This content has been marked as final. Show 16 replies
check the /etc/sysctl.conf file and check if contains following line:
net.ipv4.ip_local_port_range = 1024 65000
If not, then add that line save the file and reboot the system.
Also check the wheter some service is not running on required ports that was mentioned in the error messages.
You can check this issuing the "netstat -an | grep tcp | grep LISTEN" command.
Im trying to install Oracle Application Server 10g, when trying to configure assistants I receive the same error but on a Windows 2003 Server.
DataBase Configuration Assistant:
Failed to allocate port(s) in the specified range for the following process(es): JMS [5540-5559],RMI [5520-5539],Database Control [5500-5519],EM agent [1830-1849]
I set (in windows registry) the MaxPortUser key to 65534 and TcpTimedWaitDelay key to 30 secs , but it still not working (same error).
thanks in advance,
I had the same trouble when creating an Oracle 10.1.0.4 database on RedHat Enterprise Linux 4.
I was successfully install a database on a machine, but not to the other.
I found the difference between the two machine. It was the /etc/hosts file.
I fix that file and after, emca finished without any error.
On windows, the hosts file is located in system32/drivers/etc I think
Hope that helps.
i 've triple checked all requirements and my /etc/services file and everyhting is fine except runInstaller breaks with the same (posted by multiple people above) message.
FYI Ivan: your post from april 20th doesn't help at all. i mean it might help somebody who doesn't read docs, but the rest of the story is missing from there.
while searching online i found that a number of people ran into the same problem.
the proposed workaround is to run emca -h etc..
emca -RMI_port<port> etc
but what is the problem with hosts file? anybody?
I'm joining this thread late as I've just encountered the same problem.
None of the ports specified in the error message are in the etc/services file, and none of them are in use according to netstat either. sysctl.conf and ip_local_port_range are both correct too.
SEVERE: Failed to allocate port(s) in the specified range for the following process(es): JMS [5540-5559],RMI [5520-5539],Database Control [5500-5519],EM agent [1830-1849]
Werner Puschitz has something about this on his Suse installation page in the Troubleshooting section (http://www.puschitz.com/InstallingOracle10gOnSUSE.shtml ) , but is does not seem to have got into his Red Hat page:
The problem is that ports listed in /etc/services are discounted.
The workaround is to specify the ports manually, see emca -h.
Use the following options and use the port ranges as specified in the above error message:
emca -RMI_PORT <port> -JMS_PORT <port> -AGENT_PORT <port> -DBCONSOLE_HTTP_PORT <port>
so it may be worth giving that a try.
I experienced a same failure recently during my trial installations. I didn't have the error at the first two tries. But when I tried third time, I got this error. That's why I found this thread. I checked everything suggested here and found no solution still. Later, I figured out that the error actually was an indication of network failure. It turned out my eth0 mysterously quited working prior my installation. (Do a "ifconfig" and you can see it.) And the database configuration scripts actually looked for an active card and therefore failed. Once I fixed the problem in the network card, this problem went away.
Just want to let you know that may hints and clues provided here are correct, but sometimes may not be "complete".
I recently ran into this problem and found the response from email@example.com on May 10, 2005 very helpful. Thanks a lot.
My problem had to deal with an incorrect entry in the /etc/hosts file. We had originally loaded this machine using a dynamic IP. Later we changed it to a static IP and when I went to reload Oracle I ran into this problem. Once I corrected the /etc/hosts file emca ran without any problems.