cancel
Showing results for 
Search instead for 
Did you mean: 

HANA Snapshot no end time

Former Member
0 Kudos
516

Hi all,

I try to execute a point-in-time restore using snapshots. During the recovery I want to roll forward the database to a specific time - meaning combination of storage snapshots and log backups.

The snapshot is in state "successful" in M_BACKUP_CATALOG.

After attaching the storage (EBS volumes) from the snapshot, the state changes to "failed". The backup catalog is located on a different volume as /hana/data and /hana/log. That means, the catalog is not reset by the snapshot.

Is it possible to load the backup catalog back to the system DB?

Any advice, why the snapshot state changes to "failed" and the end time (SYS_END_TIME and UTC_END_TIME) is empty?

HANA version: 2.00.020.00 with one single tenant.

Any help is appreciated.

Regards Arne

Accepted Solutions (0)

Answers (1)

Answers (1)

0 Kudos

Hi,

a combination of storage snapshot of HANA Log-Volumes and Point In Time-Recovery is not possible. See SAP-Note https://launchpad.support.sap.com/#/notes/2039883

Which volumes of the SAP HANA database are affected by a storage snapshot?

A storage snapshot has to consider the data volumes of all SAP HANA database services. The log area must not be included in a storage snapshot to be able to perform a database recovery including the point-in-time option. Thus, log backups and not yet backed up log segments can be recovered after a snapshot has been restored.

BR

Markus

Former Member
0 Kudos

Hi Markus,

the snapshot of /hana/log is not executed with the intention of a point-in-time recovery. In the AWS context log and data are sharing the same disks. I was trying to use the log backups for the recovery with the snapshot of /hana/data and CLEAR LOG command.

In the meantime I was able to solve the issue. The tenant DB must not start after attaching the snapshot and before executing the recover process. It is no issue to include the log area to the snapshot, if you use 'clear log' during the recovery and backup log files.

Regards

Arne

0 Kudos

Hi Arne,

We are facing similar issues while recovering SAP Hana DB post migration to a new host. We register the system with the new host using hdbreg, the snapshot gets abandoned as the system comes up. Below is the snapshot status :-

1577939744796,"data snapshot",1577939744796,"2020-01-02 04:35:44.796000000","2020-01-02 04:35:44.796000000","2020-01-02 04:35:45.077000000","2020-01-02 04:35:45.077000000","successful","snapshot_2020-01-02_10:05:44","abandoned because of service restart","QAS"

How do we bring up the system from the same snapshot, I cannot reapply the same snapshot as the system would go down while applying that and does not come up with the below command

hana/shared/<SID>/HDB00/exe/Python/bin/python /hana/shared/<SID>/HDB00/exe/python_support/recoverSys.py --command "RECOVER DATA USING SNAPSHOT CLEAR LOG"