[#LU-4284] Unable to provision SLES SP2 as a Lustre server

[LU-4284] Unable to provision SLES SP2 as a Lustre server Created: 20/Nov/13 Updated: 04/Mar/14 Resolved: 04/Mar/14
Status:	Closed
Project:	Lustre
Component/s:	None
Affects Version/s:	None
Fix Version/s:	None

Type:

Bug

Priority:

Critical

Reporter:

Brett Lee (Inactive)

Assignee:

Bob Glossman (Inactive)

Resolution:

Fixed

Votes:

Labels:

None

Severity:

Rank (Obsolete):

11759

Description

Unable to install the SLES builds of Lustre. Dependency missing.

Malcolm found that the RHEL spec file for e2fsprogs has:
Provides: ldiskfsprogs = %

{version}

This was not found in the SLES version.

Comments

Comment by Bob Glossman (Inactive) [ 20/Nov/13 ]

To be a little more explicit, the top level file in the source tree e2fsprogs-SUSE_LINUX-11.spec.in is missing the line

Provides:       ldiskfsprogs = %{version}

that is found in the file e2fsprogs-RHEL-6.spec.in

Since the .spec file used in the rpmbuild command line in our SUSE package builds for e2fsprogs is e2fsprogs-SUSE_LINUX-11.spec I strongly suspect that adding that line to e2fsprogs-SUSE_LINUX-11.spec.in will put the necessary addition in our built RPMs for SLES 11 SP2 and SP3.

Comment by Bob Glossman (Inactive) [ 21/Nov/13 ]

proposed fix: http://review.whamcloud.com/8367

Comment by Brett Lee (Inactive) [ 22/Nov/13 ]

Restarted the entire process to ensure a clean bug report.
OS is SLES 11 SP2 for SAP.
Machine is a KVM VM - 2 GB RAM.

Seems to have at least two problems:
1. The "nothing provides" ldiskfsprogs as noted previously.
2. Upon reboot into the Lustre kernel, the root file system could not be mounted or fsck'd.
3. Perhaps due to what I installed, after the "grub update" there were two entries referring to the Lustre kernel, and none for the original kernel. Thus, could not roll back to the Failsafe kernel.

Captured some notes. A bit verbose, but as this is a new OS for the Lustre server it seemed reasonable to capture.

sles11sp2:~/lustre/# cat sles-lustre-install.txt
Lustre on SLES 11 SP 2

Get "Lustre" from here:
http://build.whamcloud.com/job/lustre-b2_5/arch=x86_64,build_type=server,distro=sles11sp2,ib_stack=inkernel/

Get "e2fsprogs" from here:
http://build.whamcloud.com/job/e2fsprogs-master/arch=x86_64,distro=sles11/

Scripted to reproduce easier:

wget \
\
http://build.whamcloud.com/job/e2fsprogs-master/arch=x86_64,distro=sles11/lastSuccessfulBuild/artifact/_topdir/RPMS/x86_64/e2fsprogs-1.42.7.wc2-0suse.x86_64.rpm \
\
http://build.whamcloud.com/job/lustre-b2_5/arch=x86_64,build_type=server,distro=sles11sp2,ib_stack=inkernel/lastSuccessfulBuild/artifact/artifacts/RPMS/x86_64/kernel-default-3.0.93-0.5_lustre.x86_64.rpm \
http://build.whamcloud.com/job/lustre-b2_5/arch=x86_64,build_type=server,distro=sles11sp2,ib_stack=inkernel/lastSuccessfulBuild/artifact/artifacts/RPMS/x86_64/kernel-default-base-3.0.93-0.5_lustre.x86_64.rpm \
http://build.whamcloud.com/job/lustre-b2_5/arch=x86_64,build_type=server,distro=sles11sp2,ib_stack=inkernel/lastSuccessfulBuild/artifact/artifacts/RPMS/x86_64/lustre-2.5.0-3.0.93_0.5_lustre_default.x86_64.rpm \
http://build.whamcloud.com/job/lustre-b2_5/arch=x86_64,build_type=server,distro=sles11sp2,ib_stack=inkernel/lastSuccessfulBuild/artifact/artifacts/RPMS/x86_64/lustre-modules-2.5.0-3.0.93_0.5_lustre_default.x86_64.rpm \
http://build.whamcloud.com/job/lustre-b2_5/arch=x86_64,build_type=server,distro=sles11sp2,ib_stack=inkernel/lastSuccessfulBuild/artifact/artifacts/RPMS/x86_64/lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64.rpm

RC=$?
echo $RC

Install e2fsprogs:

sles11sp2:~/lustre/e2fsprogs # ll
total 1316
~~rw-r~~r- 1 root root 1342882 Nov 22 01:56 e2fsprogs-1.42.7.wc2-0suse.x86_64.rpm
sles11sp2:~/lustre/e2fsprogs # zypper install *.rpm
Loading repository data...
Reading installed packages...
Resolving package dependencies...

