There seems to be some additional mitigating factors for this to work in 2-seconds vs 2-hours.
I just redid our SAN storage on another box that has 6.2.3.0 installed and it is formatting as before....gonna take hours for 300GB The big differences between this system and the other one: FAST = RH 6.1 filesystems=ext4 SLOW = RH 5.6 filesystem=ext3 Thinking it might be ext3 vs ext4, we tried to perform a mkfs.ext4 but that function/format doesn't exist until a higher kernel level that what is on our SLOW system. So, what are the details / requirements for FAST formatting vs SLOW? Is there a doc/tid to read? Zoltan Forray TSM Software & Hardware Administrator Virginia Commonwealth University UCC/Office of Technology Services zfor...@vcu.edu - 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: Hans Christian Riksheim <bull...@gmail.com> To: ADSM-L@VM.MARIST.EDU Date: 11/02/2011 09:49 AM Subject: Re: [ADSM-L] When hours become seconds Sent by: "ADSM: Dist Stor Manager" <ADSM-L@VM.MARIST.EDU> Yes, this was finally fixed for the UNIX platforms in 6.2.3. This update was released just in time for me when I had to format 40 TB of disk/file on a new server. Would have taken a week otherwise. I guess it does the same as on the windows platform so no need to write zeros on all blocks on unix anymore. Hans Chr. On Wed, Nov 2, 2011 at 1:49 PM, Zoltan Forray/AC/VCU <zfor...@vcu.edu> wrote: > I had an odd situation occur involving formatting disk storage volumes on > a Linux server. > > In the past, when creating and formatting storage pool volumes, a 300GB > volume would normally take an hour or more. I could watch the space being > allocated, piece-by-piece until it reached 300GB and then wait for the > formatting to complete. This process was usually so resource intensive > (never understood why???) doing something like an LS would take many > seconds to respond. > > Yesterday, I had a need to create new volumes on a recently rebuilt server > - RH Linux 5.6 and TSM server 6.2.3.0. > > Imagine my surprise when a 300GB volume was created in 2-seconds! > Furthermore, I create 9-300GB volumes as fast as I could enter the > commands! > > I thought for sure there was some kind of problem but all the error logs I > checked were clean. I figured if I started using them they would start > registering errors but nothing happened! Everything seems to be working > just fine. > > So, what gives? > > Is there some dramatic change in 6.2.3.0 (this is my first server upgraded > to this level) effecting the behavior of creating disk storage pool > volumes? It can't be the hardware since this is a 3+ year old machine. I > have newer, bigger, beefier, faster servers that still take a long time to > format storage volumes? > > > Zoltan Forray > TSM Software & Hardware Administrator > Virginia Commonwealth University > UCC/Office of Technology Services > zfor...@vcu.edu - 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 >