UCMDB and UD Practitioners Forum (Previously CMS)
cancel
Showing results for 
Search instead for 
Did you mean: 

What are the steps to find the detailed errors relating to ucmdb discovery probes?

Highlighted
lwvirden
Frequent Visitor

What are the steps to find the detailed errors relating to ucmdb discovery probes?

I am trying to learn about ucmdb.

We have several discovery probes defined.

From those, we see there are a number of errors.

When I click on the number of errors, I get a brief (less than 5 words usually) indication of some issue that has occurred - connection time out, authentication failed, etc.

 

However, I don't know how to see the specifics of the failure so that when I talk to a DBA, I can say "on machine xyz, the probe attempted to log in as ABC but got the SQL Server error "password failed"

or whatever the specifics are.

 

If I go and say "hey, some sql server just says that a ucmdb authentication failed, but we have a dozen or more logins doing stuff - can you see what server and instance is reporting the problem" I get a look like I am crazy and am told "No, I need more information if I am to help you".

 

So,  how do I get more information about the errors?

10 REPLIES
Strawberry1
Super Collector

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

Hi

 

The information you provide is very sumier. Which version uCMDB are you using. On what platform?

 

If you look in the Data Flow Probe Setup, all probes are connected and no errors or warnings are shonw.

Is this correct? 

 

I assume that the probes are installed correctly.

Then...

 

I assume that get errors when running jobs, because you say you have some issues with connection time out, access denied etc.

 

More information you might find in the communication log. Go to the job, select a destination, right click, select Debug and choose Communication Log. Scroll down and look for the error. 

Or the following log files on the probe: probe-error.log and WrapperProbeGW.log located in <Ucmdb Home folder>\DataFlowProbe\runtime\log

 

Little help with the errors:

- connection timeout: Can you ping the destination from the probe. This might be a network issue and the probe can not reach the destination. That is where I would start.

- Acces Denied: Wrong password is usualy the case. Check the credentials.  Again. Data Flow Probe Setup > DefaultDomain (Default) > Credentials > Choose the protocol > Select the Credentials > Right click > Check Credentials.

Fill in the ip of the destination and press ok. See if it is Successful

 

Add screenshots if nothing helps.

 

Good luck

lwvirden
Frequent Visitor

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

This is HM uCMDB 10.20 on Windows.

lwvirden
Frequent Visitor

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

You wrote:

More information you might find in the communication log. Go to the job, select a destination, right click, select Debug and choose Communication Log. Scroll down and look for the error.

 

I don't understand the steps.

I am in Data Flow Management > Un iversal Discovery > Zone Based Discovery

I expand the zone where our probes are set.

I click rignt on one of these and there is no destination.

 

So then I click on Discovery Modules/Jobs tab.

I expand the Discovery MOdules > Network Infrastructure > Host Connection for example, because of some of the connection errors.

I click right on Host Connection - again, nothing called destination there.

 

So I don't see where to select Debug and Communication Log.

 

 

Strawberry1
Super Collector

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

Hi,

 

I never use the Zone-Based Discovery. But that is personaly, I think.

In "Discovery Modules/Jobs" and locate the job you want to run. In this case "Host connection by ?"

 

You select the job and in the right "Discovery Progress" pane you see numbers. Click on the Total of Total number (bottom left one).

In the status column you see the status. Select one that is not success. From there you right click and select Debug and choose Communication Log. Scroll down and look for the error. 

See the screenshot I attached.

 

Did you manage to set and check the credentials in Data Flow Probe Setup?

 

Let me know if you need more assistance.

lwvirden
Frequent Visitor

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

Thank you for your ideas.

 

The primary on this effort set up zone based discovery initially based on recommendations by someone with more experience than we had.

 

When I click over to Discovery Modules > Database > Connections using Host Credentials (one of the places I know has an error), I see this.

 

As you can see, the total of totals is 0.

I want to see the specifics of the error - I figure it would be good to resolve all errors and then see how things go.

 

So I click right on the 1. I see Show trigger count and Show errors.

My menu looks nothing like your screen shot.

I go ahead and select Show errors.

I get this.

An "Internal error".

 

That isn't, by itself, a helpful message.

 

So how do I get to the information that either tells me what is wrong, tells me how to fix it, or at least gives me something useful to report to support?

 

 

 

lwvirden
Frequent Visitor

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

When I look into the wrapperProbeGW.log file, I see this in relationship to the machine in question:

 

jvm 1    | <2015-06-07 11:09:20,760> 325952036 [ERROR] [JobExecuterWorker-76:MSSQL Server Connection by SQL_134.243.94.18] (?:?) - Failed to connect using all protocolsWhen

 

I don't know how to get details such as what specific action is being attempted with what specific user id.

