Bug 763320 (GLUSTER-1588)

Summary: No space left on device error
Product: [Community] GlusterFS Reporter: Anush Shetty <anush>
Component: distributeAssignee: Amar Tumballi <amarts>
Status: CLOSED WORKSFORME QA Contact:
Severity: high Docs Contact:
Priority: low    
Version: mainlineCC: gluster-bugs, vijay, vraman
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description Anush Shetty 2010-09-10 12:08:20 UTC
This was on a dht+afr setup. Even touching a file gives this error

domU-12-31-39-14-04-C2:/mnt/gluster# touch dot
touch: cannot touch `dot': No space left on device

domU-12-31-39-14-04-C2:/mnt/gluster# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/sda1             9.9G  2.5G  6.9G  27% /
tmpfs                 3.8G     0  3.8G   0% /lib/init/rw
udev                  3.8G   24K  3.8G   1% /dev
tmpfs                 3.8G     0  3.8G   0% /dev/shm
/dev/sdb              414G   17G  377G   5% /mnt
glusterfs#domU-12-31-39-14-04-C2:1156_10_09
                      827G   33G  753G   5% /mnt/gluster

[2010-09-10 11:57:43.572344] D [dht-common.c:232:dht_revalidate_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)
[2010-09-10 11:57:43.575005] D [dht-common.c:1889:dht_fd_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)
[2010-09-10 11:57:43.579051] D [dht-common.c:232:dht_revalidate_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)
[2010-09-10 11:57:43.579186] D [client3_1-fops.c:4745:client3_1_readdirp] 1156_10_09-client-0: (1): failed to get fd ctx. EBADFD
[2010-09-10 11:57:43.579251] D [client3_1-fops.c:4745:client3_1_readdirp] 1156_10_09-client-3: (1): failed to get fd ctx. EBADFD
[2010-09-10 11:57:43.579333] D [stat-prefetch.c:3581:sp_release] 1156_10_09: cache hits: 0, cache miss: 0
[2010-09-10 11:59:49.697502] W [dht-diskusage.c:216:dht_is_subvol_filled] 1156_10_09-dht: disk space on subvolume '1156_10_09-replicate-0' is getting full (100.00 %), consider adding more nodes
[2010-09-10 11:59:49.697555] D [dht-diskusage.c:253:dht_free_disk_available_subvol] 1156_10_09-dht: no subvolume has enough free space to create
[2010-09-10 11:59:49.697888] D [afr-transaction.c:960:afr_post_nonblocking_entrylk_cbk] 1156_10_09-replicate-0: Non blocking entrylks done. Proceeding to FOP
[2010-09-10 11:59:49.699761] E [afr-transaction.c:699:afr_changelog_pre_op_cbk] 1156_10_09-replicate-0: xattrop failed on child 1156_10_09-client-0: No space left on device
[2010-09-10 11:59:49.699936] E [afr-transaction.c:699:afr_changelog_pre_op_cbk] 1156_10_09-replicate-0: xattrop failed on child 1156_10_09-client-2: No space left on device
[2010-09-10 11:59:49.699967] E [afr-transaction.c:699:afr_changelog_pre_op_cbk] 1156_10_09-replicate-0: xattrop failed on child 1156_10_09-client-1: No space left on device
[2010-09-10 11:59:49.699987] D [afr-lk-common.c:410:transaction_lk_op] 1156_10_09-replicate-0: lk op is for a transaction
[2010-09-10 11:59:49.700245] W [fuse-bridge.c:1500:fuse_create_cbk] glusterfs-fuse: 511: /dot => -1 (No space left on device)
[2010-09-10 11:59:50.859936] D [dht-common.c:232:dht_revalidate_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)
[2010-09-10 11:59:50.863831] D [dht-common.c:1889:dht_fd_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)
[2010-09-10 11:59:50.869349] D [dht-common.c:232:dht_revalidate_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)
[2010-09-10 11:59:50.869552] D [client3_1-fops.c:4745:client3_1_readdirp] 1156_10_09-client-0: (1): failed to get fd ctx. EBADFD
[2010-09-10 11:59:50.869594] D [client3_1-fops.c:4745:client3_1_readdirp] 1156_10_09-client-3: (1): failed to get fd ctx. EBADFD
[2010-09-10 11:59:50.869696] D [stat-prefetch.c:3581:sp_release] 1156_10_09: cache hits: 0, cache miss: 0
[2010-09-10 12:00:08.446079] D [dht-common.c:232:dht_revalidate_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)
[2010-09-10 12:00:08.449855] D [dht-common.c:232:dht_revalidate_cbk] 1156_10_09-dht: subvolume 1156_10_09-replicate-1 returned -1 (No space left on device)

Comment 1 Vijay Bellur 2010-09-13 10:44:57 UTC
PATCH: http://patches.gluster.com/patch/4735 in master (No space left on device error)

