Unsolved
This post is more than 5 years old
2 Posts
0
2783
multipe DataDomain Dedup information in mminfo
Hi all,
I got some ss information and I think I know what the most of it is, but I dont understand everything.
Im new to Networker and DataDomain.
I extracted one single result block from a mminfor report, it shows 2 blocks with dedupe information for the dd (the 2 blocks starting with "v1:" in the attributes) , I masekd some uninteresting parts and filled in some return for better readability:
3922136649
volumename1
clientname2
poolname3
0 KB
examplessname
08/09/15 10:24:41
08/09/15 10:24:49
08/09/15 10:24:34
manual
448542272
1
09/09/15 23:59:59
2
vF
09/09/15 23:59:59
1439108681
08/09/15 10:24:41
438 MB
438 MB
08/09/15 10:24:51
09/09/15 23:59:59
*ss clone retention: 1439108681: 1439108691: 2727308, 1439110810: 1439110163: 2725836;
*ss data domain backup cloneid:1439108681;
*ss data domain dedup statistics:
v1:1439108681:450037520:450037520:193016691,
v1:1439110810:450039776:1506123:1504649
For the numbers, please correct me if I am wrong with something.
is the #Bytes on disk (client)
The "*ss data domain dedup statistics" contains 2 blocks of information (starting with "v1.") , with 4 numbers each. The numbers of one block are :
timestamp,
size of the ss with meta data (thats why its bigger than total)
size of the ss after global compression (dedupe)
size of the ss after local compression (another compression alorithm like gz,gzfast,lz,...its used on the data after dedupe)
Why do I have 2 blocks with this 4 numbers ? And what does "v1" mean ?
Its only a guess, but the second row could either be for a clone or for a second DataDomain or a clone on a second DataDomain.
In case of 2 diffrent DataDomains I dont understand why the compression is so much better for the other DataDomain.
Another answer would be that a clone and a backup would go to the same DataDomain (thats why the dedupe compression is so good) but why should one do that? And wouldnt be there another a seperat ss for that case?
I hope someone can help me, best regards!
ble1
14.3K Posts
2
October 14th, 2015 00:00
There is KB article about, but... v1 ignore - this is not relevant for anything. Values mentioned are cloneid, backup size, pre-dedupe size and post-dedupe size. Normally you can get those with -r attrs or -S parameter when checking individual ssid record.
babochabo1
2 Posts
0
October 14th, 2015 03:00
Thank you for your quick answer Hrvoje Crevlin ! Can you remember where you found this article in the KB?
That was nearly the verification of what I thought.
But what interests me too is : Why do I have second block of all these numbers ? (the second "v1:")
Do you know anything about that too ?
ble1
14.3K Posts
0
October 14th, 2015 08:00
As hinted before, if you search for mminfo and attrs you will find article https://support.emc.com/kb/192743
I can't say much about your second block as I do not know how the query was made. It is part of answer to the query and one needs to know exact query to be able to interpret results. Only you know that as you never posted query part used to get information.