cancel
Showing results for 
Search instead for 
Did you mean: 

RSTIRIDX job is not indexing any document in Solution Manager

Former Member
0 Kudos

Hello All,

I'm configuring TREX with Solution Manager for Full Text Search Configuration for SOLAR01. But not successful.

I've checked RFC with Solution Manager & TREX is working.

Did index activation for IBWSOLAR in SKPR06 ( DRFUZZY & VERTIFY)

Created Search Server Relation in SRMO & checked RFC is working.

Did , complete re-index through SKPR07 for IWBSOLAR document area.

Executed report RSTIRIDX_REINDEX for IWBSOLAR document area.

Finally created batch job for RSTIRIDX

But when checking the spool it's not indexing any document.

Looking for help.

Regards,

Sumanta

Accepted Solutions (1)

Accepted Solutions (1)

Thomas_S
Advisor
Advisor
0 Kudos

Hello Sumata,

make sure that you have the latest version of note:

1750162 - PHIOs are not found by Full Text Search Engine(TREX)

implemented in your system.

And then, after you have maintained the table SDOKMIME_C with the correct mime types as mentioned note 1237688, run the  report: SOLMAN_UPDATE_PHIO_MIMETYPE

from note:  1750162

for the Document extensions: DOCX, XLSX, PPTX

(you need to run the report for every Extension one after the other)

Best regards,

Thomas

Answers (1)

Answers (1)

Former Member
0 Kudos

Hello All,

Can anyone please comment on my above posted issue ?

Regards,

Sumanta

former_member217429
Active Contributor
0 Kudos

Hi Sumanta,

could you please apply the notes 650521 and 685521 .

Best regards,
Mikhail

Former Member
0 Kudos

Hello Mikhail,

Already above 2 SAP Note has been implemented.

Still the job is not indexing documents.

Regards,

Sumanta

former_member217429
Active Contributor
0 Kudos

Hello Sumanta,

may I ask you to open the OSS incident for this issue (component is BC-TRX).

Thanks and best regards,
Mikhail

francois_keen
Participant
0 Kudos

Hello Sumanta

What version of solution manager are you using?

After I upgraded from 7.1 SP10 to SP12, indexing stopped working for me.

I have opened an OSS incident end of last week which according to the message processor reached the SAP development team.

I will keep you posted,

@Mikhail the incident number for me is 1140570

cheers

Francois

francois_keen
Participant
0 Kudos

Hello,

The issue I was facing is now solved.

With some help from SAP AGS it was discovered that it was office 2007 documents which stopped being indexed after I upgraded from solman 7.1 SP10 to SP12.

the solution consisted of multiple steps which I can share provided you're facing the same symptom.

best regards

-

Former Member
0 Kudos

Hi Keen,

System is SP12 & facing issue after the upgrade.

Please send me the steps > sumanta.here@gmail.com

Regards,

Sumanta

francois_keen
Participant
0 Kudos

Hello Sumanta 

I wont send you the steps to your personal email, I dont think this is fair for the other SCN members.

I will share what I had to do to solve our problem.

1] implement note 1953873

2] add the office 2007 MIME type entries in tables SDOKMIME and SDOKFEXT as per note 1237688

3] ensure there was the  technical  user CONTENTSERV assigned to logon data tab for the ICM node /default_host/sap/bc/contentserver ; make sure the external alias is reflecting that too

4] I added the entry the entry "unescape_url=no" in the TREXPreprocessor.ini in the section [httpclient] as per note 650521

All of the above should help get new documents being indexed.

In regards to documents created before the upgrade to solman SP12, which cant be found using the full text search you have to do the following:

5] fix wrong mime entries

5.a] identify how many office 2007 documents were showing as MIME type application/octet-stream. in my solman production system, it was hundreds of documents.

to do so, run report RSIR_PHIO_MODIFY_CONTENT_INFO with PHIO class = SOLARGNSRC; Category of Temporary Storage = MIME ; original category = SOLAR01

5.b] in the displayed table, export results to spreadsheet, doing a right click, and fiter by MIME type=application/octet-stream

5c.] update each wrong entry using the Change button, example for a .docx, the MIME type will have to be updated to application/vnd.openxmlformats-officedocument.wordprocessingml.document

5.d] once the inconsitencies have been fixed, 2 options:

* if there are not so many documents, you can re-index documents manually by feeding the Document Field in SKPR07 with the PHIO ID displayed in the report SIR_PHIO_MODIFY_CONTENT_INFO results

* else, if many many documents to re-index, you have to drop your index and recreate it. 

Hope this helps

best regards

Francois

Former Member
0 Kudos

Hello Francois,

There are more than thousand documents were showing as MIME type application/octet-stream. in my system. I've deleted physical index from SRMO & recreated but still from SOLAR01 it's not showing.

Do you know any other process to delete index ?

Regards,

Sumanta

francois_keen
Participant
0 Kudos

Hi Sumanta

first, sorry to hear you got thousands of office 2007 documents showing as application/octet-stream, this will take you a long time to correct the entries to the right mime types using RSIR_PHIO_MODIFY_CONTENT_INFO.

2ndly, did you see your index being deleted at the TREXADMIN level?

this is how I deleted my index.

Call transaction SKPR07.

Select the required document class => SOLARGNSRC

Use Button "clean up table Entries" provided in SKPR07

Run report RSTIRIDX_INDXCAT_DEL to delete the active index

the index should be now be gone from TREXADMIN

Call transaction SKPR07 and click on marked documents, select the index category and click on delete.

Click on the Re-index button, this will mark all the documents for indexing, it took 20 mins on my system for several thousands docs, it will display “entries copied to search engine” on the status bar.

Click on trigger process and schedule it as a background job which will trigger the INDEXING job.

Hope this helps.

best regards

francois_keen
Participant
0 Kudos

Hi Sumanta

Any update on your progress, it has been 3 weeks I provided you with all the steps required to fix the problem...

Best regards

Francois

Former Member
0 Kudos

Hello Francois Keen,

Still few documents are not indexing yet. Result is not as expected.

I've opened a OSS message for this.

Regards,

Sumanta Chatterjee

francois_keen
Participant
0 Kudos

Thanks for your feedback Sumanta. so it means at least you made some progress, so I would however appreciate if you could please mark the replies which have been helpful!

thanks for your consideration.