Yeah, I don't understand why the routing problems, because from the fw
machine I can get to both servers. The fw machine is also the default
gateway, it is 192.168.1.1
sysadmin@fw:~$ telnet 192..168.1.204 25
Trying 192.168.1.204...
Connected to 192.168.1.204..
Escape character is '^]'.
220 email1.americasnet.com ESMTP
sysadmin@fw:~$ telnet 192.168.1.102 25
Trying 192.168..1.102...
Connected to 192.168.1.102.
Escape character is '^]'.
220 americasnet.com ESMTP MailCleaner (Community Edition 2010 beta 3)
Sun, 06 Mar 2011 07:44:58 -0800
At Sunday, 03-06-2011 on 6:50 "Henrique Fernandes" wrote:
sysadmin@fw
What ip has that machine is that ?
It seens you are having some problems routing the packages,
Which are the default gateways in mail1 and mail2 ?
[]'sf.rique
On Sun, Mar 6, 2011 at 1:27 AM, Ricardo Kleemann wrote:
Hello Henrique,
The problem is very strange.
From the ldirectord server itself, I'm able to get to BOTH smtps as I
showed in the original post.
So I'll do what you suggested.
First, with BOTH servers enabled:
sysadmin@fw:~$ telnet 192.168.1.200 25
Trying 192.168.1.200...
Connected to 192.168.1.200.
Escape character is '^]'.
220 email1.americasnet.com [1] ESMTP
Second attempt, still with BOTH enabled:
sysadmin@fw:~$ telnet 192.168.1.200 25
Trying 192.168.1.200...
telnet: Unable to connect to remote host: Connection timed out
I tried a third time, got the same result as above. Then I tried a
fourth time, and was successful, on server email1 as above.
Now, I turn off the email1 server.
The result is the connection timeout
sysadmin@fw:~$ telnet 192.168.1.200 25
Trying 192.168..1.200...
telnet: Unable to connect to remote host: Connection timed out
I tried it 4 times, with same result, connection timeout
Now I turn off the second server (the one that doesn't seem to work
from ldirectord). The result is that I get an immediate connection
refused (rather than a timeout). That would make sense because there
are no real servers at all available.
As soon as I bring up the second server again (the one that doesn't
work with ldirectord) then the situation goes back to the timeout.
But, as I said before, if I manually telnet to it, it works fine.
It's only through ldirectord that it doesn't work.
Ricardo
At Saturday, 03-05-2011 on 18:36 "Henrique Fernandes" wrote:
[]'sf.rique
On Sat, Mar 5, 2011 at 10:16 PM, Ricardo Kleemann wrote:
Hi,
I'm trying to make ldirectord work for smtp and having some issues.
Basically I have 2 servers which themselves run virtualization (via
virtualbox).
Server1 has a virtualized server running smtp (lets call it smtp1 ->
192.168.1.204)
Server2 also has a virtualized server running smtp (lets call it
smtp2 -> 192.168.1.102)
ldirectord is running on Server1, ip 192.168.1.200
So I have in my config:
virtual=192.168.1.200:25 [2]
real=192.168.1.102:25 [3] masq
real=192.168.1.204:25 [4] masq
service=smtp
scheduler=wlc
protocol=tcp
#persistent=600
# #netmask=255.255.255.255
checktype=connect
I have no problems getting to smtp1 and smtp2 from the ldirectord
server:
$ telnet 192.168.1.204 25
Trying 192.168.1.204...
Connected to 192.168.1.204.
Escape character is '^]'.
220 email1 ESMTP
$ telnet 192.168.1.102 25
Trying 192.168.1.102...
Connected to 192.168.1.102.
Escape character is '^]'.
220 maillgw2 ESMTP MailCleaner (Community Edition 2010 beta 3) Sat,
05 Mar 2011 17:08:46 -0800
However, ldirectord only flows connections to the virtual server that
is local to itself (smtp1). No matter what I do, it will never use
the
other smtp server (smtp2)..
ldirectord considers both as valid and adds weight to them
sysadmin@vbox:/etc/bind/master$ sudo ipvsadm -L
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight
ActiveConn InActConn
TCP vbox:smtp wlc
-> email1:smtp Masq
1 1 1
-> mailgw2:smtp Masq
1 0 1
But again, never routes to the other smtp.
So I tried removing the smtp1 from the equation, and then even though
ipvsadm reports it is available for forwarding, that never works. I
never get the connection. If smtp1 is in the forwarding list, then it
works.
What do you do to take it out of equation ?
What do you mean it never flows to the other server ?
Try like this:
From another server do it about 5 times and post the results!
$ telnet 192.168.1.200 25
After it, go into the first server and shut down mail server and see
if ldirector set it weigh to 0 if it does, it should only goes to next
server.
So for tests porpuses:
Shut down one mail sever and repet the test above.
WITHOUT restarting it, shutdown the other mail server, and repeat
thet test
Now you can turn one the firt server again and repeat the teste.
When you post here the results i might be able to help!
Any idea what could be wrong?
I did not understand pretty well what is wrong. Lets see if you got
my ideia and will test it well!
Thanks
Ricardo
Good luck!
_______________________________________________
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
Links:
------
[1] http://email1.americasnet.com
[2] http://192.168.1.200:25
[3] http://192.168.1.102:25
[4] http://192.168.1.204:25
_______________________________________________
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
|