The following package is going to be upgraded:
e2fsprogs

The following package is going to change vendor:
e2fsprogs SUSE LINUX Products GmbH, Nuernberg, Germany ->

The following package is not supported by its vendor:
e2fsprogs

1 package to upgrade, 1 to change vendor.
Overall download size: 1.3 MiB. After the operation, additional 3.0 MiB will be used.
Continue? [y/n/?] : y
Retrieving package e2fsprogs-1.42.7.wc2-0suse.x86_64 (1/1), 1.3 MiB (4.9 MiB unpacked)
Retrieving package e2fsprogs-1.42.7.wc2-0suse.x86_64 (1/1), 1.3 MiB (4.9 MiB unpacked)
Installing: e2fsprogs-1.42.7.wc2-0suse [done]
sles11sp2:~/lustre/e2fsprogs #

Install Lustre (ignore the dep, it should be included but not marked as such):

sles11sp2:~/lustre/lustre # ll
total 40068
~~rw-r~~r- 1 root root 18410610 Nov 11 11:53 kernel-default-3.0.93-0.5_lustre.x86_64.rpm
~~rw-r~~r- 1 root root 10930548 Nov 11 11:53 kernel-default-base-3.0.93-0.5_lustre.x86_64.rpm
~~rw-r~~r- 1 root root 7800341 Nov 11 12:13 lustre-2.5.0-3.0.93_0.5_lustre_default.x86_64.rpm
~~rw-r~~r- 1 root root 3443661 Nov 11 12:13 lustre-modules-2.5.0-3.0.93_0.5_lustre_default.x86_64.rpm
~~rw-r~~r- 1 root root 374587 Nov 11 12:13 lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64.rpm
sles11sp2:~/lustre/lustre # zypper install *.rpm
Loading repository data...
Reading installed packages...
Resolving package dependencies...
2 Problems:
Problem: nothing provides ldiskfsprogs >= 1.42.7.wc1 needed by lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64
Problem: nothing provides ldiskfsprogs >= 1.42.7.wc1 needed by lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64

Problem: nothing provides ldiskfsprogs >= 1.42.7.wc1 needed by lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64
Solution 1: do not install lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64
Solution 2: break lustre-osd-ldiskfs by ignoring some of its dependencies

Choose from above solutions by number or skip, retry or cancel [1/2/s/r/c] (c): 2

Problem: nothing provides ldiskfsprogs >= 1.42.7.wc1 needed by lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64
Solution 1: do not install lustre-2.5.0-3.0.93_0.5_lustre_default.x86_64
Solution 2: break lustre-osd-ldiskfs by ignoring some of its dependencies

Choose from above solutions by number or skip, retry or cancel [1/2/s/r/c] (c): 2
Resolving dependencies...
Resolving package dependencies...

The following NEW packages are going to be installed:
lustre lustre-modules lustre-osd-ldiskfs

The following packages are going to be upgraded:
kernel-default kernel-default-base

The following packages are going to change vendor:
kernel-default SUSE LINUX Products GmbH, Nuernberg, Germany ->
kernel-default-base SUSE LINUX Products GmbH, Nuernberg, Germany ->

The following packages are not supported by their vendor:
kernel-default kernel-default-base lustre lustre-modules lustre-osd-ldiskfs

2 packages to upgrade, 3 new, 2 to change vendor.
Overall download size: 39.1 MiB. After the operation, additional 63.7 MiB will be used.
Continue? [y/n/?] : y
Retrieving package kernel-default-base-3.0.93-0.5_lustre.x86_64 (1/5), 10.4 MiB (22.3 MiB unpacked)
Retrieving package kernel-default-3.0.93-0.5_lustre.x86_64 (2/5), 17.6 MiB (78.8 MiB unpacked)
Retrieving package lustre-modules-2.5.0-3.0.93_0.5_lustre_default.x86_64 (3/5), 3.3 MiB (20.3 MiB unpacked)
Retrieving package lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64 (4/5), 366.0 KiB (1.6 MiB unpacked)
Retrieving package lustre-2.5.0-3.0.93_0.5_lustre_default.x86_64 (5/5), 7.4 MiB (40.0 MiB unpacked)
Retrieving package kernel-default-base-3.0.93-0.5_lustre.x86_64 (1/5), 10.4 MiB (22.3 MiB unpacked)
Installing: kernel-default-base-3.0.93-0.5_lustre [done]
Additional rpm output:
Scanning scripts ...
Resolve dependencies ...
Install symlinks in /lib/mkinitrd/setup ...
Install symlinks in /lib/mkinitrd/boot ...

