3717996 - Secondary site HANA DB cannot start after unregistering it from System Replication: "Could not load HA/DR Provider"

After unregistering a HANA System Replication site from its primary, it cannot start up as a standalone system.
Errors similar to below can be found in the HANA DB daemon and nameserver trace files.

daemon_hostname.3xx00.xxx.trc

[1704723]{-1}[-1/-1] 2026-02-12 09:49:34.874911 i Daemon           SignalsUNIX.cpp(00601) : signo 2=SIGINT from user. errno 0 code 0. Requested 'QUIT'. Sender pid xxx, real user 'xxxadm'=xxxx, executable 'hdbnameserver'
[1704723]{-1}[-1/-1] 2026-02-12 09:49:34.875217 i Daemon           Daemon.cpp(01068) : Comment file contains: nameserver: Could not load HA/DR Provider

nameserver_hostname.3xx01.xxx.trc

[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.232019 i ha_dr_provider   HADRProviderManager.cpp(00080) : loading HA/DR Provider 'SAPHanaSR' from <full_path_to_the_hadr_provider>
[1704837]{-1}[-1/-1] 2026-02-12 09:49:34.855848 e python           PythonUtil.cpp(00483) : loading of module 'SAPHanaSR' failed
[1704837]{-1}[-1/-1] 2026-02-12 09:49:34.855892 e ha_dr_provider   PythonProxyImpl.cpp(00142) : import of SAPHanaSR failed: No module named 'SAPHanaSR'
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.864054 e ha_dr_provider   HADRProviderManager.cpp(00093) : could not load HA/DR Provider 'SAPHanaSR' from <full_path_to_the_hadr_provider>
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.864313 f NameServer       TREXNameServer.cpp(03491) : exception  1: no.7010003  (TREXNameServer/TREXNameServer.cpp:1578) TID: 1704834
    service is aborting: Could not load HA/DR Provider
exception throw location:
...
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.869495 f NameServer       TREXNameServer.cpp(03512) : Could not load HA/DR Provider -> stopping instance ...
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.869572 i DaemonClient     Client.cpp(00090) : Sending signal 2
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.874780 i DaemonClient     Client.cpp(00078) : Sending signal 2 to 1704723
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.874819 i DaemonClient     Client.cpp(00080) : Sent signal 2 to 1704723 rc 0
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.874820 i DaemonClient     Client.cpp(00135) : Send signal 2
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.874832 f NameServer       TREXNameServer.cpp(03525) :  stopping topology job
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.874837 f NameServer       TREXNameServer.cpp(03527) :  got shutdown scope
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.874843 f NameServer       TREXNameServer.cpp(03531) :  stopped topology job
[1704834]{-1}[-1/-1] 2026-02-12 09:49:34.874849 e Basis            TREXNameServer.cpp(03536) : Process exited due to an error via explicit exit call with exit code 1 , no crash dump will be written

Removing HA/DR provider configuration sections from the HANA DB configuration files manually fixes the issue and the database starts up. Example:

[ha_dr_provider_SAPHanaSR]
path = /usr/share/SAPHanaSR
execution_order = 1
provider = SAPHanaSR

Note, that "SAPHanaSR" is the provider name in the examples above but it can have any other name.

SAP HANA, platform edition 2.0

ha_dr_provider SAPHanaSR , KBA , HAN-DB-HA , SAP HANA High Availability (System Replication, DR, etc.) , Problem