You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi!Our company uses ambry as a file storage service. Now I need to migrate the ambry service to a new server. However, both servers do not have public network IP and are no longer the same network segment. Therefore, data cannot be copied through fragmentation or clustering. Can we find the storage path of ambry data, export the data directory directly and upload it to the new server After that, start the ambry service and access the data normally?Thank you
The text was updated successfully, but these errors were encountered:
Hi, I suppose you are not able to migrate data through replication (although this is recommended for online migration). I don't quite understand the way you proposed but if you want to know the data directory, the mountPath defined in HardwareLayout.json should be the place to look at (in that dir, each partition has its own dir). Assuming you succeed in exporting data and transfer them to new server, you also need to update the hostname in your HardwareLayout.json and PartitionLayout.json to indicate the location change. After that, you might have to reboot all frontends and servers to honor such change(if you are using static cluster manager). Hope this helps.
Hi!Our company uses ambry as a file storage service. Now I need to migrate the ambry service to a new server. However, both servers do not have public network IP and are no longer the same network segment. Therefore, data cannot be copied through fragmentation or clustering. Can we find the storage path of ambry data, export the data directory directly and upload it to the new server After that, start the ambry service and access the data normally?Thank you
The text was updated successfully, but these errors were encountered: