From: Oliver Rath <rath@mglug.de>
To: linux-lvm@redhat.com
Subject: Re: [linux-lvm] LVM2 [lvcreate --type raid5] issue
Date: Mon, 03 Feb 2014 11:17:24 +0100 [thread overview]
Message-ID: <52EF6CB4.7010505@mglug.de> (raw)
In-Reply-To: <BLU0-SMTP1435605BEFA0CCEC48A9CC4FBA80@phx.gbl>
[-- Attachment #1: Type: text/plain, Size: 14958 bytes --]
Hi James,
i was able to reproduce this on a Ubuntu 12.04.3 machine with lvm
2.02.105+ (compiled from git). It seems, that the -s 4kib parameter
causes this error. Normally, if not set explicitly, it es a 4MB size
assumed, so that you choosed a really tiny size (1/1000th of standard
size). Omitting this all seems running fine.
Im not a developer of this, but i could imagine, that based on this tiny
size the metadata is running out of space. Setting this to at least
64kib seems running fine. I think, smaller than 64kib doesnt make sense,
because this is the smallest reading size of most harddisks.
Hth
Oliver
On 02.02.2014 20:53, James Woolliscroft wrote:
>
> Hi
>
>
>
> New to Linux, not sure if this is a legitimate fault, or if I am doing
> something wrong. *_Please can you advise on which release of LVM2 (if
> any) fixes this issue._*
>
>
>
> Initially tested using SL6.4, now testing with SL6.5. I have used both
> LiveDVD and basic "Minimal Desktop" Installations on the target
> computer and inside a VirtualBox Virtual Machine.
>
>
>
> I am preparing to install SL6.5 on a small system with 3x750GB Disks.
> As a pre-requisite, I wish to testing the performance of the disks
> using different chunk/block sizes using LVM2 RAID5.
>
>
>
> It is a preference to use LVM2 RAID5 rather than mdraid and LVM2 to
> reduce the complexity of setting up and managing the system.
>
>
>
> However I am experiencing a problem when creating the Logical Volume.
> This issue occurs with both SL6.4 and SL6.5:
>
>
>
> * /dev/vg_system1/lv_system1: read failed after 0 of 4096 at 0:
> Input/output error*
>
> * /dev/vg_system1/lv_system1: write failed after 0 of 4096 at 0:
> Input/output error*
>
>
>
> There was one additional issue with SL6.4 which also appears to be
> partially resolved in SL6.5. Certain commands function without an
> issue, however once the issue manifests, it appears in most or all
> commands:
>
>
>
> [root@localhost ~]# pvscan -vd
>
> Wiping cache of LVM-capable devices
>
> Wiping internal VG cache
>
> Walking through all physical volumes
>
> * /dev/vg_system1/lv_system1: read failed after 0 of 4096 at
> 5368643584: Input/output error*
>
> * /dev/vg_system1/lv_system1: read failed after 0 of 4096 at
> 5368700928: Input/output error*
>
> * /dev/vg_system1/lv_system1: read failed after 0 of 4096 at 0:
> Input/output error*
>
> * /dev/vg_system1/lv_system1: read failed after 0 of 4096 at 4096:
> Input/output error*
>
> * /dev/vg_system1/lv_system1: read failed after 0 of 4096 at 0:
> Input/output error*
>
> PV /dev/sdb1 VG vg_system1 lvm2 [16.00 GiB / 13.50 GiB free]
>
> PV /dev/sdc1 VG vg_system1 lvm2 [16.00 GiB / 13.50 GiB free]
>
> PV /dev/sdd1 VG vg_system1 lvm2 [16.00 GiB / 13.50 GiB free]
>
> PV /dev/sda2 VG VolGroup lvm2 [11.51 GiB / 0 free]
>
> Total: 4 [59.50 GiB] / in use: 4 [59.50 GiB] / in no VG: 0 [0 ]
>
>
>
> I am at a loss to explain this, I cannot find any information relating
> to these errors.
>
>
>
> If I cannot resolve this within a reasonable period of time, I will be
> forced to use mdraid which is much less flexible -- i.e. if I wish to
> use different RAID modes, I have to use separate partitions/volume
> groups for example I would consider using unraided partitions for
> /boot, RAID 0 for SWAP and /tmp and RAID 5 for the rest of the system.
> Separating these into different partitions means the system will have
> to span further across the disk at times.
>
>
>
> I am considering possibly testing a later revision of LVM2 as I
> understand that SL6.5 has *_LVM2-2.02.100-8_* while *_subversions 101
> -- 105_* are available. I understand that to do this I may have to
> download the source code of this and dependant packages and compile
> it. This is not something I can easily do, but am prepared to try if I
> can get some guidance.
