error sda: WRITE SAME failed when installing package

Bug #1266737 reported by Harini M Patgar
18
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Confirmed
Medium
Andy Whitcroft

Bug Description

Description:
==========
Was installing perl packages using apt-get install command. I saw WRITE SAME failed displayed during the process.

Command used : apt-get install perl*
on my Ubuntu guest to install all perl packages.

I saw the message as below
.
.
.
.
Setting up perlindex (1.606-1) ...
Generating index (this may take a while)...sda: WRITE SAME failed. Manually zeroing.
 done.

it took few seconds to complete and later on the process continued.

Environment:
============

Host: Power8
Guest : ubuntu hosted on Power8

used below qemu command :

[root@localhost ubuntu_alpha1]# cat kvmg.p8.h
KERNEL=trusty-server-cloudimg-ppc64el-vmlinuz-generic
#FSIMAGE=trusty-server-cloudimg-ppc64el.img
FSIMAGE=trusty-server-cloudimg-ppc64el.img.qcow2
MEM=8G
qemu-system-ppc64 \
        -enable-kvm -M pseries -cpu POWER8 -smp cores=4,threads=1 -m $MEM \
        -net nic -net user \
        -device spapr-vscsi -drive file=$FSIMAGE \
        -kernel $KERNEL \
        -append "earlyprintk root=/dev/sda console=hvc0 --no-sessions" \
        -nographic

Revision history for this message
Brad Figg (brad-figg) wrote : Missing required logs.

This bug is missing log files that will aid in diagnosing the problem. From a terminal window please run:

apport-collect 1266737

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

Changed in linux (Ubuntu):
status: New → Incomplete
Changed in linux (Ubuntu):
status: Incomplete → Confirmed
Revision history for this message
Andy Whitcroft (apw) wrote :

The Ubuntu-3.13.0-0.15 kernel has been copied into the bootstrap archive and should be available for update. This kernel is based on the v3.13-rc7 kernel. If we could retest with that kernel that would help. Please report any testing here.

Changed in linux (Ubuntu):
importance: Undecided → Medium
tags: added: kernel-da-key
Changed in linux (Ubuntu):
assignee: nobody → Andy Whitcroft (apw)
Revision history for this message
Harini M Patgar (hapatgar) wrote :

I tested by upgrading my kernel using "apt-get dist-upgrade " command. I tried the same scenario that is mentioned in the bug after upgrade.

This is the state on my guest

root@ubuntu:~# dpkg --list | grep linux
ii libselinux1:ppc64el 2.2.1-1ubuntu2 ppc64el SELinux runtime shared libraries
ii linux-headers-3.13.0-0 3.13.0-0.15 all Header files related to Linux kernel version 3.13.0
ii linux-headers-3.13.0-0-generic 3.13.0-0.15 ppc64el Linux kernel headers for version 3.13.0 on PowerPC 64el SMP
ii linux-headers-3.13.0-1 3.13.0-1.16 all Header files related to Linux kernel version 3.13.0
ii linux-headers-3.13.0-1-generic 3.13.0-1.16 ppc64el Linux kernel headers for version 3.13.0 on PowerPC 64el SMP
ii linux-headers-generic 3.13.0.1.3 ppc64el Generic Linux kernel headers
ii linux-headers-virtual 3.13.0.1.3 ppc64el Transitional package.
ii linux-image-3.13.0-0-generic 3.13.0-0.15 ppc64el Linux kernel image for version 3.13.0 on PowerPC 64el SMP
ii linux-image-3.13.0-1-generic 3.13.0-1.16 ppc64el Linux kernel image for version 3.13.0 on PowerPC 64el SMP
ii linux-image-virtual 3.13.0.1.3 ppc64el This package will always depend on the latest minimal generic kernel image.
ii linux-virtual 3.13.0.1.3 ppc64el Minimal Generic Linux kernel and headers
ii util-linux 2.20.1-5.1ubuntu11 ppc64el Miscellaneous system utilities
root@ubuntu:~#

But still I am seeing the same issue again on my guest.

