[Bugs] [Bug 1724754] fallocate of a file larger than brick size leads to increased brick usage despite failure

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1724754



--- Comment #2 from Raghavendra Bhat <rabhat at redhat.com> ---

Description of problem:
======================
when we use fallocate to create a file which is >= or to the max disk capacity, 
while we get a CLI error "fallocate: fallocate failed: No space left on device" 

However, the file gets created, and if you check the file size  on the mount 
shows zero size, but if you check the volume space on the client (df -h) , it
can be seen that the file is occupying significant space, that is because on
the backend bricks  the file is created upto size of about 90% of the disk
size( may be because of storage reserve space)


How reproducible:
===================
always


Steps to Reproduce:
1.create a 1x3 volume and fuse mount it
2. use fallocate to create a file which is >=size of of the brick
3. you would get error "fallocate: fallocate failed: No space left on device" 

Actual results:
============
however file is created and it shows as zero size file from mount point,
but the file does occupy about 90% of the brick size backend and the same
reflects in df -h of the mount point




>From client:

[root at hostname2]# pwd
/mnt/nfnas/falloc-test
[root at hostname2]# df -h
Filesystem                                Size  Used Avail Use% Mounted on
/dev/mapper/rhel_dhcp42--60-root           44G  1.6G   43G   4% /
devtmpfs                                  3.9G     0  3.9G   0% /dev
tmpfs                                     3.9G     0  3.9G   0% /dev/shm
tmpfs                                     3.9G  8.5M  3.9G   1% /run
tmpfs                                     3.9G     0  3.9G   0% /sys/fs/cgroup
/dev/sda1                                1014M  188M  827M  19% /boot
tmpfs                                     783M     0  783M   0% /run/user/0
hostname1:nfnas  2.2T  453G  1.8T  21% /mnt/nfnas ====> NOTICE THE USED SIZE OF
Storage space 
[root at hostname2]# fallocate test -l 600GB
fallocate: fallocate failed: No space left on device
[root at hostname2]# df -h
Filesystem                                Size  Used Avail Use% Mounted on
/dev/mapper/rhel_dhcp42--60-root           44G  1.6G   43G   4% /
devtmpfs                                  3.9G     0  3.9G   0% /dev
tmpfs                                     3.9G     0  3.9G   0% /dev/shm
tmpfs                                     3.9G  8.5M  3.9G   1% /run
tmpfs                                     3.9G     0  3.9G   0% /sys/fs/cgroup
/dev/sda1                                1014M  188M  827M  19% /boot
tmpfs                                     783M     0  783M   0% /run/user/0
hostname1:nfnas  2.2T  925G  1.3T  43% /mnt/nfnas ===>NOTICE THE INCREASE IN
USED SPACE
[root at hostname2]# ls
test
[root at dhcp42-60 falloc-test]# du -sh test
0       test
[root at hostname2]# stat test
  File: ‘test’
  Size: 0               Blocks: 0          IO Block: 131072 regular empty file
Device: 26h/38d Inode: 13717993992350864287  Links: 1
Access: (0644/-rw-r--r--)  Uid: (    0/    root)   Gid: (    0/    root)
Context: system_u:object_r:fusefs_t:s0
Access: 2019-05-08 18:24:52.342403239 +0530
Modify: 2019-05-08 18:24:52.342403239 +0530
Change: 2019-05-08 18:24:52.342403239 +0530
 Birth: -
[root at hostname2]# 



from server:
[root at hostname1]# ls /gluster/brick1
nfnas
[root at hostname1]# ls /gluster/brick1
brick1/  brick10/ brick11/ 
[root at hostname1]# ls /gluster/brick1/nfnas/
falloc-test  IOs  logs
[root at hostname1]# ls /gluster/brick1/nfnas/falloc-test/
test
[root at hostname1]# ls /gluster/brick1/nfnas/falloc-test/test 
/gluster/brick1/nfnas/falloc-test/test
[root at hostname1]# du -sh /gluster/brick1/nfnas/falloc-test/test
473G    /gluster/brick1/nfnas/falloc-test/test
[root at hostname1]# stat /gluster/brick1/nfnas/falloc-test/test
  File: ‘/gluster/brick1/nfnas/falloc-test/test’
  Size: 0               Blocks: 990030216  IO Block: 4096   regular empty file
Device: fd17h/64791d    Inode: 1749722171  Links: 2
Access: (0644/-rw-r--r--)  Uid: (    0/    root)   Gid: (    0/    root)
Context: system_u:object_r:glusterd_brick_t:s0
Access: 2019-05-08 18:24:52.343774310 +0530
Modify: 2019-05-08 18:24:52.343774310 +0530
Change: 2019-05-08 18:24:52.366773892 +0530
 Birth: -
[root at hostname1]# df -h /gluster/brick1/
Filesystem                           Size  Used Avail Use% Mounted on
/dev/mapper/GLUSTER_vg1-GLUSTER_lv1  547G  541G  6.9G  99% /gluster/brick1

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.