Gosh...
Well, if you are lucky, maybe in TSM the objects are not yet deleted...
If that's the case, then you can retrieve the objects from TSM with the following command:
Let say that you load ID is the following:
arsadmin retrieve -h archive -u admin -p password -n <pri_nid>-0 -g applGroup index_Doc_NAME
Let say that you load-ID is 112233-2-0-1234FAA-218921-218921
Then it should be something like:
arsadmin retrieve -h archive -u admin -p password -n 2-0 -g applGroup 1234FAA1
Once you got the 1234FAA1, then decompress it:
arsadmin decompress -s 1234FAA1 -o 1234FAA1.txt
Once decompressed, you can look inside the file 1234FAA1.txt, and you will see the entries of your segment table in your database. You can use this file, with some "work" to insert the rows in a segment table, or to recreate a new generic index.
Also in this file, you will find the column DOC_NAME, and you need to get all the different DOC_NAME available.
Let say that you find the following list (uniq doc_name):
- 1234FAAA
- 1234FAAB
- 1234FAAB$
- 1234FAAC
- 1234FAAC$
Then remove all the entries with the "$"
So you get the final list:
And then you can retrieve the objects in TSM with the following command:
arsadmin retrieve -h archive -u admin -p password -n 2-0 -g applGroup 1234FAAA 1234FAAB 1234FAAC
Now you should have the following files available:
- 1234FAA1
- 1234FAAA
- 1234FAAB
- 1234FAAC
Meaning that you can do many things with them. But at least you saved them, and you have time to rearchive them again in CMOD.
I will post a second entry on how to reinject in CMOD.
But at least you are in a situation, where you can decide in a more calm mood, what to do next...
IF the command "arsadmin retrieve" doesn't give you ANYTHING....
Well, in that case, you must hope that you have a backup of your cache for the objects of the load ID you deleted.
Or that you have somewhere a backup of TSM with the old data which contains your objects.
If you don't have them, and they are not retrievable anymore.... then you are f*cked up... *sorry for the bad word*.... because you might recover the database table for the indexes... but you have lost the documents behind forever...