Title: | VMSTUNING |
Notice: | Welcome to VMSTUNING |
Moderator: | EVMS::HALLYB |
Created: | Sat Feb 15 1986 |
Last Modified: | Wed May 14 1997 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 1539 |
Total number of notes: | 7984 |
Hello, my customer has a following disk space question: on openvms 6.2-1h2 typing a "dir/siz=all/gran $1$dua8:[*...]*:*;*" command, he get this report: gran total 53 directories,16100 files,3337112/6511212 blocks from "show device/full $1$dua8:" he get following: cluster size=9 free blocks=1825668 total blocks=8378028 host name=hsd30 $1$dua8: is a mirrorset of 2xrz29b on hsd30 controller 2.5 fw. Consideration: customer has recently performed a backup/restore of disk; the cluster_size parameter is at minumun size. The Question : is possible to reduce the allocation space on disk ?
T.R | Title | User | Personal Name | Date | Lines |
---|---|---|---|---|---|
1535.1 | RAID Software for OpenVMS for partitioning | OSOV03::KAGEYAMA | Trust, but Verify | Thu Apr 03 1997 01:37 | 9 |
>$1$dua8: is a mirrorset of 2xrz29b on hsd30 controller 2.5 fw. HSD50 allows partitioning but HSD30 seems it doesn't. So replace them? ;-) How about RAID Software for OpenVMS for partitioning? Additional disks might be cheaper than this software, though... - Kazunori | |||||
1535.2 | news on disk space | ITPD01::TERRIN | Thu Apr 03 1997 10:47 | 6 | |
hello, customer call me today and he has solved the problem. He has detected that a program allocates disk space, but it doesn't release the blocks unused. Thank's Maurizio | |||||
1535.3 | dir/sel=siz=min=#/size=all | CSC32::BUCKLEY | ski fast,take chances,die young | Thu Apr 03 1997 13:51 | 8 |
> hello, customer call me today and he has solved the problem. > He has detected that a program allocates disk space, but it doesn't > release the blocks unused. $dir/sel=size=min=1000/size=all disk:[000000...] would probibly have helped here Dan Buckley, digital csc | |||||
1535.4 | EPS::VANDENHEUVEL | Hein | Thu Apr 03 1997 22:50 | 21 | |
Just to beat that dead horse a little more... >gran total 53 directories,16100 files,3337112/6511212 blocks >cluster size=9 free blocks=1825668 total blocks=8378028 host name=hsd30 The absolute maximum loss due to the largish cluster size is 16,100 * 8 = 128,800 blocks. This is a far cry from those missing 6,511,212 - 3,337,112 = 3,174,100 blocks. Therefor, cluster size is NOT to blame and partitioning would NOT have resolved this particular problem. yes, Dan's command DIR/SIZE=ALL/SELE=SIZE=MINI=10000 would probably have found the offending files IFF.. the customer had the patience to wait for the result. A rather speedy, and highly recommended alternative to this is the VMS freeware distributed tool DFU. DFU SEARC/SIZE=MIN=1000 Cheers, Hein. |