[Gluster-users] can't write to gluster when local disk full

m.c.wilkins at massey.ac.nz m.c.wilkins at massey.ac.nz
Wed Apr 29 03:34:06 UTC 2009



Hi,

I have three machines in a NUFA arrangment.  When the local disk is
full no writes can be made to the gluster disk, even though there is
space on the other disks?!  Thanks for any help.

Perhaps I have done something wrong, this is very odd.  Here is what
df says:

tur-awc1:~> df -hl
Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/sysvg-root
                      7.7G  3.1G  4.3G  42% /
/dev/sda1              92M   25M   62M  29% /boot
tmpfs                 7.9G     0  7.9G   0% /dev/shm
/dev/sdb2             6.3T  3.4T  2.9T  55% /data
/dev/sdb1             5.4T  2.8T  2.7T  51% /export/brick0
glusterfs              26T   12T   15T  46% /awc/data
/dev/mapper/sysvg-newgluster
                      886G  886G   63M 100% /export/brick-newgluster
glusterfs#/usr/local/glusterfs-2.0.0rc9/etc/glusterfs/glusterfsd.vol
                      2.8T  887G  1.9T  32% /ng

Looks like plenty of room on the gluster, but my local brick is full,
and I can't write to /ng.  Here is the log:


================================================================================
Version      : glusterfs 2.0.0git built on Apr 29 2009 11:08:01
TLA Revision : git://git.sv.gnu.org/gluster.git
Starting Time: 2009-04-29 11:35:46
Command line : /usr/local/glusterfs-2.0.0rc9/sbin/glusterfsd -f /usr/local/glusterfs-2.0.0rc9/etc/glusterfs/glusterfsd.vol /ng 
PID          : 22711
System name  : Linux
Nodename     : tur-awc1
Kernel Release : 2.6.18-92.1.13.el5
Hardware Identifier: x86_64

Given volfile:
+------------------------------------------------------------------------------+
  1: volume posix0
  2:   type storage/posix
  3:   option directory /export/brick-newgluster
  4: end-volume
  5: 
  6: volume locks0
  7:   type features/locks
  8:   subvolumes posix0
  9: end-volume
 10: 
 11: volume brick0
 12:   type performance/io-threads
 13:   subvolumes locks0
 14: end-volume
 15: 
 16: volume server
 17:   type protocol/server
 18:   option transport-type tcp
 19:   option listen-port 16996
 20:   option auth.addr.brick0.allow *
 21:   subvolumes brick0
 22: end-volume
 23: 
 24: volume tur-awc1-0
 25:   type protocol/client
 26:   option transport-type tcp
 27:   option remote-port 16996
 28:   option remote-host tur-awc1
 29:   option remote-subvolume brick0
 30:   option ping-timeout 120
 31: end-volume
 32: 
 33: volume tur-awc2-0
 34:   type protocol/client
 35:   option transport-type tcp
 36:   option remote-port 16996
 37:   option remote-host tur-awc2
 38:   option remote-subvolume brick0
 39:   option ping-timeout 120
 40: end-volume
 41: 
 42: volume tur-awc3-0
 43:   type protocol/client
 44:   option transport-type tcp
 45:   option remote-port 16996
 46:   option remote-host tur-awc3
 47:   option remote-subvolume brick0
 48:   option ping-timeout 120
 49: end-volume
 50: 
 51: volume nufa
 52:    type cluster/nufa
 53:    option local-volume-name tur-awc1-0
 54:    subvolumes tur-awc1-0 tur-awc2-0 tur-awc3-0
 55: end-volume
 56: 
 57: volume writebehind
 58:    type performance/write-behind
 59:    option cache-size 1MB
 60:    subvolumes nufa
 61: end-volume
 62: 
 63: # before or after writebehind?
 64: volume ra
 65:    type performance/read-ahead
 66:    subvolumes writebehind
 67: end-volume
 68: 
 69: volume cache
 70:    type performance/io-cache
 71:    option cache-size 512MB
 72:    subvolumes ra
 73: end-volume
 74: 

+------------------------------------------------------------------------------+
2009-04-29 11:35:46 W [xlator.c:555:validate_xlator_volume_options] server: option 'listen-port' is deprecated, preferred is 'transport.socket.listen-port', continuing with correction
2009-04-29 11:35:46 N [glusterfsd.c:1152:main] glusterfs: Successfully started
2009-04-29 11:35:46 E [socket.c:743:socket_connect_finish] tur-awc2-0: connection failed (Connection refused)
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc2-0: disconnected
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc2-0: disconnected
2009-04-29 11:35:46 E [socket.c:743:socket_connect_finish] tur-awc2-0: connection failed (Connection refused)
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc2-0: disconnected
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc2-0: disconnected
2009-04-29 11:35:46 E [socket.c:743:socket_connect_finish] tur-awc1-0: connection failed (Connection refused)
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc1-0: disconnected
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc1-0: disconnected
2009-04-29 11:35:46 E [socket.c:743:socket_connect_finish] tur-awc1-0: connection failed (Connection refused)
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc1-0: disconnected
2009-04-29 11:35:46 N [client-protocol.c:7014:notify] tur-awc1-0: disconnected
2009-04-29 11:35:46 N [client-protocol.c:6327:client_setvolume_cbk] tur-awc3-0: connection and handshake succeeded
2009-04-29 11:35:46 N [client-protocol.c:6327:client_setvolume_cbk] tur-awc3-0: connection and handshake succeeded
2009-04-29 11:35:51 N [server-protocol.c:7502:mop_setvolume] server: accepted client from 130.123.128.35:1020
2009-04-29 11:35:51 N [server-protocol.c:7502:mop_setvolume] server: accepted client from 130.123.128.35:1019
2009-04-29 11:35:52 N [server-protocol.c:7502:mop_setvolume] server: accepted client from 130.123.128.28:1017
2009-04-29 11:35:52 N [server-protocol.c:7502:mop_setvolume] server: accepted client from 130.123.128.28:1016
2009-04-29 11:35:57 N [server-protocol.c:7502:mop_setvolume] server: accepted client from 130.123.129.121:1022
2009-04-29 11:35:57 W [client-protocol.c:6320:client_setvolume_cbk] tur-awc1-0: attaching to the local volume 'brick0'
2009-04-29 11:35:57 N [client-protocol.c:6327:client_setvolume_cbk] tur-awc1-0: connection and handshake succeeded
2009-04-29 11:35:57 N [client-protocol.c:6327:client_setvolume_cbk] tur-awc2-0: connection and handshake succeeded
2009-04-29 11:35:57 N [server-protocol.c:7502:mop_setvolume] server: accepted client from 130.123.129.121:1016
2009-04-29 11:35:57 W [client-protocol.c:6320:client_setvolume_cbk] tur-awc1-0: attaching to the local volume 'brick0'
2009-04-29 11:35:57 N [client-protocol.c:6327:client_setvolume_cbk] tur-awc1-0: connection and handshake succeeded
2009-04-29 11:35:57 N [client-protocol.c:6327:client_setvolume_cbk] tur-awc2-0: connection and handshake succeeded
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 14:35:06 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
<snip>
2009-04-29 15:18:23 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 15:18:23 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 15:18:23 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 15:18:23 W [posix.c:1697:posix_writev] posix0: writev failed: No space left on device
2009-04-29 15:18:23 E [fuse-bridge.c:923:fuse_err_cbk] glusterfs-fuse: 4939521: FLUSH() ERR => -1 (No space left on device)




More information about the Gluster-users mailing list