Bug 1211962 - Disperse volume: Input/output errors on nfs and fuse mounts during delete operation
Summary: Disperse volume: Input/output errors on nfs and fuse mounts during delete op...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: GlusterFS
Classification: Community
Component: disperse
Version: mainline
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
Assignee: Pranith Kumar K
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks: qe_tracker_everglades 1214822 1214994 1224181 1224185 1224188 1225283
TreeView+ depends on / blocked
 
Reported: 2015-04-15 10:10 UTC by Bhaskarakiran
Modified: 2016-11-23 23:13 UTC (History)
5 users (show)

Fixed In Version: glusterfs-3.8rc2
Doc Type: Bug Fix
Doc Text:
Clone Of:
: 1224181 1225283 (view as bug list)
Environment:
Last Closed: 2016-06-16 12:51:10 UTC
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Embargoed:


Attachments (Terms of Use)
sosreport of client (7.05 MB, application/x-xz)
2015-04-15 10:19 UTC, Bhaskarakiran
no flags Details

Description Bhaskarakiran 2015-04-15 10:10:05 UTC
Description of problem:
=======================

'rm -rf <dirs>' throws Input / output errors during delete operations.

[root@rhs-client29 linux]# ls -l
total 0
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.1
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.10
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.2
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.3
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.4
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.5
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.6
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.7
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.8
drwxr-xr-x. 3 root root 0 Apr 14 13:48 linux.9
[root@rhs-client29 linux]# rm -rf linux.1
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/HD-Audio-Models.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/HD-Audio.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/Procfile.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/VIA82xx-mixer.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/alsa-parameters.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/hdspm.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/powersave.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/seq_oss.html': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/serial-u16550.txt': Input/output error
rm: cannot remove `linux.1/linux-3.19/Documentation/sound/alsa/soc/overview.txt': Input/output error
^C
[root@rhs-client29 linux]# pwd
/mnt/nfs/linux
[root@rhs-client29 linux]# mount
/dev/mapper/vg_rhsclient29-lv_root on / type ext4 (rw)
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
devpts on /dev/pts type devpts (rw,gid=5,mode=620)
tmpfs on /dev/shm type tmpfs (rw,rootcontext="system_u:object_r:tmpfs_t:s0")
/dev/sda1 on /boot type ext4 (rw)
/dev/mapper/vg_rhsclient29-lv_home on /home type ext4 (rw)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)
gvfs-fuse-daemon on /root/.gvfs type fuse.gvfs-fuse-daemon (rw,nosuid,nodev)
sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)
vertigo:/testvol on /mnt/nfs type nfs (rw,noac,addr=10.70.34.56)
[root@rhs-client29 linux]# 

NFS Log file shows:
===============

[2015-04-15 09:55:42.678352] I [ec-heal.c:546:ec_heal_init] 0-ec: Healing '<gfid:05cae3d6-feb7-4618-ae0d-c1c033ca0ae2>/devicetree', gfid 9bfe7338-e8cb-4223-b55b-7cae0cad31b9
[2015-04-15 09:55:42.865724] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:42.865768] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:42.865895] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.159871] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.159908] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.159941] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.459021] W [ec-common.c:162:ec_check_status] 0-testvol-disperse-0: Operation failed on some subvolumes (up=FFF, mask=FFA, remaining=0, good=FFA, bad=5)
[2015-04-15 09:55:43.459084] I [ec-heal.c:546:ec_heal_init] 0-ec: Healing '<gfid:05cae3d6-feb7-4618-ae0d-c1c033ca0ae2>/devicetree', gfid 9bfe7338-e8cb-4223-b55b-7cae0cad31b9
[2015-04-15 09:55:43.459880] W [ec-common.c:407:ec_child_select] 0-testvol-disperse-0: Executing operation with some subvolumes unavailable (5)
[2015-04-15 09:55:43.609652] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.609725] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.610976] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.611088] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
[2015-04-15 09:55:43.611203] W [ec-combine.c:858:ec_combine_check] 0-testvol-disperse-0: Mismatching xdata in answers of 'LOOKUP'
^C


Version-Release number of selected component (if applicable):
=============================================================
[root@vertigo ~]# gluster --version
glusterfs 3.7dev built on Apr 13 2015 07:14:27
Repository revision: git://git.gluster.com/glusterfs.git
Copyright (c) 2006-2011 Gluster Inc. <http://www.gluster.com>
GlusterFS comes with ABSOLUTELY NO WARRANTY.
You may redistribute copies of GlusterFS under the terms of the GNU General Public License.