Kernel image: /boot/vmlinuz-3.0.93-0.5_lustre-default
Initrd image: /boot/initrd-3.0.93-0.5_lustre-default
Root device: /dev/vda2 (mounted on / as ext3)
Resume device: /dev/vda1
find: `/lib/modules/3.0.93-0.5_lustre-default/kernel/drivers/scsi/device_handler': No such file or directory
modprobe: Module thermal not found.
WARNING: no dependencies for kernel module 'thermal' found.
modprobe: Module processor not found.
WARNING: no dependencies for kernel module 'processor' found.
modprobe: Module fan not found.
WARNING: no dependencies for kernel module 'fan' found.
modprobe: Module virtio_blk not found.
WARNING: no dependencies for kernel module 'virtio_blk' found.
modprobe: Module ata_generic not found.
WARNING: no dependencies for kernel module 'ata_generic' found.
modprobe: Module virtio_pci not found.
WARNING: no dependencies for kernel module 'virtio_pci' found.
modprobe: Module virtio:d00000002v00001AF4 not found.
WARNING: no dependencies for kernel module 'virtio:d00000002v00001AF4' found.
modprobe: Module usbhid not found.
WARNING: no dependencies for kernel module 'usbhid' found.
Kernel Modules: scsi_mod libata ata_piix mbcache jbd ext3 usb-common usbcore ohci-hcd uhci-hcd ehci-hcd crc-t10dif sd_mod
Features: acpi block usb resume.userspace resume.kernel
36915 blocks
>>> Network: auto
>>> Calling mkinitrd -B -k /boot/vmlinuz-3.0.93-0.5_lustre-default -i /boot/initrd-3.0.93-0.5_lustre-default-kdump -f 'kdump network' -s ''
Regenerating kdump initrd ...

Kernel image: /boot/vmlinuz-3.0.93-0.5_lustre-default
Initrd image: /boot/initrd-3.0.93-0.5_lustre-default-kdump
Root device: /dev/vda2 (mounted on / as ext3)
Resume device: /dev/vda1
find: `/lib/modules/3.0.93-0.5_lustre-default/kernel/drivers/scsi/device_handler': No such file or directory
modprobe: Module thermal not found.
WARNING: no dependencies for kernel module 'thermal' found.
modprobe: Module processor not found.
WARNING: no dependencies for kernel module 'processor' found.
modprobe: Module fan not found.
WARNING: no dependencies for kernel module 'fan' found.
modprobe: Module virtio_blk not found.
WARNING: no dependencies for kernel module 'virtio_blk' found.
modprobe: Module ata_generic not found.
WARNING: no dependencies for kernel module 'ata_generic' found.
modprobe: Module virtio_pci not found.
WARNING: no dependencies for kernel module 'virtio_pci' found.
modprobe: Module virtio:d00000002v00001AF4 not found.
WARNING: no dependencies for kernel module 'virtio:d00000002v00001AF4' found.
modprobe: Module usbhid not found.
WARNING: no dependencies for kernel module 'usbhid' found.
modprobe: Module virtio:d00000001v00001AF4 not found.
WARNING: no dependencies for kernel module 'virtio:d00000001v00001AF4' found.
modprobe: Module nls_utf8 not found.
WARNING: no dependencies for kernel module 'nls_utf8' found.
Kernel Modules: scsi_mod libata ata_piix mbcache jbd ext3 usb-common usbcore ohci-hcd uhci-hcd ehci-hcd af_packet crc-t10dif sd_mod
Features: acpi block usb network resume.userspace resume.kernel kdump
50530 blocks
Don't refresh the bootloader. You may have to do that manually!

Retrieving package kernel-default-3.0.93-0.5_lustre.x86_64 (2/5), 17.6 MiB (78.8 MiB unpacked)
Installing: kernel-default-3.0.93-0.5_lustre [done]
Additional rpm output:

Kernel image: /boot/vmlinuz-3.0.93-0.5_lustre-default
Initrd image: /boot/initrd-3.0.93-0.5_lustre-default
Root device: /dev/vda2 (mounted on / as ext3)
Resume device: /dev/vda1
Kernel Modules: hwmon thermal_sys thermal processor fan virtio_ring virtio virtio_blk scsi_mod libata ata_piix ata_generic virtio_pci scsi_dh scsi_dh_hp_sw scsi_dh_rdac scsi_dh_alua scsi_dh_emc mbcache jbd ext3 usb-common usbcore ohci-hcd uhci-hcd ehci-hcd hid usbhid crc-t10dif sd_mod
Features: acpi block usb resume.userspace resume.kernel
38137 blocks
>>> Network: auto
>>> Calling mkinitrd -B -k /boot/vmlinuz-3.0.93-0.5_lustre-default -i /boot/initrd-3.0.93-0.5_lustre-default-kdump -f 'kdump network' -s ''
Regenerating kdump initrd ...