>
>
>
> It may be useful to test this in RHEL 6.5 and CentOS 6.5 though I
> could only do this on the latter myself.
>
>
>
> Below are the scripts to destroy the test LV/VG/PV and a test pass,
> and a script to create the PV/VG/LV as executed on a test system. Each
> attempt I execute the destroy script prior to running the create
> script. There is also a dump of the partition tables.
>
>
>
> I hope you can point me in the right direction!
>
>
>
> Thanks and Regards,
>
>
>
> J Woolliscroft
>
> [root@localhost ~]# cat d_lv
>
> lvremove /dev/vg_system1/lv_system1
>
> vgremove vg_system1
>
> pvremove /dev/sdb1 /dev/sdc1 /dev/sdd1
>
> =====
>
> [root@localhost ~]# . ./d_lv
>
> Do you really want to remove active logical volume lv_system1? [y/n]: y
>
> Logical volume "lv_system1" successfully removed
>
> Volume group "vg_system1" successfully removed
>
> Labels on physical volume "/dev/sdb1" successfully wiped
>
> Labels on physical volume "/dev/sdc1" successfully wiped
>
> Labels on physical volume "/dev/sdd1" successfully wipeda
>
>
>
> [root@localhost ~]# cat c_lv
>
> parted -l
>
>
>
> pvcreate -M2 -v /dev/sdb1 /dev/sdc1 /dev/sdd1
>
>
>
> pvs -v -d
>
> pvscan -v -d
>
>
>
> vgcreate -s 4kib -v vg_system1 /dev/sdb1 /dev/sdc1 /dev/sdd1
>
>
>
> vgs -v -d
>
> vgscan -v -d
>
>
>
> lvcreate --type raid5 -i 2 -I 4 -L 5G -ay -Zy -v -n lv_system1 vg_system1
>
>
>
> lvs -v -d
>
> lvscan -v -d
>
> [root@localhost ~]# . ./c_lv
>
> Model: ATA VBOX HARDDISK (scsi)
>
> Disk /dev/sda: 12.9GB
>
> Sector size (logical/physical): 512B/512B
>
> Partition Table: msdos
>
>
>
> Number Start End Size Type File system Flags
>
> 1 1049kB 525MB 524MB primary ext4 boot
>
> 2 525MB 12.9GB 12.4GB primary lvm
>
>
>
>
>
> Model: ATA VBOX HARDDISK (scsi)
>
> Disk /dev/sdb: 17.2GB
>
> Sector size (logical/physical): 512B/512B
>
> Partition Table: gpt
>
>
>
> Number Start End Size File system Name Flags
>
> 1 1049kB 17.2GB 17.2GB primary lvm
>
>
>
>
>
> Model: ATA VBOX HARDDISK (scsi)
>
> Disk /dev/sdc: 17.2GB
>
> Sector size (logical/physical): 512B/512B
>
> Partition Table: gpt
>
>
>
> Number Start End Size File system Name Flags
>
> 1 1049kB 17.2GB 17.2GB primary lvm
>
>
>
>
>
> Model: ATA VBOX HARDDISK (scsi)
>
> Disk /dev/sdd: 17.2GB
>
> Sector size (logical/physical): 512B/512B
>
> Partition Table: gpt
>
>
>
> Number Start End Size File system Name Flags
>
> 1 1049kB 17.2GB 17.2GB primary lvm
>
>
>
>
>
> Model: Linux device-mapper (linear) (dm)
>
> Disk /dev/mapper/VolGroup-lv_swap: 2147MB
>
> Sector size (logical/physical): 512B/512B
>
> Partition Table: loop
>
>
>
> Number Start End Size File system Flags
>
> 1 0.00B 2147MB 2147MB linux-swap(v1)
>
>
>
>
>
> Model: Linux device-mapper (linear) (dm)
>
> Disk /dev/mapper/VolGroup-lv_root: 10.2GB
>
> Sector size (logical/physical): 512B/512B
>
> Partition Table: loop
>
>
>
> Number Start End Size File system Flags
>
> 1 0.00B 10.2GB 10.2GB ext4
>
>
>
>
>
> Set up physical volume for "/dev/sdb1" with 33552351 available sectors
>
> Zeroing start of device /dev/sdb1
>
> Writing physical volume data to disk "/dev/sdb1"
>
> Physical volume "/dev/sdb1" successfully created
>
> Set up physical volume for "/dev/sdc1" with 33552351 available sectors
>
> Zeroing start of device /dev/sdc1
>
> Writing physical volume data to disk "/dev/sdc1"
>
> Physical volume "/dev/sdc1" successfully created
>
> Set up physical volume for "/dev/sdd1" with 33552351 available sectors
>
> Zeroing start of device /dev/sdd1
>
> Writing physical volume data to disk "/dev/sdd1"
>
> Physical volume "/dev/sdd1" successfully created
>
> Scanning for physical volume names
>
> Wiping cache of LVM-capable devices
>
> PV VG Fmt Attr PSize PFree DevSize PV
