Hello Dojo,
When migrating a standalone DMA (version 10.4.0.0-14772) to a new Hardware, I am facing the following issue:
I'm following the procedure as described in https://docs.dataminer.services/user-guide/Getting_started/Installing_a_DMA/Installing_DM_using_the_DM_Installer.html
under "Restoring a backup onto the new installed DataMiner Agent":
Everything goes well (files creation etc.) until the restore procedure tries (and fails) to restore the Cassandra data.
The batch file (Executing 'C:\Skyline DataMiner\Backup\{5B8F34AF-xxx-xxx}.bat' with params '' (working dir: 'C:\Skyline DataMiner\Backup\')) doesn't come to an end, the restore procedure would stay in "Still working..." status forever.
Actually I have tried and checked a lot of things, but without success.
There is one thing I've realized which makes me wonder if this is really the right procedure to migrate data from one Hardware server to another: When checking the options of the tool "CassandraBackup.exe" (which is called from the batch file with -restore option in this case) it's printing a comment: "only backup and consequently restore on the same node is supported!". In my opinion, this contradicts to the suggestion to use this procedure for restoring a DMA on a different node.
However, can anybody help and give a hint what's missing or what I might be doing wrong?
Best Regards
Jörg
Hi Brent,
Thanks for your response and for the clarification of the warning.
Here is some information about the sequence of failure:
A DMA is freshly installed on the host, and license is applied already.
Cassandra 3.11.19 is installed. On the source host, it's Cassandra 3.7.
Since the last restore action, all directories and files from the source DMA are available.
DataMiner is started, and during startup it has created the SLDMADB keyspace and tables in the Cassandra data directory.
So, in this stage, I have a fully started DMA available, but without database.
When I now execute "CassandraBackup.exe", the output and logfile produces the lines as shown here:
The SLDMADB schema is gone, and also the root user I created initially on Cassandra is gone.
Any further support is appreciated.
Best Regards - Jörg
Hi Jörg,
The procedure you linked is the correct one to migrate hardware. The warning about the same node applies when the cassandra is part of a cluster, where you should not restore data across nodes. It is fine to replace a node (remove the old node and add a new node) and restore the data to that node. For cassandra this counts as "the same node" Which is what the backup/restore procedure does.
As to why it is stuck in "still working" a bit hard to say from the screenshot. There should be a logfile at "C:\skyline dataminer\logging\CassandraBackup.txt" which may give more information.