Setting up perlbrew (0.66-1) ...
Setting up perlindex (1.606-1) ...
Generating index (this may take a while)...[ 1468.347473] sda: WRITE SAME failed. Manually zeroing.
 done.
Setting up perlmagick (8:6.7.7.10-6ubuntu2) ...
Setting up perlprimer-doc (1.1.21-1) ...
Setting up perltidy (20120701-1) ...

Revision history for this message
Harini M Patgar (hapatgar) wrote :

you can get the guest access by logging into 9.3.189.114(root/.pasroot)
cd /var/lib/libvirt/images/harini/bug_verify_image
run ./kvmg_p8 <-- this has qemu command and this will get the console for guest.

Revision history for this message
Harini M Patgar (hapatgar) wrote :

I had same error of sda: WRITE SAME failed. Manually zeroing. when running the dio tests of LTP test suite on my ubuntu .

on other terminal of the guest I had error displayed as below :

sda: WRITE SAME failed. Manually zeroing.

and in dmesg also the error was collected , I will be pasting only part of the dmesg,

type=1006 audit(1389258733.373:16): pid=1675 uid=0 old auid=4294967295 new auid=1000 old ses=4294967295 new ses=2 res=1
sda: WRITE SAME failed. Manually zeroing.
type=1006 audit(1389264359.283:17): pid=29045 uid=0 old auid=4294967295 new auid=1000 old ses=4294967295 new ses=3 res=1
root@ubuntu:/opt/ltp#

previous to dio test , I ran fsx and filecpas tests from LTP suite.

Revision history for this message
Harini M Patgar (hapatgar) wrote :

For more info I have the below kernel

root@ubuntu:/opt/ltp# dpkg --list | grep linux
ii libselinux1:ppc64el 2.2.1-1ubuntu2 ppc64el SELinux runtime shared libraries
ii libv4l-0:ppc64el 1.0.0-1 ppc64el Collection of video4linux support libraries
ii libv4lconvert0:ppc64el 1.0.0-1 ppc64el Video4linux frame format conversion library
ii linux-headers-3.13.0-0 3.13.0-0.15 all Header files related to Linux kernel version 3.13.0
ii linux-headers-3.13.0-0-generic 3.13.0-0.15 ppc64el Linux kernel headers for version 3.13.0 on PowerPC 64el SMP
ii linux-headers-generic 3.13.0.0.2 ppc64el Generic Linux kernel headers
ii linux-headers-virtual 3.13.0.0.2 ppc64el Transitional package.
ii linux-image-3.13.0-0-generic 3.13.0-0.15 ppc64el Linux kernel image for version 3.13.0 on PowerPC 64el SMP
ii linux-image-extra-3.13.0-0-generic 3.13.0-0.15 ppc64el Linux kernel extra modules for version 3.13.0 on PowerPC 64el SMP
ii linux-image-virtual 3.13.0.0.2 ppc64el This package will always depend on the latest minimal generic kernel image.
ii linux-libc-dev:ppc64el 3.13.0-0.15 ppc64el Linux Kernel Headers for development
ii linux-sound-base 1.0.25+dfsg-0ubuntu4 all base package for ALSA and OSS sound systems
ii linux-virtual 3.13.0.0.2 ppc64el Minimal Generic Linux kernel and headers
ii util-linux 2.20.1-5.1ubuntu11 ppc64el Miscellaneous system utilities
root@ubuntu:/opt/ltp#

Revision history for this message
Andy Whitcroft (apw) wrote :

Has this kernel been copied out of the guest into the host? The way to confirm from inside the host is to use:

    cat /proc/version_signature

Revision history for this message
Harini M Patgar (hapatgar) wrote :

Yes I copied the kerenel image to the guest.

the below is used to open my guest

KERNEL=vmlinux-3.13.0-0-generic
#FSIMAGE=trusty-server-cloudimg-ppc64el.img
FSIMAGE=trusty-server-cloudimg-ppc64el.img.qcow2
MEM=8G
qemu-system-ppc64 \
        -enable-kvm -M pseries -cpu POWER8 -smp cores=4,threads=1 -m $MEM \
        -net nic -net user \
        -device spapr-vscsi -drive file=$FSIMAGE \
        -kernel $KERNEL \
        -append "earlyprintk root=/dev/sda console=hvc0 --no-sessions" \
        -nographic