> UUID
>
> /dev/sda2 VolGroup lvm2 a-- 11.51g 0 11.51g
> JjV0kz-rW98-zTxe-X1jm-T9h3-i6SG-Frd325
>
> /dev/sdb1 lvm2 a-- 16.00g 16.00g 16.00g
> k7kxWU-W1bh-Eqda-xVj8-aBva-efN4-Et5B94
>
> /dev/sdc1 lvm2 a-- 16.00g 16.00g 16.00g
> 9M29V5-qldE-mSVl-Z2UY-h9nM-1F3i-AEsNYu
>
> /dev/sdd1 lvm2 a-- 16.00g 16.00g 16.00g
> zPpXzZ-ptwp-qxok-ubNq-asmz-Ifcj-QtRvfv
>
> Wiping cache of LVM-capable devices
>
> Wiping internal VG cache
>
> Walking through all physical volumes
>
> PV /dev/sda2 VG VolGroup lvm2 [11.51 GiB / 0 free]
>
> PV /dev/sdb1 lvm2 [16.00 GiB]
>
> PV /dev/sdc1 lvm2 [16.00 GiB]
>
> PV /dev/sdd1 lvm2 [16.00 GiB]
>
> Total: 4 [59.50 GiB] / in use: 1 [11.51 GiB] / in no VG: 3 [48.00 GiB]
>
> Wiping cache of LVM-capable devices
>
> Wiping cache of LVM-capable devices
>
> Adding physical volume '/dev/sdb1' to volume group 'vg_system1'
>
> Adding physical volume '/dev/sdc1' to volume group 'vg_system1'
>
> Adding physical volume '/dev/sdd1' to volume group 'vg_system1'
>
> Archiving volume group "vg_system1" metadata (seqno 0).
>
> Creating volume group backup "/etc/lvm/backup/vg_system1" (seqno 1).
>
> Volume group "vg_system1" successfully created
>
> Finding all volume groups
>
> Finding volume group "vg_system1"
>
> Finding volume group "VolGroup"
>
> VG Attr Ext #PV #LV #SN VSize VFree VG
> UUID VProfile
>
> VolGroup wz--n- 4.00m 1 2 0 11.51g 0
> oaJmPd-ElY9-a0E9-hUlu-SLxE-XTje-LfpO31
>
> vg_system1 wz--n- 4.00k 3 0 0 47.99g 47.99g
> hZTC0w-k0bJ-6M2A-Fmky-hHLo-hzHM-k8gdPl
>
> Wiping cache of LVM-capable devices
>
> Wiping internal VG cache
>
> Reading all physical volumes. This may take a while...
>
> Finding all volume groups
>
> Finding volume group "vg_system1"
>
> Found volume group "vg_system1" using metadata type lvm2
>
> Finding volume group "VolGroup"
>
> Found volume group "VolGroup" using metadata type lvm2
>
> Setting logging type to disk
>
> Finding volume group "vg_system1"
>
> Archiving volume group "vg_system1" metadata (seqno 1).
>
> Creating logical volume lv_system1
>
> Creating logical volume lv_system1_rimage_0
>
> Creating logical volume lv_system1_rmeta_0
>
> Creating logical volume lv_system1_rimage_1
>
> Creating logical volume lv_system1_rmeta_1
>
> Creating logical volume lv_system1_rimage_2
>
> Creating logical volume lv_system1_rmeta_2
>
> activation/volume_list configuration setting not defined: Checking
> only host tags for vg_system1/lv_system1_rmeta_0
>
> Creating vg_system1-lv_system1_rmeta_0
>
> Loading vg_system1-lv_system1_rmeta_0 table (253:2)
>
> Resuming vg_system1-lv_system1_rmeta_0 (253:2)
>
> Clearing metadata area of vg_system1/lv_system1_rmeta_0
>
> Clearing start of logical volume "lv_system1_rmeta_0"
>
> Removing vg_system1-lv_system1_rmeta_0 (253:2)
>
> activation/volume_list configuration setting not defined: Checking
> only host tags for vg_system1/lv_system1_rmeta_1
>
> Creating vg_system1-lv_system1_rmeta_1
>
> Loading vg_system1-lv_system1_rmeta_1 table (253:2)
>
> Resuming vg_system1-lv_system1_rmeta_1 (253:2)
>
> Clearing metadata area of vg_system1/lv_system1_rmeta_1
>
> Clearing start of logical volume "lv_system1_rmeta_1"
>
> Removing vg_system1-lv_system1_rmeta_1 (253:2)
>
> activation/volume_list configuration setting not defined: Checking
> only host tags for vg_system1/lv_system1_rmeta_2
>
> Creating vg_system1-lv_system1_rmeta_2
>
> Loading vg_system1-lv_system1_rmeta_2 table (253:2)
>
> Resuming vg_system1-lv_system1_rmeta_2 (253:2)
>
> Clearing metadata area of vg_system1/lv_system1_rmeta_2
>
> Clearing start of logical volume "lv_system1_rmeta_2"
>
> Removing vg_system1-lv_system1_rmeta_2 (253:2)
>
> Creating volume group backup "/etc/lvm/backup/vg_system1" (seqno 3).
