Re: Error -908 on network connection (IDS 7.30)
Posted in 1999
Topics: Installation, Setup & Upgrades, Connectivity: ODBC / JDBC / .NET, Server Administration, Security, Permissions & Auditing, Transactions, Locking & Isolation, Networking & sqlhosts Configuration, Platform-Specific Issues, Versions, Editions & End-of-Life
Hi,
I suppose you've forgot to post "netstat -na"
I'd like to see there something like:
tcp 0 0 192.168.0.229.:1526 *:* LISTEN
If you get instead of this the entry:
tcp 0 0 127.0.0.2:1526 *:* LISTEN
you will not be able to make the connection, since the server
is listening for connections on your local interface 127.0.0.2
and don't care about your network interface 192.168.0.229.
If it is true - change the host name in your sqlhosts file to
be a name for the IP address 192.168.0.229
In fact 127.0.0.2 looks for me very strange. But this is another problem
and if it doesn't make you any problems - it's Ok.
Best Regards,
Octav
Kai Mehrtens wrote:
>
> Hello everybody,
>
> I have an (obviously quite common) problem with connecting from a
> client (for test purposes ILogin Demo) to an Informix Dynamic Server
> 7.30 database server. Here is my configuration:
>
> Server side:
> Linux (SuSE 6.1 with Kernel 2.2.10)
> hostname: bassin.hec.de
> IP-address: 192.168.0.229
> uname -a:
> Linux bassin 2.2.10 #4 Thu Oct 7 09:50:15 MEST 1999 i586 unknown
> $INFORMIXDIR=/opt/informix
> $INFORMIXSERVER=bassin
>
> sqlhosts file (/opt/informix/etc/sqlhosts):
> bassin onsoctcp bassin idstcp
> bassin2 onipcshm bassin idstcp>
> onconfig file (/opt/informix/etc/onconfig, extraction):
> [...]
> SERVERNUM 0
> DBSERVERNAME bassin
> DBSERVERALIASES bassin2
> NETTYPE onsoctcp,1,20,NET
> DEADLOCK_TIMEOUT 60
> RESIDENT 0
> [...]>
> /etc/hosts.equiv file:
> MEA.hec.de
> 192.168.0.101
>
> /etc/services file (extraction):
> [...]
> idstcp 1526/tcp
>
> .rhosts file of user mea (~mea/.rhosts):
> MEA.hec.de
> 192.168.0.101
> bassin.hec.de
>
> client side:
> Hostname: MEA.hec.de
> IP-dddress: 192.168.0.101
> Windows NT 4.0 Service Pack 4
> Informix Client SDK Version 2.30 tc1.1 for intel pentium nt 4.0.
> (I installed the whole package, including the two ODBC-drivers
> Informix 3.30.00.10139 32 BIT and INTERSOLV 3.11.00.21 32 BIT
> INFORMIX 9).
>
> Setnet32 Settings:
> -Environment:
> INFORMIXDIR=d:\\programme\\informix\\sdk
> INFORMIXSERVER=bassin> All other entries are left blank.
> -Server Information:
> Informix Server: bassin
> HostName: bassin.hec.de
> Protocolname: onsoctcp
> Service Name: idstcp
> Options: (left blank)
> -Host Information
> Current Host: bassin
> User Name: mea
> Password Option: Password
> Password: My correct password that I won't post here ;-)
>
> Registry entries:
> HKEY_LOCAL_MACHINE/Software/Informix/SqlHosts/bassin:
> -HOST "bassin"
> -PROTOCOL "onsoctcp"
> -SERVICE "idstcp"
> HKEY_CURRENT_USER/Software/Informix/Environment:
> -DBANSIWARN "n"
> -DELIMIDENT "n"
> -INFORMIXDIR "d:\\Programme\\Informix\\sdk"
> -INFORMIXSERVER "bassin"
>
> Services file (C:\\WINNT\\System32\\drivers\\etc, extraction):
> [...]
> idstcp 1526/tcp
> [...]
>
> There is no sqlhost or onconfig file on the client.
