailman/listinfo/lvs-users looking for standard utilities
$ECHO=/bin/echo
$PING=/bin/ping -U -c 1
testing ping
ping can send one packet. is OK.
$FPING=/bin/ping -U -c 1
$IFCONFIG=/sbin/ifconfig
$NETSTAT=/bin/netstat
$ROUTE=/sbin/route
$AWK=/bin/awk
$AWK=/usr/bin/awk
$GREP=/bin/grep
$HOSTNAME_CMD=/bin/hostname
$UNAME_CMD=/bin/uname
$CAT=/bin/cat
$CUT=/bin/cut
$CUT=/usr/bin/cut
$TAIL=/usr/bin/tail
$XARGS=/usr/bin/xargs
$PS=/bin/ps
$KILL=/bin/kill
$WC=/usr/bin/wc
$TRACEROUTE=/usr/sbin/traceroute
$ARP=/sbin/arp
$ROUTE=/sbin/route
$TR=/usr/bin/tr
$EXPR=/usr/bin/expr
$CHMOD=/bin/chmod
$MV=/bin/mv
$RM=/bin/rm
$MKDIR=/bin/mkdir
$SSH=/usr/bin/ssh
$NTPD=/usr/sbin/ntpd
$IP=/sbin/ip
rc.lvs version 0.9.2 Aug 2001
(C) 2000-2001 Joseph Mack jmack@xxxxxxxx, distributed under GPL license
This file is part of the LVS project http://www.linuxvirtualserver.org
setting up www1.resfeber.se
find_System_map
System.map
$SYSTEM_MAP=/boot/System.map
$IPTABLES=/sbin/iptables
$LSMOD=/sbin/lsmod
$RMMOD=/sbin/rmmod
$INSMOD=/sbin/insmod
number nics on director 1
LVS realserver type vs-dr
adding route to real-server network 255.255.255.224
route: bogus netmask 212.75.72.63
Usage: route [-nNvee] [-FC] [<AF>] List kernel routing tables
route [-v] [-FC] {add|del|flush} ... Modify routing table for AF.
route {-h|--help} [<AF>] Detailed usage syntax for
specified AF.
route {-V|--version} Display version/author and exit.
-v, --verbose be verbose
-n, --numeric don't resolve names
-e, --extend display other/more information
-F, --fib display Forwarding Information Base (default)
-C, --cache display routing cache instead of FIB
<AF>=Use '-A <af>' or '--<af>'; default: inet
List of possible address families (which support routing):
inet (DARPA Internet) inet6 (IPv6) ax25 (AMPR AX.25)
netrom (AMPR NET/ROM) ipx (Novell IPX) ddp (Appletalk DDP)
x25 (CCITT X.25)
installing default gw 212.75.72.1 for vs-dr
deleting current default gw 212.75.72.1
setting default gw to 212.75.72.1
showing routing table
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
212.75.72.0 0.0.0.0 255.255.255.192 U 40 0 0 eth1
192.168.0.0 0.0.0.0 255.255.255.0 U 40 0 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 40 0 0 lo
0.0.0.0 212.75.72.1 0.0.0.0 UG 40 0 0 eth1
checking if DEFAULT_GW 212.75.72.1 is reachable - PING 212.75.72.1
(212.75.72.1) from 212.75.72.31 : 56(84) bytes of data.
64 bytes from 212.75.72.1: icmp_seq=0 ttl=255 time=430 usec
--- 212.75.72.1 ping statistics ---
1 packets transmitted, 1 packets received, 0% packet loss
round-trip min/avg/max/mdev = 0.430/0.430/0.430/0.000 ms
good
set_realserver_ip_forwarding to OFF (1 on, 0 off).
proc/sys/net/ipv4/ip_forward 0
searching for ipchains
ipchains not loaded, good
loading ip_tables module
find_kernel_function_name_2_4
find_kernel_name_2_4: parameter ip_tables
check_function_in_kernel
function ipt_tables not in kernel
find_module_name_2_4
find_module_name_2_4: parameter ip_tables
module name ip_tables is ip_tables
attempting to load module: ip_tables
module ip_tables already loaded
setting default policy to ACCEPT for LVS devices
clearing iptables/ipchain rules
showing iptables nat rules
Chain PREROUTING (policy ACCEPT)
target prot opt source destination
Chain POSTROUTING (policy ACCEPT)
target prot opt source destination
Chain OUTPUT (policy ACCEPT)
target prot opt source destination
showing iptables rules
Chain INPUT (policy ACCEPT)
target prot opt source destination
Chain FORWARD (policy ACCEPT)
target prot opt source destination
Chain OUTPUT (policy ACCEPT)
target prot opt source destination
correct VIP entry in /etc/iproute2/rt_tables: 200 VIP
deleting ip rule from 212.75.72.15
correct RIP entry in /etc/iproute2/rt_tables: 201 RIP
deleting ip rule from 212.75.72.31 to 255.255.255.224/25
deleting ip rule from 212.75.72.31
ip rules (RIP and VIP tables should be empty)
0: from all lookup local
32766: from all lookup main
32767: from all lookup 253
route table name<->id translation table:
#
# reserved values
#
#255 local
#254 main
#253 default
#0 unspec
#
# local
#
#1 inr.ruhep
200 VIP
201 RIP
clearing priority routing table
deleting table VIP
ip route table VIP already empty
routing table VIP should be empty:
deleting table RIP
ip route table RIP already empty
routing table RIP should be empty:
device lo:110 has VIP 212.75.72.15
new VIP device == old VIP device, don't reinstall
device lo:110 has VIP 212.75.72.15 and is UP
removing 212.75.72.15 from lo:110
looking for DIP 212.75.72.24
PING 212.75.72.24 (212.75.72.24) from 212.75.72.31 : 56(84) bytes of data.