Kernel image: /boot/vmlinuz-3.0.93-0.5_lustre-default
Initrd image: /boot/initrd-3.0.93-0.5_lustre-default-kdump
Root device: /dev/vda2 (mounted on / as ext3)
Resume device: /dev/vda1
Kernel Modules: hwmon thermal_sys thermal processor fan virtio_ring virtio virtio_blk scsi_mod libata ata_piix ata_generic virtio_pci scsi_dh scsi_dh_hp_sw scsi_dh_rdac scsi_dh_alua scsi_dh_emc mbcache jbd ext3 usb-common usbcore ohci-hcd uhci-hcd ehci-hcd hid usbhid af_packet virtio_net nls_utf8 crc-t10dif sd_mod
Features: acpi block usb network resume.userspace resume.kernel kdump
51833 blocks
Don't refresh the bootloader. You may have to do that manually!

Retrieving package lustre-modules-2.5.0-3.0.93_0.5_lustre_default.x86_64 (3/5), 3.3 MiB (20.3 MiB unpacked)
Installing: lustre-modules-2.5.0-3.0.93_0.5_lustre_default [done]
Retrieving package lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default.x86_64 (4/5), 366.0 KiB (1.6 MiB unpacked)
Installing: lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default [done]
Retrieving package lustre-2.5.0-3.0.93_0.5_lustre_default.x86_64 (5/5), 7.4 MiB (40.0 MiB unpacked)
Installing: lustre-2.5.0-3.0.93_0.5_lustre_default [done]
sles11sp2:~/lustre/lustre #

sles11sp2:~/lustre/lustre # rpm -qa|egrep 'kernel-default|lustre|e2fs'|sort
e2fsprogs-1.42.7.wc2-0suse
kernel-default-3.0.93-0.5_lustre
kernel-default-base-3.0.93-0.5_lustre
kernel-default-devel-3.0.13-0.27.1
lustre-2.5.0-3.0.93_0.5_lustre_default
lustre-modules-2.5.0-3.0.93_0.5_lustre_default
lustre-osd-ldiskfs-2.5.0-3.0.93_0.5_lustre_default
sles11sp2:~/lustre/lustre #

sles11sp2:/boot/grub # cat menu.lst

Modified by YaST2. Last modification on Fri Nov 22 12:43:41 MST 2013
default 0
timeout 8
##YaST - generic_mbr
##YaST - activate

###Don't change this comment - YaST2 identifier: Original name: linux###
title SLES for SAP Applications - 3.0.93-0.5_lustre
root (hd0,1)
kernel /boot/vmlinuz-3.0.93-0.5_lustre-default root=/dev/vda2 resume=/dev/vda1 splash=silent crashkernel=256M-:128M text console=ttyS0,115200 sshd=1
showopts
initrd /boot/initrd-3.0.93-0.5_lustre-default

###Don't change this comment - YaST2 identifier: Original name: failsafe###
title Failsafe – SLES for SAP Applications - 3.0.93-0.5_lustre
root (hd0,1)
kernel /boot/vmlinuz-3.0.93-0.5_lustre-default root=/dev/vda2 showopts ide=nodma apm=off noresume edd=off powersaved=off nohz=off highres=off proces
sor.max_cstate=1 nomodeset x11failsafe
initrd /boot/initrd-3.0.93-0.5_lustre-default
sles11sp2:/boot/grub #

sles11sp2:~ # uname -a
Linux sles11sp2 3.0.13-0.27-default #1 SMP Wed Feb 15 13:33:49 UTC 2012 (d73692b) x86_64 x86_64 x86_64 GNU/Linux
sles11sp2:~ # reboot

...
...
...

Start Unicode mode done

fsck failed. Please repair manually and reboot. The root
file system is currently mounted read-only. To remount it
read-write do:

bash# mount -n -o remount,rw /

Attention: Only CONTROL-D will reboot the system in this
maintanance mode. shutdown or reboot will not work.

Give root password for maintenance
(or type Control-D to continue):
(repair filesystem) # mount -n -o remount,rw /
[ 135.376104] EXT3-fs (vda2): using internal journal
(repair filesystem) # fsck -yv /
fsck from util-linux 2.19.1
fsck.ext3: symbol lookup error: fsck.ext3: undefined symbol: set_com_err_gettext
(repair filesystem) #

Comment by Brett Lee (Inactive) [ 04/Mar/14 ]

Resolved.

Generated at Sat Feb 10 01:41:20 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.

[LU-4284] Unable to provision SLES SP2 as a Lustre server Created: 20/Nov/13 Updated: 04/Mar/14 Resolved: 04/Mar/14