#6c:ae:8b:69:21:bc

This the version on that image

root@ubuntu:~# cat /proc/version_signature
Ubuntu 3.13.0-0.15-generic 3.13.0-rc7
root@ubuntu:~#

Revision history for this message
Abdul (abdhalee) wrote :
Download full text (4.3 KiB)

Hello,

Verified this problem with 3.13.0-14-generic OS on a KVM guest running on Power 8 and the issue is no more reproducible

I tried the mentioned scenarios like
- apt-get install perl*
- dio test run from LTP

Both scenarios were successfull with out any problem or error message in dmesg

More details:

root@ubuntu:~# dmesg | tail
[ 12.026629] type=1400 audit(1393859454.002:8): apparmor="STATUS" operation="profile_replace" profile="unconfined" name="/sbin/dhclient" pid=744 comm="apparmor_parser"
[ 12.026640] type=1400 audit(1393859454.002:9): apparmor="STATUS" operation="profile_replace" profile="unconfined" name="/usr/lib/NetworkManager/nm-dhcp-client.action" pid=744 comm="apparmor_parser"
[ 12.026647] type=1400 audit(1393859454.002:10): apparmor="STATUS" operation="profile_replace" profile="unconfined" name="/usr/lib/connman/scripts/dhclient-script" pid=744 comm="apparmor_parser"
[ 12.027157] type=1400 audit(1393859454.002:11): apparmor="STATUS" operation="profile_replace" profile="unconfined" name="/usr/lib/NetworkManager/nm-dhcp-client.action" pid=744 comm="apparmor_parser"
[ 12.583881] init: plymouth-stop pre-start process (848) terminated with status 1
[ 13.361227] e1000: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
[ 13.362023] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
[ 94.161888] random: nonblocking pool is initialized
[ 575.930046] audit_printk_skb: 9 callbacks suppressed
[ 575.930054] type=1006 audit(1393863692.439:15): pid=9047 uid=0 old auid=4294967295 new auid=1000 old ses=4294967295 new ses=1 res=1
root@ubuntu:~#

cat diolog
Test Start Time: Mon Mar 3 22:05:30 2014
-----------------------------------------
Testcase Result Exit Value
-------- ------ ----------
dio01 PASS 0
dio02 PASS 0
dio03 PASS 0
dio04 PASS 0
dio05 PASS 0
dio06 PASS 0
dio07 PASS 0
dio08 PASS 0
dio09 PASS 0
dio10 PASS 0
dio11 PASS 0
dio12 PASS 0
dio13 PASS 0
dio14 PASS 0
dio15 PASS 0
dio16 PASS 0
dio17 PASS 0
dio18 PASS 0
dio19 PASS 0
dio20 PASS 0
dio21 PASS 0
dio22 PASS 0
dio23 PASS 0
dio24 PASS 0
dio25 PASS 0
dio26 PASS 0
dio27 PASS 0
dio28 PAS...

Read more...

Revision history for this message
Kalpana S Shetty (kalshett) wrote :

This issue no more seen on daily builds. Please close this issue.

Revision history for this message
Kalpana S Shetty (kalshett) wrote :

Today's build (mar 17) I installed on a NFS shared location and after the installation while bringing up the VM I see message "sda2: WRITE SAME failed. Manually zeroing." console and even dmesg has this message logged.

Attached the dmesg.?field.comment=Today's build (mar 17) I installed on a NFS shared location and after the installation while bringing up the VM I see message "sda2: WRITE SAME failed. Manually zeroing." console and even dmesg has this message logged.

Attached the dmesg.?field.comment=Today's build (mar 17) I installed on a NFS shared location and after the installation while bringing up the VM I see message "sda2: WRITE SAME failed. Manually zeroing." console and even dmesg has this message logged.

Attached the dmesg.

Revision history for this message
Abdul (abdhalee) wrote :

I could see this warning message again : dm-1: WRITE SAME failed. Manually zeroing. while creating file system on a LVM device.

Can you please have a look if this can be fixed or Ignore ?

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.