64 bytes from 212.75.72.24: icmp_seq=0 ttl=255 time=291 usec
--- 212.75.72.24 ping statistics ---
1 packets transmitted, 1 packets received, 0% packet loss
round-trip min/avg/max/mdev = 0.291/0.291/0.291/0.000 ms
found, good
not local, good
looking for VIP on director from realserver
looking for VIP on director from realserver
director is accepting packets on VIP 212.75.72.15, device eth1:1
No VIP on real-server, VIP will be on director.
pinging VIP 212.75.72.15 from RIP
server gw not on director
OK is vs-dr
LVS_TYPE is vs-dr and VIP 212.75.72.15 not in same network as RIP
Warning: You haven't set "$ROUTER_FORWARDS"
in the User configurable section of configure.pl,
so we don't know if the VIP should be pingable or not.
This ping test is just extra checking and the LVS
will work whether or not this ping test is properly setup.
If this test is properly setup and fails,
it will help diagnose problems with your LVS.
VIP should not be pingable if forwarding is off in the box router/test_client
VIP should be pingable if forwarding is on in the box router/test_client
PING 212.75.72.15 (212.75.72.15) from 212.75.72.31 : 56(84) bytes of data.
From 212.75.72.31: Destination Host Unreachable
--- 212.75.72.15 ping statistics ---
1 packets transmitted, 0 packets received, +1 errors, 100% packet loss
212.75.72.15 not pingable.
If you find that your LVS works
AND you aren't going to be changing the routing
on the real-server's default gw 212.75.72.1,
then set "$ROUTER_FORWARDS"="N"
LVS_TYPE = vs-dr, VIP device installed
OS verion Linux-2.4.7-10custom being treated as minor version 6
lo is not a tunl device, OK
lo:110 is local, up'ing lo.
install_realserver_vip: configuring Linux 2.4.6
ifconfig output
lo:110 Link encap:Local Loopback
inet addr:212.75.72.15 Mask:255.255.255.255
UP LOOPBACK RUNNING MTU:16436 Metric:1
installing route for VIP 212.75.72.15 on device lo:110
listing routing info for VIP 212.75.72.15
212.75.72.15 0.0.0.0 255.255.255.255 UH 40 0 0 lo
hiding interface lo:110, will not arp
Installing priority routing
lvs type vs-dr, SERVER_GW 212.75.72.1
priority rules
0: from all lookup local
32766: from all lookup main
32767: from all lookup 253
routes VIP
routes RIP
adding rules to RIP with priority 100 from 212.75.72.31 to 255.255.255.224/25
adding routing rules to RIP with priority 100 from 212.75.72.31 to 0/0
showing ip rules for priority 100
100: from 212.75.72.31 to 255.255.255.224/25 lookup RIP
100: from 212.75.72.31 lookup RIP
set src device for packets from VIP
packets with src=VIP are sent out eth1
ip route show table RIP (should be empty):
no gateway - is link route: ip route add table RIP to 255.255.255.224/25 dev
eth1 from 212.75.72.31
RTNETLINK answers: Invalid argument
route add: return_code 2
Error: table RIP already has entry: via dev eth1
cannot overwrite entry, you'll have to delete the old one first.
current entry: set src device for packets from VIP
packets with src=VIP are sent out eth1
ip route show table RIP (should be empty):
adding route to gateway 212.75.72.24 in table RIP
ip route add table RIP to 0/0 via 212.75.72.24 dev eth1 from 212.75.72.31
route add: return_code 0
ip route for table RIP
default via 212.75.72.24 dev eth1
adding VIP routing rules with priority 99 to realserver.
showing ip rules for prio 99
99: from 212.75.72.15 lookup VIP
set src device for packets from VIP
packets with src=VIP are sent out eth1
ip route show table VIP (should be empty):
adding route to gateway 212.75.72.1 in table VIP
ip route add table VIP to 0/0 via 212.75.72.1 dev eth1 from 212.75.72.15
route add: return_code 0
ip route for table VIP
default via 212.75.72.1 dev eth1
removing default gw
showing priority routing
0: from all lookup local
99: from 212.75.72.15 lookup VIP
100: from 212.75.72.31 to 255.255.255.224/25 lookup RIP
100: from 212.75.72.31 lookup RIP
32766: from all lookup main
32767: from all lookup 253
212.75.72.15 dev lo scope link src 212.75.72.15
212.75.72.0/26 dev eth1 proto kernel scope link src 212.75.72.31
192.168.0.0/24 dev eth0 scope link
127.0.0.0/8 dev lo scope link
routing for table VIP
default via 212.75.72.1 dev eth1
routing for table RIP
default via 212.75.72.24 dev eth1
ntpd not running, won't be restarted
not adding filter rules.
The location of the output files rc.lvs, mon.cf and ntp.conf is the default = ./
You can change this by editing the variables $rc_lvs_home, $rc_mon_home,
$rc_ntp_home
Errors: 1
Some of these errors are from tests that failed.
If you're experimenting, it's possible that the LVS will work.
If you're doing production, you can't assume that the LVS will work.
Your LVS may not be setup properly.
Warnings: 1
The configure script has encountered an unexpected situation.
Your LVS will probably run (it may not do what you want).
It would be reasonable to try your LVS first,
since it will likely work at least partially.
It is possible that the conf file has invalid information
(or the configure script has a bug).
If you suspect a bug in the script, please contact jmack@xxxxxxxx
or post to the LVS mailing list lvs-users@xxxxxxxxxxxxxxxxxxxxxx
(You can always run rc.lvs manually on director/real-servers.)
|
