Oracle EM. emctl start dbconsole fails: connection refused.

deanbrown3d · 08-24-2004, 10:37 AM

Please could someone help me here? I put Ora10g onto Fedore Core 2 Linux and all is well I can start up a database and create tables etc.

Now I want to run OEM Database Control on the same local machine. I start the listener (however it says status unknow, see below if this could be a problem) and then run:

$ emctl start dbconsole

It puts up all the "........" dots for about 2 mins, and then states FAILED. Here is the first error message in the trace

/app/oracle/product/10.1.0/db_1/localhost.localdomain_ACT1/sysman/log/emdctl.trc)

TRACE:

2004-08-24 10:09:21 Thread-4144096448 WARN http: snmehl_connect: connect failed to (localhost.localdomain:1830): Connection refused (error = 111)
2004-08-24 10:09:26 Thread-4144653504 WARN http: snmehl_connect: connect failed to (localhost.localdomain:5500): Connection refused (error = 111)
2004-08-24 10:09:29 Thread-4144239808 WARN http: snmehl_connect: connect failed to (localhost.localdomain:5500): Connection refused (error = 111)
2004-08-24 10:09:32 Thread-4144194752 WARN http: snmehl_connect: connect failed to (localhost.localdomain:5500): Connection refused (error = 111)

And the emagent trace starts with:

2004-08-24 10:10:53 Thread-4144567232 WARN http: snmehl_connect: connect failed to (localhost.localdomain:5500): Connection refused (error = 111)
2004-08-24 10:10:53 Thread-4144567232 ERROR pingManager: nmepm_pingReposURL: Cannot connect to http://localhost.localdomain:5500/em/upload/: retStatus=-32
2004-08-24 10:10:53 Thread-4144567232 WARN command: Job Subsystem Timeout set at 600 seconds
2004-08-24 10:10:53 Thread-66694064 WARN http: snmehl_connect: connect failed to (localhost.localdomain:5500): Connection refused (error = 111)
2004-08-24 10:10:53 Thread-66694064 ERROR command: nmejcn: failed http connection to http://localhost.localdomain:5500/em/upload/: retStatus=-32
2004-08-24 10:10:53 Thread-4144567232 WARN upload: Upload manager has no Failure script: disabled

Could anyone please tell me where to start here? Is there a serive or something that I need to start before running the command?

Thanks very much,

Dean Brown

LSNRCTL> start
Starting /app/oracle/product/10.1.0/db_1/bin/tnslsnr: please wait...

TNSLSNR for Linux: Version 10.1.0.2.0 - Production
System parameter file is /app/oracle/product/10.1.0/db_1/network/admin/listener.ora
Listening on: (DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.0.177)(PORT=1521)))
Listening on: (DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(KEY=ORCL)))

Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=TCP)(HOST=192.168.0.177)(PORT=1521)))
STATUS of the LISTENER
------------------------
Alias LISTENER
Version TNSLSNR for Linux: Version 10.1.0.2.0 - Production
Start Date 24-AUG-2004 11:18:19
Uptime 0 days 0 hr. 0 min. 0 sec
Trace Level off
Security ON: Local OS Authentication
SNMP OFF
Listener Parameter File /app/oracle/product/10.1.0/db_1/network/admin/listener.ora
Listening Endpoints Summary...
(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.0.177)(PORT=1521)))
(DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(KEY=ORCL)))
Services Summary...
Service "ACT1" has 1 instance(s).
Instance "ACT1", status UNKNOWN, has 1 handler(s) for this service...
The command completed successfully

deanbrown3d · 08-24-2004, 03:46 PM

Anyone? I've been trying all day to get this to work

I installed the oracle database companion CD and then tried it again and it worked! But then I had to restart and since then the damn thing has gone back to failing again.....