>
> activation/volume_list configuration setting not defined: Checking
> only host tags for vg_system1/lv_system1
>
> Creating vg_system1-lv_system1_rmeta_0
>
> Loading vg_system1-lv_system1_rmeta_0 table (253:2)
>
> Resuming vg_system1-lv_system1_rmeta_0 (253:2)
>
> Creating vg_system1-lv_system1_rimage_0
>
> Loading vg_system1-lv_system1_rimage_0 table (253:3)
>
> Resuming vg_system1-lv_system1_rimage_0 (253:3)
>
> Creating vg_system1-lv_system1_rmeta_1
>
> Loading vg_system1-lv_system1_rmeta_1 table (253:4)
>
> Resuming vg_system1-lv_system1_rmeta_1 (253:4)
>
> Creating vg_system1-lv_system1_rimage_1
>
> Loading vg_system1-lv_system1_rimage_1 table (253:5)
>
> Resuming vg_system1-lv_system1_rimage_1 (253:5)
>
> Creating vg_system1-lv_system1_rmeta_2
>
> Loading vg_system1-lv_system1_rmeta_2 table (253:6)
>
> Resuming vg_system1-lv_system1_rmeta_2 (253:6)
>
> Creating vg_system1-lv_system1_rimage_2
>
> Loading vg_system1-lv_system1_rimage_2 table (253:7)
>
> Resuming vg_system1-lv_system1_rimage_2 (253:7)
>
> Creating vg_system1-lv_system1
>
> Loading vg_system1-lv_system1 table (253:8)
>
> Resuming vg_system1-lv_system1 (253:8)
>
> Monitoring vg_system1/lv_system1
>
> Clearing start of logical volume "lv_system1"
>
> * /dev/vg_system1/lv_system1: read failed after 0 of 4096 at 0:
> Input/output error*
>
> * /dev/vg_system1/lv_system1: write failed after 0 of 4096 at 0:
> Input/output error*
>
> Creating volume group backup "/etc/lvm/backup/vg_system1" (seqno 3).
>
> Logical volume "lv_system1" created
>
> Finding all logical volumes
>
> LV VG #Seg Attr LSize Maj Min KMaj KMin Pool
> Origin Data% Meta% Move Cpy%Sync Log Convert LV
> UUID LProfile
>
> lv_root VolGroup 1 -wi-ao---- 9.51g -1 -1 253
> 0
> BG77rR-C2lS-UXzm-V4KI-ptcG-1BZu-0yqVcV
>
> lv_swap VolGroup 1 -wi-ao---- 2.00g -1 -1 253
> 1
> i6uo0r-9T2P-OW60-Z1i7-6o7y-id1Q-1R7Lli
>
> lv_system1 vg_system1 1 rwi-a-r-r- 5.00g -1 -1 253
> 8 100.00
> GM0hKc-BHNV-CUAR-y0US-5Iw9-Ihi0-GSFD1V
>
> Finding all logical volumes
>
> ACTIVE '/dev/vg_system1/lv_system1' [5.00 GiB] inherit
>
> ACTIVE '/dev/VolGroup/lv_root' [9.51 GiB] inherit
>
> ACTIVE '/dev/VolGroup/lv_swap' [2.00 GiB] inherit
>
> [root@localhost ~]#
>
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@redhat.com
> https://www.redhat.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
[-- Attachment #2: Type: text/html, Size: 56020 bytes --]
prev parent reply other threads:[~2014-02-03 10:17 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-02 19:53 [linux-lvm] LVM2 [lvcreate --type raid5] issue James Woolliscroft
2014-02-03 10:17 ` Oliver Rath [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=52EF6CB4.7010505@mglug.de \
--to=rath@mglug.de \
--cc=linux-lvm@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.