We have some very large LVM2 filesystems, and have only seen one issue.
As you add PV's to a VG, the time it takes for the utilities (pvscan, pvs, etc.) to run increases exponentially with the number of volumes. This is because LVM2 puts metadata on every volume by default, and the utilities seem to process the metadata recursively. It was recommended that the "--metadatacopies=0" parameter be used on pvcreate for all but the first couple of PV's in a VG to avoid this. We also found that when you have more than a handful of PV's to work with, striping makes a HUGE difference in performance, so consider that. -----Original Message----- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Scott Rohling Sent: Monday, September 29, 2008 12:50 PM To: [email protected] Subject: [LINUX-390] Unusual amount of overhead on large volume group I'm confused about how much overhead seems to be involved in creating a volume group that approaches a terabyte with ECKD devices (A mix of some 3390-27 and mostly 3390-9): dxxxml01: ~ > df -h Filesystem Size Used Avail Use% Mounted on /dev/mapper/sysvg-root 2.0G 460M 1.5G 24% / /dev/dasda1 109M 34M 71M 33% /boot tmpfs 1005M 0 1005M 0% /dev/shm /dev/mapper/sysvg-var 4.0G 210M 3.6G 6% /var /dev/mapper/sysvg-usr 2.0G 864M 1.1G 46% /usr /dev/mapper/sysvg-home 3.0G 319M 2.5G 12% /home /dev/mapper/sysvg-opt 6.9G 359M 6.2G 6% /opt /dev/mapper/sysvg-tmp 2.0G 68M 1.9G 4% /tmp /dev/mapper/appvg-lotus 9.9G 151M 9.2G 2% /opt/ibm/lotus /dev/mapper/appvg-logdir 20G 173M 19G 1% /opt/ibm/lotus/logdir */dev/mapper/appvg-notesdata 897G 200M 870G 1% /opt/notesdata * dxxxml01: ~ > sudo lvdisplay -v /dev/appvg/notesdata Using logical volume(s) on command line --- Logical volume --- LV Name /dev/appvg/notesdata VG Name appvg LV UUID ub9Gqh-kAEc-wfxb-lXXi-i9ER-3F9Y-gjyHC2 LV Write Access read/write LV Status available # open 1 * LV Size 911.02 GB * Current LE 233222 Segments 71 Allocation inherit Read ahead sectors auto - currently set to 256 Block device 253:8 We're going from 911G for the logical volume to 897G displayed in 'df' to only 870G being available in the filesystem.. That's 41G of 'overhead'. Am I just naive about how much space it takes to manage this? Any input welcome! Scott Rohling ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 -------------------------------------------------------- This message w/attachments (message) may be privileged, confidential or proprietary, and if you are not an intended recipient, please notify the sender, do not use or share it and delete it. Unless specifically indicated, this message is not an offer to sell or a solicitation of any investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Merrill Lynch. Subject to applicable law, Merrill Lynch may monitor, review and retain e-communications (EC) traveling through its networks/systems. The laws of the country of each sender/recipient may impact the handling of EC, and EC may be archived, supervised and produced in countries other than the country in which you are located. This message cannot be guaranteed to be secure or error-free. This message is subject to terms available at the following link: http://www.ml.com/e-communications_terms/. By messaging with Merrill Lynch you consent to the foregoing. -------------------------------------------------------- ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390
