No index created for backup

What you do is not officially supported. I assume you are backing up filer so I really can't see why you can't do it? If people having filer won't give you an access to then they should also live with the fact they won't have backups as well. I hope you do realize that NFS backups are kind of double backups.

In your case I would check:
- are backups saved at all (use mminfo and check the saveset name as well)
- which index got updated (if any) when I saved this
- check the save manual for -L and -c options and see if that changes anything

But most of all - I would try to get the environment to be supported. Without it - you kiss goodbye to all of the official support (and there are always times when we need it).

Chang

My first question here would be are locally mounted file systems generating indexes or not - the easiest check for this is to going into nwrecover on the client and see if you can browse the file system. Just because a backup crosses a mount point does not mean it should not create an index.

Check that the browse time is not set to zero.

The "No Index Save" flag will do what it says - it will not back up the client indexes with the group; however it will generate the indexes in the first place.

I'm not too sure whether or not this is something that will be supported as I have had different lines from EMC in the past; my understanding is that it is supported but discouraged because:

1. it is a way of avoiding licensing fees
2. your data will be moved across the network twice

Official statement from Legato was that this is supported only for systems not having backup solution like module, clientpak, etc coming ftom Legato (eg. some BSD clients and few other falvors of not that spread OSes). NFS way of backing up filer partitions was fine until Legato introduced support for NDMP backups - then whole story changed.

Hi,

Thanks for so many feedback. Just realised that it is not due to NFS mounted filesystem. I tried to backup local directories and still no indexes are created. /etc is the local directory i tried to backup. These are the information:

[root@networker root]# mminfo -r client,name,savetime,nsavetime -vot -N /etc -c clueless.datastorage.bii
client name date save time
clueless.datastorage.bii /etc 03/21/06 1142930822

[root@networker root]# nsrinfo -t 1142930822 clueless.datastorage.bii
scanning client `clueless.datastorage.bii' for savetime 1142930822(Tue Mar 21 16:47:02 2006) from the backup namespace
0 objects found

Anyone of you encountered this problem before?

Talking about supported environment, is Solaris 10 supported in networker 7.2.1? We do have another Solaris 10 machine and it is backing up fine.

Was any data backed up at all? Try this one to see if we don't have an empty saveset:

mminfo -r client,name,savetime,nsavetime,level,totalsize -vot -N /etc -c clueless.datastorage.bii

If it was backed up at incremental level with no files having changed you would still be able to browse the file system in nwrecover based upon previous backups - can you see anything there?

If we ascertain that there are files that have been backed up but there are no indexes, next thing we could try doing is scanning indexes in and seeing if they are visible....

Another thing to do - take a look in the /nsr/index directory and see if there are more than one folder for this particular client

Did we confirm:

* browse time is not zero
* whether anything at all is backed up (by attempting to browse indexes using nwrecover)

Hi,

[root@networker root]# mminfo -r client,name,savetime,nsavetime,level,totalsize -vot -N /etc -c clueless.datastorage.bii
client name date save time lvl total
clueless.datastorage.bii /etc 03/21/06 1142930822 full 48433356


Seems that data have been saved. In /nsr/index , there are two directories , clueless.datastorage.bii and clueless.dev.bii . Both point to the same machine; clueless.dev.bii is an alias. I am clueless (no pun intended) why clueless.dev.bii was created in the first place. Both directories do not contain the indexes.

I tried running nsrck -L 7 clueless.datastorage.bii but still doesn't work.

Thanks in advance.

One more thing, I tried to recover the files through the Legato user program in windows. It gave an error "No index was found. You have not backed up any files from the source machine, or you need to connect to the server as the supervisor."

Hi,

It works! I put save -c clueless.datastorage.bii under the backup command and the indexes are created! Thank you all for helping me out with this. Definitely faster than logging cases to Legato support. Will be posting few more issues in the near future.