The message is that NW cannot find any full for this save set. As a consequence it must run a full to begin the backup cycle.
Why? - no idea. Has this client ever successfully been backed up?
You better check the index with nsrck.
I would suggest you run a mminfo query for this machine for the saveset in question for the past 1 week and check the backups. Ensure that the retentions are set properly.
backup is getting successful. in the next run for incr backup it is not taking incr backup.
again FULL backup is running with below error.
138851:save: 'clirnt server:F:\XXXXXXXXXXX' is being reset/promoted to level 'Full' as no backup is found.
Now, as CrazyRov metioned it, I remember a case in our data zone which caused the same problems.
And yes, we had to merge the CFI databases.
So nsrclientfix should be the command to use for reporting and repair.
without other info a bit difficult to assess but one would guess that the backups might not ne administered under the name of the client. either by having merged it by accident due to using one of its names for a different client. can also occur with incorrect config in a clustered setup (which also can lead to the reference to another host). clientfix indeed can give you an idea about the host or if so required your whole datazone. does the client or its aliases possibly show up in another client? what is the reported clientname according to mminfo for a backup being made? doea that match the expected client. I also tend to check the name of the client local nsrla db (from backup server end via "nsradmin -s <client> -p nsrexecd" and the printing the contents of nsrla. "pr . type:nsrla" and looking for the name). we tend to add that name as an aliases automated but then also wrong setups like localhost could end up as an alias making things worse if the same occurs for yet another host... are there any options used in the backup command using another clientname?
Try changing the backup command in the client ressource to "save -c clientname". I've had this issue a couple of times as well. The 'fix' from Dell EMC was doing this.
My suggestion is to get the SSID from the latest successful Full backup from the backup logs.
With that in hand, query the Media DB using the SSID: mminfo -q ssid=ssid_number -r client,level,ssid
Is the client name what you expect?
IE: Does the client name in the Incr backup error match what you get in the mminfo output for the full backup?
138851:save: 'client server:F:\XXXXXXXXXXX' is being reset/promoted to level 'Full' as no backup is found.
If not then you may have client ID issues.