 

In the probeMgr-adaptersDebug.log I see this

 

<2015-06-08 16:58:18,273> [DEBUG] [JobExecuterWorker-67:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.198.4] - Starting script HostConnectionByShell.py^M

<2015-06-08 16:58:18,367> [DEBUG] [JobExecuterWorker-0:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by PowerShell_134.243.198.136] - Fixing encoding.. from :Windows-1252, to: IBM437^M

<2015-06-08 16:58:18,367> [DEBUG] [JobExecuterWorker-0:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by PowerShell_134.243.198.136] - Fixed string: Active code page: 437^M^M

<2015-06-08 16:58:18,382> [WARN ] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - Error while performing stop and uninstall HPCmd service on 134.243.94.18. details: Access is denied.^M

^M

<2015-06-08 16:58:18,398> [ERROR] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - Failed to cleanup after task execution failed. details: Access is denied.^M

^M

<2015-06-08 16:58:18,398> [DEBUG] [JobExecuterWorker-9:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by PowerShell_134.243.5.84] - powershell: Invalid user name or password^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - NTCMD: Internal error. Details: Access is denied.^M

^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - Reporting warning code 205 to framework.^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-9:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by PowerShell_134.243.5.84] - Reporting error code 302 to framework.^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - Warning message is: SSH: Connection refused.^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-9:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by PowerShell_134.243.5.84] - Error message is: powershell: Invalid user name or password^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - Reporting warning code 205 to framework.^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - Warning message is: Telnet: Connection refused.^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-54:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by Shell_134.243.94.18] - Reporting error code 103 to framework.^M

<2015-06-08 16:58:18,414> [DEBUG] [JobExecuterWorker-9:MZ_Test Mgmt Zone_New Infrastructure Discovery_Network_Host Connection by PowerShell_134.243.5.84] - Finished script Host_Connection_by_powershell.py^M

 

I see an accessed denied message in this stream of messages. I'm stil not clear on what specific action(s) are being attempted with what user id . I certainly could send these messages to the DBA of the machine. I am pretty certain the DBA won't know what it means, however, or what to do about it.

 

I see errors referring to NTCMD, powershell, and telnet. But I can't see enough details to see what specific commands were being executed when the errors occurred. 

 

Does any of this seem familar?

Is there some flag or setting I can turn on that would give me more information?

 

 

 

Strawberry1
Super Collector

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

Hi,

 

In the first log (wrapperProbeGW.log) you took out the following error:

Failed to connect using all protocols.

 

The protocols for this job are NTCMD, SSH, Telnet etc.

 

When you place this next to the other log (probeMgr-adaptersDebug.log), I see only credential problems:

 

powershell: Invalid user name or password

NTCMD: Internal error. Details: Access is denied.

Warning message is: SSH: Connection refused.

 

The job tries to login via xcmd to retrieve information of the queried host, but fails.

 

Check the credentials.

 

To make sure that you use the correct credentials check them in the Data Flow  Probe Setup. (See attachment)

 

Good luck

 

 

 

 

 

lwvirden
Frequent Visitor

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

1. I have never configured any part of ucmdb before. 

2. The person who has, did so using the Zone based discovery, not the modules/jobs directly,

3. Following your last comment, I went into the Data Flow Management > Data Flow Probe Setup > Domains and Probes > Default Domain > Data Flow Probes.

4. There were 9 protocols which had entries in the right hand panel.

5. I selected the check credentials menu option on each entry

6. Each entry reported

The credentials cannot be checked because there are no available Data Flow Probes

 

Now, on the other hand, when I go to Universal Discovery, and look in either tab, what I see are the results of probes.  So there are probes available.

 

Of those probes, I see 28 errors - which is what I am trying to resolve.

I am getting connection failed, Failed to connect using all protocols, Internal error, in valud user name or password, and no agent on the remote machine.

 

I can't fix a user name and password until I can figure out what user name and password is not working.

Which is why I am trying to figure out where the credentials are that are being used.

 

Sorry - I just am having quite a bit of trouble understanding this application.

 

Strawberry1
Super Collector

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

Hi,

 

I looked at the zond based discovery. It looks more like a wizard, which enables the right jobs for the discovery.

Just a tip - Try to use the modules and jobs directly. You can actived the jobs one by one and therefore have more control over what you are discovering.  

 

Any way.

The credentials are set in the Data Flow Probe Setup under credentials and protocol.

The same place where you checked.

 

Depending on your company security policies, the account may lock if you try invalid credentials after three times. 

 

When a succesfull connection is made to a host with a protocol, uCMDB stores and use this information. Other credentials for this protocol are skipped.

 

Did you try to login on the host manually, outside the uCMDB? Ie. via microsoft Terminal Client (mstsc). To verify you have the right credentials.

 

Also, did you checked if the probes are connected to the uCMDB server?
I do not know your configuration, Once a customer configured the probes service not to start automatically. Therefore after a reboot of the probe server the service did not start the probe. 

 

It is normal that this application gives you headaches in the beginning. It will pass after six months.  ;o)

 

Good luck

lwvirden
Frequent Visitor

Re: What are the steps to find the detailed errors relating to ucmdb discovery probes?

I am getting errors from probe discoveries, as well as CIs, etc., so the probes appear to be connected.

I have figured out how to find the specific logins being used - now I need to talk to people to determine why the logins appear to be failing.

 

Thank you for your help.

//Add this to "OnDomLoad" event