Amazon Web Services
The following section is intended to help you troubleshoot problems while settings AWS Credentials and Permissions.
Topics
ImportError: No module named awscli.clidriver
-
Symptom
You try to configure AWS profile as the authentication method and you get the following error:
_10Traceback (most recent call last):_10File "/usr/bin/aws", line 19, in <module>_10import awscli.clidriver_10ImportError: No module named awscli.clidriver -
Cause
HANA Uses Python version 2.x, AWS CLI driver uses Python3.
-
Solution
To be able to configure credentials, we have to temporarily unset
Python 2
and enablePython 3
environment variables. To do so:_10$ unset PYTHONHOME_10$ unset PYTHONPATH_10$ unset PYTHONSTARTUP_10$ export PATH=/usr/bin/python3:$PATH_10$ aws configure --profile emory
TotalPartsExceeded: exceeded total allowed configured MaxUploadParts (10000)
-
Symptom
Backups fails when uploading to S3 and you see the following error on emory log file:
_10ErrorMsg="aws.UploadObject: Error while uploading objects to AWS. Additional Information: MultipartUpload: upload multipart failed_10upload id: _EHp7lQOJEXA4TJB9xhDXmzb06UCNftERuTegG7.DT2QM_SlCq0ffV0h......UXIz6lJaYTl5SYWn5UAtPycxiTQaU.jaaFyckX5.v4gV1T9ClBzw--_10caused by:* TotalPartsExceeded: exceeded total allowed configured MaxUploadParts (10000). Adjust PartSize to fit in this limit"_10Hint="Check that the server has the correct permissions"* -
Cause
AWS S3 SDK upload function can handle a file which size is at maximum MemoryBufferSize * 10.000 bytes. So your MemoryBufferSize configuration parameter will define which is your maximum file size that emory will be able to upload correctly. By default, depending on your database engine, this parameter will be:
Engine Size (Bytes) Maximum file size to upload (GBytes) HANA 83886080 781.25 Oracle 5242880 48.82 Sybase 26214400 244.14 ImportantMinimum value for MemoryBuffer size is 5242880 bytes.
-
Solution: Depending on the database:
- SAP HANA
- Oracle
- SAP ASE
You can configure your HANA database engine to open more channels that just one when backup/restoring the database. This will launch several emory processes in parallel that each one will handle less amount of data as the whole data will be split onto smaller parts.
You can get more information on the following SAP Community Blog:
https://blogs.sap.com/2017/05/04/hana-backup-and-recovery-multi-streaming-data-backups-with-third-party-backup-toolsUse another configuration profile that uses stripe values higher than actual ones
This problem will not happen as maximum data file size is 32 GBytes.