>
> The network connection between both computers works fine. On the
> server side, I can do ping MEA or ping MEA.hec.de:
> PING MEA.hec.de (192.168.0.101): 56 data bytes
> 64 bytes from 192.168.0.101: icmp_seq=0 ttl=128 time=0.556 ms
> 64 bytes from 192.168.0.101: icmp_seq=1 ttl=128 time=0.382 ms
> 64 bytes from 192.168.0.101: icmp_seq=2 ttl=128 time=0.387 ms
> --- MEA.hec.de ping statistics ---
> 3 packets transmitted, 3 packets received, 0% packet loss
> round-trip min/avg/max = 0.382/0.441/0.556 ms
>
> The same thing works the other way around:
> PING wird ausgef'hrt f'r bassin [192.168.0.229] mit 32 Bytes Daten:
>
> Antwort von 192.168.0.229: Bytes=32 Zeit<10ms TTL=255
> Antwort von 192.168.0.229: Bytes=32 Zeit<10ms TTL=255
> Antwort von 192.168.0.229: Bytes=32 Zeit<10ms TTL=255
> Antwort von 192.168.0.229: Bytes=32 Zeit<10ms TTL=255
>
> I first made a test with the dbaccess tool on the linux host to
> determine if the database server works correctly, and found out that it
> works fine. I started dbaccess as user "mea", select
> Connection/Connect, pressed enter to select the first database
> (bassin), entered my user name ("mea") and my password and after a few
> seconds, I got the choice of available databases on the server. So I
> could e.g. choose "store7@bassin" and everything was fine. To be sure
> the connection was made via tcp/ip sockects, I watched the socket ports
> using netstat -tuonc. Here I could observe the following behaviour:
> 1. Before the connection was established, I got the output:
>
> Active Internet connections (w/o servers)
> Proto Recv-Q Send-Q Local Address Foreign Address State
> Timer
> tcp 0 0 127.0.0.2:1526 127.0.0.2:1101 CLOSE
> on2 (2.04/0/7)
> tcp 0 130 192.168.0.229:23 192.168.0.101:1043 ESTABLISHED
> on1 (0.36/0/0)
> tcp 0 0 192.168.0.229:23 192.168.0.101:1042 ESTABLISHED
> off (0.00/0/0)
>
> [The first entry shows the state of the Informix server (the other two
> entries are probably my two telnet sessions on the linux host)]
>
> 2. After starting dbaccess, the output changed to:
>
> Active Internet connections (w/o servers)
> Proto Recv-Q Send-Q Local Address Foreign Address State
> Timer
> tcp 0 0 127.0.0.2:1526 127.0.0.2:1102 ESTABLISHED
> off (0.00/0/0)
> tcp 0 0 127.0.0.2:1102 127.0.0.2:1526 ESTABLISHED
> off (0.00/0/0)
> tcp 0 130 192.168.0.229:23 192.168.0.101:1043 ESTABLISHED
> on1 (0.40/0/0)
> tcp 0 0 192.168.0.229:23 192.168.0.101:1042 ESTABLISHED
> off (0.00/0/0)
>
> [As one can see, the connection has been properly established using
> tcp sockets.]
>
> 3. After disconnecting, the state changed again to:
>
> Active Internet connections (w/o servers)
> Proto Recv-Q Send-Q Local Address Foreign Address State
> Timer
> tcp 0 0 127.0.0.2:1526 127.0.0.2:1102 CLOSE
> on2 (1.44/0/7)
> tcp 0 0 127.0.0.2:1102 127.0.0.2:1526 TIME_WAIT
> unkn-3 (7.50/0/0)
> tcp 0 130 192.168.0.229:23 192.168.0.101:1043 ESTABLISHED
> on1 (0.38/0/0)
> tcp 0 0 192.168.0.229:23 192.168.0.101:1042 ESTABLISHED
> off (0.00/0/0)
>
> And after a timeout, the second entry disappeard, so that a state
> similar to that before the connection was reached.
>
> As you can see, the database server generally accepts connections (at
> least from the local host). Unfortunately I do not have a second Linux
> box to test a real (non-loopback) network connection using dbaccess.
> However, I have an NT
Hello Octav, In article <7uhsn9$ngu$1@news.xmission.com>, Octav Chiriac <com@netinfo-moldova.com> wrote: > > Hi, > > I suppose you've forgot to post "netstat -na" > I'd like to see there something like: > tcp 0 0 192.168.0.229.:1526 *:* LISTEN > > If you get instead of this the entry: > > tcp 0 0 127.0.0.2:1526 *:* LISTEN > > you will not be able to make the connection, since the server > is listening for connections on your local interface 127.0.0.2 > and don't care about your network interface 192.168.0.229. > > If it is true - change the host name in your sqlhosts file to > be a name for the IP address 192.168.0.229 > > In fact 127.0.0.2 looks for me very strange. But this is another problem > and if it doesn't make you any problems - it's Ok. > [...] In fact, you are right. After I changed the IP-address of my Linux-host called bassin from 127.0.0.2 to 192.168.0.229 in /etc/hosts, everything worked fine. The strange 127.0.0.2 address is the default setting made by the SuSE installation. I suppose this has something to do with the fact that we use DHCP to assign IP-addresses to hosts. But that's another problem and not subject to this newsgroup. Thanks a lot for your quick and competent help :-) Best regards, Kai Sent via Deja.com http://www.deja.com/ Before you buy.