01-25-2012 09:23 AM
After restarting netbackup services, VMD & LTID is not starting. VMD came up after starting manually. LTID is not starting. lots of errors on debug logs
10:26:37.761 [2572484] <4> ltid: INITIATING
10:26:37.763 [2572484] <2> mm_getnodename: cached_hostname memcf-master, cached_method 3
10:26:37.827 [2572484] <2> mm_ncbp_gethostname: GetNBUName <memcf-master>
10:26:37.827 [2572484] <2> mm_getnodename: (5) hostname memcf-master (from mm_ncbp_gethostname)
10:26:37.827 [2572484] <4> ltid: EMM tuning parameters: emm_retries = 1, emm_connect_timeout = 20, emm_request_timeout = 300
10:26:37.827 [2572484] <4> ltid: using Scan Host status interval 300
10:26:37.828 [2572484] <4> ltid: auto_correct_paths set to 0
10:26:37.853 [2572484] <16> tapelib: shmat() failed attaching shared memory, The process file table is full.
10:26:37.854 [2572484] <16> InitLtidCreateShmObjects: Error in IPC SHMAT call
10:26:37.854 [2572484] <16> TermLtid: Shutting down ltid, ToAnswer = no
10:26:39.854 [2572484] <4> TermLtid: chmod dev files
10:26:39.868 [2572484] <4> TermLtid: emmlib_uninitialize failed, status=328
01-25-2012 09:47 AM
Please tell us more about this environment:
01-25-2012 10:52 AM
Send us the details requested by Marianne.
On the other hand if you are not able to start the LTID check the tape drive status (stuck tapes ? , turned off ?) calling the commands robtest or vmoprcmd.
Once you are 100% sure that the hard is fine, while no backups are running run the command NBRBUTIL -resetall.
Try restarting the Netbackup services and check for the LTID to be UP and running.
01-25-2012 09:32 PM
Have you make any change like changing kernel parameters, installing new software?
By shmop(2) manual on Solaris 10,
EMFILE The number of shared memory segments attached to the calling process would exceed the system-imposed limit.
01-25-2012 11:37 PM
Is this not a nofiles issue ?
What is the output of ulimit -a ?
martin