This post is more than 5 years old
20 Posts
0
2254
DD-Boost from Networker
Hi!
I'm trying to add new DD-Boost Storage Unit on a DD4200, well. creating the SU is not a issue(I just add it from the GUI). But when trying to add it from networker(which already has one SU added to it (but with multiple folders/devices under it).
The reason why I want a new SU is that it would be easier to monitor diskusage from the GUI of DataDomain specially historic info).
I've already read the " EMC ® NetWorker ® and EMC Data Domain ® Boost Deduplication Devices Version 8.2Integration Guide" But I find that it lacks examples.
Can anyone help? is there something more to be done on the DD than just to create a SU, and then from networker start the new device wizard under the Datadomain in Networker?
A
coganb
736 Posts
0
November 6th, 2014 23:00
Hi,
If you do decide to go with creating a new SU, this example should help:
174255 : How to Create a NetWorker DDBoost Device in a non-default MTree
https://support.emc.com/kb/174255
-Bobby
anthome
20 Posts
0
November 6th, 2014 11:00
Well.. I initially also wanted the use of multiple SUs to use some with replication(to another DD) and some without. But perhaps you are right.
And I think you are right, that I would need to create the files under the SU. and I think that this is where it fails for me.
Another reason why we are investigating the use of multiple SUs is the multi-tenant option, so maybe we should enable that.. beacuse then it seems like it uses different SUs for different username. (I'm unsure about how enabling the multi-tenant feature affects the already configured SUs.)
I'm not that familiar with the mminfo-command but I guess I can run it with some different parameters to filter on specific devices or pools. I did notice the dd-cli command: "ddboost storage-unit show compression suname" and that gave me some info, but not historical.
ble1
14.3K Posts
1
November 6th, 2014 11:00
By default, NW will use SU which is named after hostname of the backup server. If you add manually another one, you must create devices manually with device access information to the new files under new SU. I'm not sure how do you create files under SU, but perhaps that happens by magic after labeling process - never tried that as when using wizard this is all automated.
I really do not support you approach as this is not the way to monitor disk usage. If you wish to monitor it, use mminfo which will give you data on how much data is written (raw), how much is it after global compression and how much is it at the end after local compression.
ble1
14.3K Posts
3
November 6th, 2014 12:00
I plan to write blog about it as many people still ignore this mminfo feature, but here is hint:
mminfo -av -q
This will give you raw total for whatever your query was. Field 4 and 5 are de-dupe and final size after compression. You can export those and create sort of overview on overall compression ratio per ssid which may help identify bad boys.
anthome
20 Posts
0
November 11th, 2014 23:00
Hrvoje Crvelin Thank you for that, but I suspect those commands are for Unix? This environment is running on windows and the awk-command is not present.. is it possible to run a similar script on windows?
R
ble1
14.3K Posts
0
November 13th, 2014 00:00
If PowerShell is your thing - yes
I only used awk to show certain data at specific point - I suspect you can do the same thing with PS.