Bjorkli · 09-14-2004, 05:30 AM

Did you follow these instructions?

http://staff.in2.hr/denis/oracle/10g...edora1_en.html

or these:

http://www.oracle-base.com/articles/...nOnFedora1.php

(Ok. These are for Fedora 1, not 2, but still... should work... )

Then I made a file called oracle in /etc/init.d folder to make it start at reboot, and symlink it to rc5.d (ln -s /etc/init.d/oracle /etc/rc5.d/S95oracle I think it was) Please know that the folder structure used below might be different than standardsetup, and user might not be "oracle" on your machine). Below is db_1 the oracle installation and db_2 is the companion CD with Apache.

Code:

[oracle@linux init.d]$ cat oracle
#!/bin/bash
#
# chkconfig: 35 95 1
# description: init script to start/stop oracle database 10g, TNS listener, EMS
#
#
# match these values to your environment:
export ORACLE_BASE=/u01/app/oracle
export ORACLE_HOME=$ORACLE_BASE/product/10.1.0/db_1
export ORACLE_TERM=xterm
export PATH=$ORACLE_HOME/bin:$PATH
export ORACLE_SID=TSH1
export DISPLAY=localhost:0
export ORACLE_USER=oracle

# see how we are called:
case $1 in
    start)
    su - "$ORACLE_USER"<<EOO
    lsnrctl start
    sqlplus /nolog<<EOS
    connect / as sysdba
    startup
EOS
    emctl start dbconsole
EOO
    ;;

    stop)
    su - "$ORACLE_USER"<<EOO
    lsnrctl stop
    sqlplus /nolog<<EOS
    connect / as sysdba
    shutdown immediate
EOS
    emctl stop dbconsole
EOO
    ;;

    *)
    echo "Usage: $0 {start|stop}"
    ;;
esac

I also did following to make the apache server kick in at reboot.

Code:

 [oracle@linux init.d]$ cat apache
#!/bin/bash
#
# chkconfig: 35 95 1
# description: init script to start/stop Apache 1.3 web server
#
#
# match these values to your environment:
export WEB_HOME=/u01/app/oracle/product/10.1.0/db_2/opmn/bin/
export WEB_TERM=xterm
export DISPLAY=localhost:0
export WEB_USER=oracle

# see how we are called:
case $1 in
    start)
    su - "$WEB_USER"<<EOO
    /u01/app/oracle/product/10.1.0/db_2/opmn/bin/opmnctl start
EOO
    ;;

    stop)
    su - "$WEB_USER"<<EOO
    /u01/app/oracle/product/10.1.0/db_2/opmn/bin/opmnctl shutdown
EOO
    ;;

    status)
    su - "$WEB_USER"<<EOO
    /u01/app/oracle/product/10.1.0/db_2/opmn/bin/opmnctl status
EOO
    ;;

    *)
    echo "Usage: $0 {start|stop|status}"
    ;;
esac

exit 0

(Not sure where I picked up these scripts, but thanks whoever it was).

Again, symlink it using ln -s /etc/init.d/apache /etc/rc5.d/S96apache

And of course you have to make sure that chown (owner) and chmod (privs) is set like the other files in the /etc/init.d and /etc/rc5.d folder or else root or oracle user might not be able to run it.

Hmmm. Not sure if this starts the 5500 server and the 1830 server however, since I use only the "sqlplus" command to do all my administration, and not the administration webpages. Um... Maybe I just wasted your time here. But hopefully the 2 links at top of this post provides better info. If not, try google with Fedora 2 Oracle 10g, found the following there:

http://www.oracle-base.com/articles/...nOnFedora2.php (Did you try that one ?)

deanbrown3d · 09-14-2004, 08:56 AM

Nope I gave up on trying. It was very obviously buggy, and out of my control. What I am doing instead is installing the enterprise manager console on a windows machine and access the database from there, and it works perfectly. (Not the web-based version, just the enterprise manager, which is perfect for my needs.)

Thanks for the help,

DeanB

RageIlluminati · 01-31-2005, 03:01 AM

I use Oracle10g on RH Linux, I had same errors in trace file, after unexpectedly my dbconsole crashed. I tried to restart it, but it wasn't start again, then I noticed that some java process were up to use protsessor.. after killing that protsess my dbconsole worked well. IMO this was some perl script in java that EM runs on start.