Re: sacrifice H310 hardware r5 for lvm2 raid5+lvmcache
lejeczek <[email protected]>
| Newsgroups | gmane.linux.hardware.dell.poweredge |
|---|---|
| Message-ID | <[email protected]> |
I assumed that if firmware is capable of this then omsa should be too, but when I tried: # omconfig storage controller action=setcontrollermode controller=0 mode=hba Operation not supported. Read, action=setcontrollermode Refer to the documentation for more information. then I asked about that notion of vdisk=pdisk Either I miss something in OMSA or it's weird again how OMSA developers are inconsistent, megacli does the trick for you. thanks. On 27/07/15 14:46, Paul Raines wrote: > > There may be a way to do it with omconfig / storcli / > MegaCLI but > we just set the disk to "Non-RAID" in the boot up BIOS > tool of the > RAID card (Ctrl-R). > > Install the 'lsscsi' tool which is useful for seeing > things in these cases > It will show something like > > [0:2:0:0] disk DELL PERC H310 2.12 /dev/sda > > for a virtual disk but after converted to non-RAID (JBOD) > it should show > something like > > [0:2:0:0] disk ATA ST2000DM001-1CH1 CC24 /dev/sda > > depending on disk make/model. /dev/sda will also suddenly > be a few > cylinders bigger. > > You want to wipe any partition table the OS finds on the > disk as it will have number of cylinders. So something like > > dd if=/dev/zero of=/dev/sdX count=1 bs=4k > > should do it. Then run parted / gdisk / whatever > partition tool you like to recreate a new one. Ignore any > error about a backup GPT table > > -- Paul Raines (http://help.nmr.mgh.harvard.edu) > > > > On Mon, 27 Jul 2015 9:26am, lejeczek wrote: > >> but would OMSA do it and without re-flashing the >> firmware? Or LSI's tool set is needed? >> >> On 27/07/15 14:21, Paul Raines wrote: >>> >>> I cannot comment on the issue of using the hardware >>> RAID5 or not. However, >>> I will say from experience that 1-disk virtual disks >>> are bad idea. If you >>> are doing that, you should put the disks in JBOD mode >>> instead. Doing a >>> 1-disk vdisk just puts an unnecessarily layer on top of >>> disk I/O that >>> gives a added failure mode for no purpose. >>> >>> -- Paul Raines (http://help.nmr.mgh.harvard.edu) >>> >>> >>> >>> On Mon, 27 Jul 2015 3:48am, lejeczek wrote: >>> >>> > what do you think? >>> > > dump raid5? change each pdisk into vdisk exposing >>> to OS and >>> > instead run lvm raid5 on those vdisks with a cache >>> device? >>> > > thanks. >>> > > _______________________________________________ >>> > Linux-PowerEdge mailing list >>> > [email protected] >>> > >>> https://lists.us.dell.com/mailman/listinfo/linux-poweredge >>> > > > >>> >>> The information in this e-mail is intended only for the >>> person to whom it >>> is >>> addressed. If you believe this e-mail was sent to you >>> in error and the >>> e-mail >>> contains patient information, please contact the >>> Partners Compliance >>> HelpLine at >>> http://www.partners.org/complianceline . If the e-mail >>> was sent to you in >>> error >>> but does not contain patient information, please >>> contact the sender and >>> properly >>> dispose of the e-mail. >>> >>> >> >> >> >