cancel
Showing results for 
Search instead for 
Did you mean: 

Error while indexing the imported product data

laurent_burtaire
Active Contributor
0 Kudos

Hello,

when trying to index imported SRM Product Master Data, for some Search Object Connectors, we have an indexing error in ESH_COCKPIT transaction:

 

I checked the job log, but there is no sufficient info to understand what is the issue...

Do any other logs are available?

If yes, where can i find them (which transaction)?

Thank you for your help.

Regards.

Laurent.

Accepted Solutions (1)

Accepted Solutions (1)

laurent_burtaire
Active Contributor
0 Kudos

Hello,

in the TREXADMIN, 'Services' tab, i can see service "httpserver" is inactive:

Is this could be a reason of my issue?

In the 'Trace' tab, non of the trace files were modified in 2015:

Regards.

Laurent.

former_member217429
Active Contributor
0 Kudos

Hi Laurent,

the HTTPServer isn't used in your scenario . If there are no any other applications connected to this TREX (f.e. Enterprise Portal or CRM ISA) you can remove this service from TREXAdminTool -> Connectivity -> HTTP -> Use Http Communication = no.
Could you please check in the python TREXAdminTool (or ta. TREXADMIN) if the ICM works at all.

Best regards,
Mikhail

laurent_burtaire
Active Contributor
0 Kudos

Hello Mikhail,

i have an error status for ICM connectivity:

With the advanced test, for ICM connectivity, i noticed for RFC communication method no problerm.

But for ICM communication method, i have an issue (with same host nama and port as for RFC).

I already checked OSS notes:

1270808 - TREX for Embedded Search: Remote communication error

1501222 - TREX 7.00 & 7.10: ICM_HTTP_CONNECTION_FAILED

1246275 - BWA/TREX: ICM connection test fails

I asked to our BC team to check this point.

Regards.

Laurent.

former_member217429
Active Contributor
0 Kudos

Hi Laurent,

The RFCServer (used for RFC connection) and IndexServer used for ICM are running on the different ports. What's the exact error code you have? You can look into the dev_icm log (f.e. from SMICM ) - it should definitely contain an error message now (f.e. host unknown , or connection broken ...).

Best regards,
Mikhail

laurent_burtaire
Active Contributor
0 Kudos

Hello Mikhail,

i did not exactly understand what was was the issue, but it was linked to the hostport called by ICM communication method

Now, good one is called.

After new full indexing, my Search Object Connector status being "Error while indexing" are now "Active".

Thanks again to all !

Regards.

Laurent.

Answers (2)

Answers (2)

former_member217429
Active Contributor
0 Kudos

Hi Laurent,

do you see any error messages in the ICM log (f.e.  ICM connection timeout)?
You can also look into the system logs (SM21) or SLG1 , but I think the best place to check the root cause of the issue is the TrexIndexServer trace. Do you have any entries there recorded on 08.01 16:14?

Best regards,
Mikhail

laurent_burtaire
Active Contributor
0 Kudos

Hello Mikhail,

i have no error in SMICM transaction.

No error too in SM21 or SLG1.

Regards.

Laurent.

ivy_li
Active Contributor
0 Kudos

Hi Laurent,

I think this is a new UI issue.
It would be better to open discussion in area SAP SRM User Interface Add-On where you will find more experts in this area.

BR,
Ivy

laurent_burtaire
Active Contributor
0 Kudos

Hello Ivy,

the thread is already assigned to category "SRM User Interface Add-On" which makes part of Supplier Relationship Management (SRM) forum.

Regards.

Laurent.

yu_feng2
Explorer
0 Kudos

Hi Laurent

I checked the log screen and found the following problem :

  "Depopulating in dependent nodes unsuccessful for index......"
  "An Exception has occurred".

From my option, I think it looks more like a BC-EIM-ESH issue.
I think it better you also open a discussion regarding this area.

Moreover, from post issues, I also found the followings,
Could you please check if it could be any help ?
********************************************

Firstly Please check in the Index Landscape in the Trex admin

tool if the Searcher is showing red. Which means that there is a problem

with the slave.The other is currently there is restrictions with

running a master/slave scenario. Please read the following regarding

this. See section "master/slave"

1628169 - Landscape restriction for TREX Master/Slave

Also there has been issues with initial indexing failing due to

memory deltas being activated. To resolve this you can deactivate

the delta during initial indexing and see if you still have issues

To find out exactly how to do this please read this note through and

all the notes referenced in it.

1737694 TREX 710: very high memory consumption in ES scenario

Try the indexing afterwards and see if the problem still presists

********************************************

Best Regards

Feng, yu


laurent_burtaire
Active Contributor
0 Kudos

Hello Yu,

thank for your help.

I have check in TREXADMIN transaction for the Index Landscape: all searchers are OK

Regarding delta, i switched it off. After this, i tried a new full indexation, but i still have the issue.

I am looking for OSS notes for application area BC-EIM-ESH. I will let you informed.

Regards.

Laurent.