Oracle Databases Instances are getting discovered without instances name

Santosh Kumar8 · ‎02-13-2020

Hi All,

I am getting this issue with Oracle database instances discovery where database instances are getting discovered without instance name. This issue i am getting when using patterns.

Its difficult to identify database without name as there can be more than one instances on one host

SiD2 · ‎04-20-2020

Thanks for confirming, this is an open issue only in Newyork version and we fixed in Orlando.

Actually the wrong identification section is getting executed because of the LINUX os type check.

Kindly remove the LINUX os type check as this identification section is mean to run only on Solaris OS. So remove that condition so that proper 2nd identification section runs.

Save and publish the pattern, After syncing pattern with mid try running fresh discovery on your linux machine, you should not see this issue anymore.

We have a KB article as well for the same.

https://hi.service-now.com/kb_view.do?sys_kb_id=551d275bdb7bcc98414eeeb5ca96196f

Please mark Helpful / Accept Solution so that it helps others with similar questions.

View solution in original post

Santosh Kumar8 · ‎06-02-2020

Thanks @DuaneNMore for explaining this in this detail.

I did check this as you suggested. In my case for solaris server i dont see any processes under that related list. However i do see this for linux servers and there i am discovering oracle instance.

For solaris none of the processes are getting added under related list. please check these snapshopt.

Should should i check or do for getting process added. In discovery logs i do see lot of PIDs getting identify post Solaris ADM but after that no pattern launch for DB instance discovery.

I also see this error sometime on solaris

<?xml version="1.0" encoding="UTF-8"?><results probe_time="301311" result_code="0"><result id="e71cd26fa9fe156100b4b355bf835b9c" name="sh ${file:ps.sh} ${zone}" order="1" topic="SSHCommand"><results error="Error; job finished with status ERROR: SSH channel 69 command timeout in state EXECUTING. (timeout = 300000) com.snc.ssh.SSHTimer@3da18941 while in state EXECUTING" probe_time="300467" result_code="900000"><result error="Error; job finished with status ERROR: SSH channel 69 command timeout in state EXECUTING. (timeout = 300000) com.snc.ssh.SSHTimer@3da18941 while in state EXECUTING"><output>PID PPID COMMAND

Sorry, try again.

sudo: 3 incorrect password attempts

5171 1

Sorry, try again.

sudo: 3 incorrect password attempts

9594 1

If i am missing something. Please do help me

On Solaris

Santosh Kumar8 · ‎06-02-2020

On Linux i do see all processes

DuaneNMore · ‎06-02-2020

Yeah Looks like ps does not have access to run on Solaris either. That is usually what happens when it complains about sudo and times out. You should find out if the sudoer file has ps in it.

SiD2 · ‎05-12-2020

Can you mark this comment helpful as well?

Please mark Helpful / Accept Solution so that it helps others with similar questions.

Santosh Kumar8 · ‎06-01-2020

Hi SID, i am still couldn't get this resolved.

This Solaris ADM probe look for lsof. so you are saying even if we dont have this access it should be fine as long as we have 'ps" access for active processes.

Solaris - Active Connections

lsof.sh

lsof -iTCP -n -P -F pcnfT -z$1