Hi everybody, I have "Storage Management Server for AIX-RS/6000 - Version 5, Release 1, Level 8.0" Right after I upgraded to this version strange things started happening to the server. Before I had disk buffer 8Gb and all backups were good. All backup usually completed before 6 or 7 am. Now I have plenty backups completed around noon and even later. If I do "q se f=d" I see a loot of sessions waiting a mount point. Looks like it happening right when migration is triggered first time in the night. So I increased the disk pool up to 24Gb ( that is all I have now) and now it looks like follows: WRKDISK DISK 24,610.0 26.4 26.0 70 5 WRKPOOL
The strange thing here is when migration starts it never ends. The same volumes is requested by nodes and is used for both migration and backup, means backup goes through migration process. Look what gives me "q pr" command: 1,812 Migration Disk Storage Pool WRKDISK, Moved Files: 96162, Moved Bytes: 22,031,622,144, Unreadable Files: 0, Unreadable Bytes: 0. Current Physical File (bytes): 87,867,392 Current output volume: 787ABP. 1,813 Migration Disk Storage Pool WRKDISK, Moved Files: 104037, Moved Bytes: 23,340,695,552, Unreadable Files: 0, Unreadable Bytes: 0. Current Physical File (bytes): 12,288 Waiting for mount of output volume 798ABP (95 seconds). 1,814 Migration Disk Storage Pool WRKDISK, Moved Files: 91987, Moved Bytes: 22,099,623,936, Unreadable Files: 0, Unreadable Bytes: 0. Current Physical File (bytes): 2,351,595,520 Current output volume: 799ABP. 1,815 Migration Disk Storage Pool WRKDISK, Moved Files: 109809, Moved Bytes: 21,499,760,640, Unreadable Files: 0, Unreadable Bytes: 0. Current Physical File (bytes): 12,288 Waiting for mount of output volume 880ABP (35 seconds). --------------------------- Pay attention to the "Moved Bytes". It is 22G on every drive. Maximum backup size may not exceed 1g per PC and usually 200-300 Mb. But I have about 500 PCs. Am I the only one who has this problem? If you know how to fix it please share your experience. I want nodes do not reserve the tapes but keep backup going to diskpool and let migration move the data to the tapes. Yuriy Us.