How reproducible:
==================
100%

Steps to Reproduce:
1. Create a distributed-disperse (2x(8+4)) volume, enable quota, USS and epoll
2. Untar a linux tarball into multiple directories on nfs mount
3. Now, delete the directories.


Actual results:
Input/ Output errors

Expected results:
No errors should be seen

Additional info:
sosreports of the node will be attached.

Comment 2 Bhaskarakiran 2015-04-15 10:19:38 UTC
Created attachment 1014668 [details]
sosreport of client

Comment 3 Bhaskarakiran 2015-04-24 04:59:10 UTC
This is seen even on fuse. 

[root@dhcp37-61 linux]# rm -rf dir.5
rm: cannot remove `dir.5/linux-3.19/Documentation/arm': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/devicetree/bindings/arm/rockchip': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/devicetree/bindings/clock': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/devicetree/bindings/drm': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/devicetree/bindings/iio': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/devicetree/bindings/mipi/dsi': Is a directory
rm: cannot remove `dir.5/linux-3.19/Documentation/devicetree/bindings/mips/cavium': Is a directory
rm: cannot remove `dir.5/linux-3.19/Documentation/devicetree/bindings/soc/fsl': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/filesystems': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/i2c': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/misc-devices': Input/output error
rm: cannot remove `dir.5/linux-3.19/Documentation/virtual/kvm': Is a directory
rm: cannot remove `dir.5/linux-3.19/Documentation/w1/masters': Is a directory
rm: cannot remove `dir.5/linux-3.19/arch/arc': Input/output error
rm: cannot remove `dir.5/linux-3.19/arch/arm/mach-gemini': Directory not empty
rm: cannot remove `dir.5/linux-3.19/arch/arm/mach-iop33x/include/mach': Input/output error
rm: cannot remove `dir.5/linux-3.19/arch/arm/mach-sa1100/include/mach': Is a directory
rm: cannot remove `dir.5/linux-3.19/arch/arm/mach-versatile/include': Input/output error


[root@dhcp37-61 linux]# df -h .
Filesystem      Size  Used Avail Use% Mounted on
ninja:/testvol  7.3T  109G  7.2T   2% /mnt/fuse1
You have new mail in /var/spool/mail/root
[root@dhcp37-61 linux]# mount
/dev/mapper/vg_dhcp3761-lv_root on / type ext4 (rw)
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
devpts on /dev/pts type devpts (rw,gid=5,mode=620)
tmpfs on /dev/shm type tmpfs (rw,rootcontext="system_u:object_r:tmpfs_t:s0")
/dev/vda1 on /boot type ext4 (rw)
none on /proc/sys/fs/binfmt_misc type binfmt_misc (rw)
sunrpc on /var/lib/nfs/rpc_pipefs type rpc_pipefs (rw)
ninja:/testvol on /mnt/fuse1 type fuse.glusterfs (rw,default_permissions,allow_other,max_read=131072)
[root@dhcp37-61 linux]#

Comment 4 Bhaskarakiran 2015-05-18 06:47:37 UTC
This seen on the latest 3.7.0.2 RHGS build too and the entries are becoming stale.

Comment 5 Anand Avati 2015-05-18 09:19:01 UTC
REVIEW: http://review.gluster.org/10806 (cluster/ec: Handle lookup failures while op in progress) posted (#1) for review on master by Pranith Kumar Karampuri (pkarampu)

Comment 6 Anand Avati 2015-05-19 09:05:16 UTC
COMMIT: http://review.gluster.org/10806 committed in master by Pranith Kumar Karampuri (pkarampu) 
------
commit 64d2a5a25b64601958c0f47ec2e6356226f97d4b
Author: Pranith Kumar K <pkarampu>
Date:   Mon May 18 14:46:55 2015 +0530

    cluster/ec: Handle lookup failures while op in progress
    
    Change-Id: Ia1834ec23d5de615526d4d4e4d2e32aff155b7f7
    BUG: 1211962
    Signed-off-by: Pranith Kumar K <pkarampu>
    Reviewed-on: http://review.gluster.org/10806
    Tested-by: NetBSD Build System
    Tested-by: Gluster Build System <jenkins.com>
    Reviewed-by: Xavier Hernandez <xhernandez>

Comment 7 Niels de Vos 2016-06-16 12:51:10 UTC
This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-3.8.0, please open a new bug report.

glusterfs-3.8.0 has been announced on the Gluster mailinglists [1], packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist [2] and the update infrastructure for your distribution.

[1] http://blog.gluster.org/2016/06/glusterfs-3-8-released/
[2] http://thread.gmane.org/gmane.comp.file-systems.gluster.user


Note You need to log in before you can comment on or make changes to this bug.