Comment 2 Anush Shetty 2010-09-14 04:18:21 UTC
This is due to extended attributes getting filled up

root@pitta:/mnt/export2# getfattr -d -m . . 
# file: .
trusted.1004_14_09-posix.gen=0sTI77ZAAAAAE=
trusted.1206_13_09-posix.gen=0sTI3GhgAAAAE=
trusted.1208_13_09-posix.gen=0sTI3G6wAAAAE=
trusted.1209_13_09-posix.gen=0sTI3HMgAAAAE=
trusted.1215_13_09-posix.gen=0sTI3ImwAAAAE=
trusted.1402_10_09-posix.gen=0sTIntOgAAAAE=
trusted.1613_10_09-posix.gen=0sTIoL/QAAAAE=
trusted.1650_13_09-posix.gen=0sTI4JOAAAAAE=
trusted.1710_13_09-posix.gen=0sTI4NwAAAAAE=
trusted.1714_09_09-posix.gen=0sTIjIowAAAAE=
trusted.1720_09_09-posix.gen=0sTIjKCQAAAAE=
trusted.1724_09_09-posix.gen=0sTIjK7AAAAAE=
trusted.1749_13_09-posix.gen=0sTI4W3gAAAAE=
trusted.1810_13_09-posix.gen=0sTI4bwQAAAAE=
trusted.1815_13_09-posix.gen=0sTI4c4wAAAAE=
trusted.1816_13_09-posix.gen=0sTI4dMQAAAAE=
trusted.1817_13_09-posix.gen=0sTI4djQAAAAE=
trusted.1822_09_09-posix.gen=0sTIjYkAAAAAE=
trusted.1851_13_09-posix.gen=0sTI4lcgAAAAE=
trusted.1926_09_09-posix.gen=0sTIjnvgAAAAE=
trusted.afr.1111_14_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1111_14_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1111_14_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1111_14_09-replace-brick=0sAAAAAAAAAAAAACcQ
trusted.afr.1135_14_09-locks=0sAAAAAAAAAAEAAAAA
trusted.afr.1135_14_09-replace-brick=0sAAAAAAAAAjkAAAR2
trusted.afr.1143_14_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1143_14_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1143_14_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1143_14_09-replace-brick=0sAAAAAAAAAAAAAFcy
trusted.afr.1154_14_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1154_14_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1154_14_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1154_14_09-replace-brick=0sAAAAAAAAAAAAABPC
trusted.afr.1159_14_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1159_14_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1159_14_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1159_14_09-replace-brick=0sAAAAAAAAAAAAAE4g
trusted.afr.1206_13_09-client-0=0sAAAAAAAAAAEAAAAA
trusted.afr.1206_13_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1206_13_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1206_13_09-replace-brick=0sAAAAAAAAAAEAAAAA
trusted.afr.1206_14_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1206_14_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1208_13_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1208_13_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1208_13_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1208_13_09-replace-brick=0sAAAAAAAAAAEAAAAA
trusted.afr.1209_13_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1209_13_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1209_13_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1209_13_09-replace-brick=0sAAAAAAAAAAAAACcQ
trusted.afr.1215_13_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1215_13_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1215_13_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1215_13_09-replace-brick=0sAAAAAAAAAAAAACcQ
trusted.afr.1402_10_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1402_10_09-replace-brick=0sAAAAAAAAAAEAAADo
trusted.afr.1650_13_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1650_13_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1650_13_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1650_13_09-replace-brick=0sAAAAAAAAAAAAAB3Y
trusted.afr.1714_09_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1714_09_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1714_09_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1714_09_09-replace-brick=0sAAAAAAAAAAAAACcQ
trusted.afr.1724_09_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1724_09_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1724_09_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1724_09_09-replace-brick=0sAAAAAAAAAAAAACcQ
trusted.afr.1822_09_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1822_09_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1822_09_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1822_09_09-replace-brick=0sAAAAAAAAAAAAAE4g
trusted.afr.1851_13_09-locks=0sAAAAAAAAAAEAAAAA
trusted.afr.1851_13_09-replace-brick=0sAAAAAAAAAWoAAAF7
trusted.afr.1926_09_09-client-0=0sAAAAAAAAAAAAAAAA
trusted.afr.1926_09_09-client-1=0sAAAAAAAAAAAAAAAA
trusted.afr.1926_09_09-locks=0sAAAAAAAAAAAAAAAA
trusted.afr.1926_09_09-replace-brick=0sAAAAAAAAAAAAAAAK
trusted.gfid=0sAAAAAAAAAAAAAAAAAAAAAQ==
trusted.glusterfs.dht=0sAAAAAQAAAAAAAAAAf////g==
trusted.glusterfs.test="working\000"

root@pitta:/mnt/export2# getfattr -d -m . . | wc

Comment 3 Amar Tumballi 2010-09-14 06:47:08 UTC
This issue won't come in case of having a fresh backend.. (or even not overused backend for different volumes).

Hence will close this issue for now. Will open a bug about extended attribute cleanup script..