This content has been marked as final. Show 9 replies
If ssh displays a banner during login (something like "you are not allowed to access this computer blah blah blah ...", it may be that you need to include the banner text in the $OMS_HOME/prov/resources/ignoreMessages.txt. ( I presume you are using Unix/Linux).
Alternatively, on Linux for example, you can edit (as root user) the /etc/ssh/sshd_config and comment out the line that start Banner /etc/...... Then stop and restart the sshd daemon.
I'm sorry, the environment is Solaris - so yes Unix.
Sadly, there is no banner. The first thing that comes up is asking for the password of the currently signed in user id (from the client).
User equivalence means one of two things:
1) The user id, group id etc. much match between the two host (this is the rule for RAC systems)
2) SSH must be setup with a shared key, so you do NOT need to provide a username/password
Check out the requirement for your installation. If in doubt make both scenarious true ;)
I have the same problem. I read the documentation "Grid Control Installation and Basic Configuration" and many metalink documents like Note:376355.1 - "Known issues with Agent Push Deploy Application".
I tried all "things". I opened a SR to and till now no answer.
What I found out is that SSH runs.
The command: 'ssh <server> date' returns the expected output and in the ssh_log writes: "Oct 16 15:59:24 devo sshd: [ID 800047 local7.info] Accepted publickey for oracle from 172.30.4.8 port 46124 ssh2"
When I start Agent Deploy I receive the error message "SSH User equivalence is not set up on the remote nodes" and in the ssh_log: "Oct 17 16:19:21 devo sshd: [ID 800047 local7.info] Did not receive identification string from 172.30.4.8"
Bye & Thanks
I found out the problem!
Examining some logs in the directory: /oracle/10.2.0_EM/oms10g/sysman/prov/agentpush/<timestamp>/prereqs/local
I discovered that Agent Deploy couldn't find scp and ssh. However the file's path in sPaths.properties was correct. Strange! As test I created a symbolic link and the problem has been solved.
Where did you setup the symbolic links?
i am also facingthe same issue can u tell me where was the symbolic link created
I had a similar problem when running runcluvfy.sh for setting RAC.
I had to add the public key to both nodes (for the oracle user) in order to allow a passwordless loginto the remote node and to the node where running runcluvfy.sh (_which means u have to be able to ssh localhost without password!_)
Hope it helps
run the sshUsersetup.sh to set the ssh User equivalance
<OMSHOME>sysman/prov/resources/scripts/ sshUserSetup.sh -user <> -hosts <hostname separate by space>