All I set up my new rac configuration. Here is the overview
- I configured scan using dns server on my openfiler box - the nslookup returns the expected results.
- Using the scan address I could NOTconnect to the instance
- using the vip address I was able to connect
I realized my hosts file had _ (underscores) rather than - (hyphens) ie and did not match my entries in the dns
so I changed them to - as below and bounced the server
opgx280.oracle.lab ~ # cat /etc/hosts
127.0.0.1 localhost.localdomain localhost
# Public Network - (eth0)
192.168.1.200 dim4700
192.168.1.202 opgx280
192.168.1.204 opgx620
192.168.1.205 openfiler
# Private Interconnect - (eth1)
192.168.2.200 dim4700-interconnect
192.168.2.202 opgx280-interconnect
192.168.2.204 opgx620-interconnect
192.168.2.225 openfiler-interconnect
# Public Virtual IP (VIP) addresses - (eth0:1)
192.168.1.220 dim4700-vip
192.168.1.222 opgx280-vip
192.168.1.224 opgx620-vip
vi /srv/named/data/oracle.lab.zone
...
@ IN NS openfiler.oracle.lab.
localhost IN A 127.0.0.1
dim3000 IN A 192.168.1.203
openfiler IN A 192.168.1.205
opgx280 IN A 192.168.1.202
dim4700 IN A 192.168.1.200
opgx620 IN A 192.168.1.204
opgx280-vip IN A 192.168.1.222
dim4700-vip IN A 192.168.1.220
opgx620-vip IN A 192.168.1.224
lab-cluster-scan IN A 192.168.1.187
IN A 192.168.1.188
IN A 192.168.1.189
since then, I lost ability to connect using the vip AND I cannot ping the vip NOR can I ping the vip ip address
From what I can tell - oracle manages resolution of the vip
So I believe my changing of the /etc/hosts entry - broke oracle's resolution.
since the ip is indeed virtual It makes sense from a system prespective that it would NOT resolve. (What would it resolve to?)
So I need to tell oracle about the new hosts name
opgx280.oracle.lab ~ # ifconfig
eth0 Link encap:Ethernet HWaddr 00:12:3F:4E:CA:A0
inet addr:192.168.1.202 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::212:3fff:fe4e:caa0/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1755 errors:0 dropped:6 overruns:0 frame:0
TX packets:1857 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:185435 (181.0 KiB) TX bytes:195883 (191.2 KiB)
Interrupt:16
eth0:1 Link encap:Ethernet HWaddr 00:12:3F:4E:CA:A0
inet addr:192.168.1.225 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
Interrupt:16
eth1 Link encap:Ethernet HWaddr 00:0F:B5:06:38:E9
inet addr:192.168.2.202 Bcast:192.168.2.255 Mask:255.255.255.0
inet6 addr: fe80::20f:b5ff:fe06:38e9/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:81795 errors:0 dropped:6 overruns:0 frame:0
TX packets:54169 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:110063686 (104.9 MiB) TX bytes:10799080 (10.2 MiB)
Interrupt:17 Base address:0x2000
I have no idea where oracle is getting the 192.168.1.225 from!
nmap -T4 -F 192.168.1.1/24
Nmap scan report for openfiler_vip (192.168.1.225)
Host is up (0.0017s latency).
Not shown: 97 closed ports
PORT STATE SERVICE
22/tcp open ssh
111/tcp open rpcbind
873/tcp open rsync
MAC Address: 00:12:3F:4E:CA:A0 (Dell)
opgx280.oracle.lab ~ # nslookup openfiler_vip
Server: 192.168.1.205
Address: 192.168.1.205#53
** server can't find openfiler_vip: NXDOMAIN
Im really confused with this. My eth0:1 is broadcasting 192.168.1.225 how I can tell oracle about the new addresses??
I notice it takes about 3-4 min for eth0:1 to show up in ifconfig, This seems to be something oracle is doing
Needless to say the scan address is NOT working
WTH!