Skip to Content

TREX recovery strategy

May 25, 2017 at 12:34 PM


avatar image

Hi colleagues,

I would like to get an advice from you on TREX recovery.

I've found enough information regarding the backup tools for TREX (python scripts, OS tools and 3rd-party tools). So, my question is what will happen when we recover TREX.

I'm interesting in the following situations:

1) We've recovered TREX and, say, TREX-related system (CRM, SolMan etc.). If we recover on the same point of time we shouldn't get any problem.

2) We've recovered TREX system only on one day before. What will happen at TREX-related system? Will we get any problem in data consistency? If assume that CRM stores some information about indexes or 'any IDs' at TREX side.

3) We've recovered CRM system. In this case, TREX stores information about the documents that don't exist in CRM yet. How to clean up this redundant information? Is it critical to have the redundant information?

In my opinion, in case 2 we can generate indexes for missed document; in case 3, we can again generate indexes for 'newly-created' documents.


10 |10000 characters needed characters left characters exceeded
* Please Login or Register to Answer, Follow or Comment.

2 Answers

Best Answer
avatar image
Former Member Jul 05, 2017 at 07:10 PM

Hi Artem.

Point number 1. If it is same hardware i thing it will not have an impact, but i am not sure, simple you can take the backup using the python script and restore the same to any other Trex instance.

Point number 2 - If your are recovering the full system OS & Trex instance one day before from the backup, check the RFC's connected to CRM & Solman, and consistence check program for CRM & Solman system missing index objects. how you are connecting the Trex instance to CRM and Solman?

Point number: 3 After recovered the CRM do the fresh installation of Trex instance and do the basic configurations or clear the Index objects in the existing Trex system.

In out environment we are using the SAP DBM 6 & 8 and Trex 7.10 SP65, every day using the python script we are taking the business objects backup and same time its get restore in the DR location different Trex instance. In case of any emergency system down switch over to DR within 15 min.



10 |10000 characters needed characters left characters exceeded
avatar image
Former Member
Sep 11, 2017 at 09:48 AM

HI Artem,

recovery post backup restore is dependent on the application as detailed in the note below:

2207563 - TREX 710: indexes recovery procedure after corruption



10 |10000 characters needed characters left characters exceeded