I can feel some of your pain. I too have a Netware server that takes forever to backup (30+ hours) eventhough it only backups up a few GB each cycle but has to process 1.5M files.
I wonder about the size of your logs. You have it set to prune at 7-days and QUIET so why is it so big? Are you sure any changes to dsm.opt have been picked up/dsmcad restarted? If I understand correctly, if there is no active backup running, you should be able to just delete/rename the dsmsched.log file and the next backup pass will create a new one. Noticed you have RESOURCEUTILZATION at 8. Are you actually running multiple sessions? Also I would think MEMORYEFFICIENT would contradict RESOUREUTILZATION since specifying the latter uses more memory? Zoltan Forray TSM Software & Hardware Administrator Virginia Commonwealth University UCC/Office of Technology Services [email protected] - 804-828-4807 Don't be a phishing victim - VCU and other reputable organizations will never use email to request that you reply with your password, social security number or confidential personal information. For more details visit http://infosecurity.vcu.edu/phishing.html From: Timothy Hughes <[email protected]> To: [email protected] Date: 07/15/2010 01:20 PM Subject: [ADSM-L] Novell backup help Sent by: "ADSM: Dist Stor Manager" <[email protected]> Hello all, We have a novell client that's backing up or least trying to back up 376GB a night the dsm.opt is below (any comments welcom) Also, I can't open the dsmsched log (s) don't know why there is more than one. One of which is 314MB. I want to get rid it and start with a new one we could just unload and reload the dsmcad.nlm correct? Thanks, for any help in advance and feel free to offer any suggestions on how to speed up the backup which runs entirely to long to the point of not completing F.Y.I - we have to work this out before I can even begin to work on the cluster part * COMMMETHOD TCPip TCPSERVERADDRESS xx.xx.xx.xx TCPCLIENTADDRESS xx.xx.x.xxx TCPPORT 1500 HTTPPORT 1581 * Setting Nodename * ------------------------------------------------------- NODENAME PilotHQ_STARK DOMAIN RESTORE: * CLUSTERNODE YES * NetWare Memory management option * --------------------------------------------------------------- * The following option uses a memory conserving algorithm for * processing incremental backups as each directory is backed up. * This option may increase the duration of time required for * backups and increase CPU utilization on the TSM Server. * But will decrease the required NetWare memory needed for backups. * MEMORYEFFICIENTBACKUP YES RESOURCEUTILIZATION 8 NWEXITNLMP YES * NWUSER server_1_name\user:password * NWUSER server_2_name\user:password * NWUSER nds_tree_name\user:password * NetWare user ID and password file * -------------------------------------------------------------- * The following option specifies that the user ID and password used * to connect to a NetWare resource (server or NDS tree) will be stored * in an encrypted file in the installation directory after the intiial * prompt. An encrypted file will be created for each resource with the * extension .PWD * * NWPWFILE YES * Silencing screen output * ----------------------------------------------------------------- * Use the QUIET option to keep messages from being displayed on the * NetWare console and the schedule log file. This option will enhance * the performance during scheduled backups. * * Note that by using this option, schedule event windows, schedule * statistics and errors during schedule processing will still be * written to the schedule log file. * * If this option is not specified, the default VERBOSE will be taken. * MANAGEDSERVICES WEBCLIENT SCHEDULE SCHEDMODE PROMPTED tcpwindowsize 64 largecommbuffers yes tcpb 32 txnb 25600 tcpnodelay yes Errorlogretention 7 Schedlogretention 7 passwordaccess generate QUIET Getting the following messas in the dsmerror.log 07/15/2010 08:12:53 (TSAFS.NLM 6.53.3 314) All resource names are already scanned, or none were found. 07/15/2010 08:12:53 ANS1999E Incremental processing of 'RESTORE:' stopped. 07/15/2010 08:12:53 ANS1028S An internal program error occurred. 07/15/2010 08:12:53 ANS1512E Scheduled event 'MIDNIGHT' failed. Return code = 12.
