Zfs out of disk space

hi,

have no a lot of experience with zfs so perhaps the question is quite simple but i didn't find any reasonable info how to resolve it.

thus, fresh solaris installation

14GB in 86 files copied after that got msg No space left on device (28) in spite of according to stat there is plenty free space.

any suggestion how to solve it.

===
zpool status
  pool: rpool
 state: ONLINE
  scan: scrub repaired 0 in 24m06s with 0 errors on Wed Jan 10 01:45:21 2018

config:

        NAME      STATE     READ WRITE CKSUM
        rpool     ONLINE       0     0     0
          c1t0d0  ONLINE       0     0     0

errors: No known data errors

===
zfs list
NAME                              USED  AVAIL  REFER  MOUNTPOINT
rpool                            28.6G   428G  4.52M  /rpool
rpool/ROOT                       18.3G   428G    31K  legacy
rpool/ROOT/solaris               18.3G   428G  18.0G  /
rpool/ROOT/solaris/var            230M   428G   226M  /var
rpool/VARSHARE                   2.53M   428G  2.44M  /var/share
rpool/VARSHARE/pkg                 63K   428G    32K  /var/share/pkg
rpool/VARSHARE/pkg/repositories    31K   428G    31K  /var/share/pkg/repositories
rpool/VARSHARE/zones               31K   428G    31K  /system/zones
rpool/dump                       8.25G   428G  8.00G  -
rpool/export                       96K   428G    32K  /export
rpool/export/home                  64K   428G    64K  /export/home
rpool/swap                       2.06G   428G  2.00G  -

===
zfs list -t snapshot
NAME                             USED  AVAIL  REFER  MOUNTPOINT
rpool/ROOT/solaris@install      56.2M      -  2.67G  -
rpool/ROOT/solaris/var@install  3.55M      -   220M  -

===
zfs get quota
NAME                             PROPERTY  VALUE  SOURCE
rpool                            quota     none   default
rpool/ROOT                       quota     none   default
rpool/ROOT/solaris               quota     none   default
rpool/ROOT/solaris@install       quota     -      -
rpool/ROOT/solaris/var           quota     none   default
rpool/ROOT/solaris/var@install   quota     -      -
rpool/VARSHARE                   quota     none   default
rpool/VARSHARE/pkg               quota     none   default
rpool/VARSHARE/pkg/repositories  quota     none   default
rpool/VARSHARE/zones             quota     none   default
rpool/dump                       quota     -      -
rpool/export                     quota     none   default
rpool/export/home                quota     none   default
rpool/swap                       quota     -      -

====
 zfs get refquota
NAME                             PROPERTY  VALUE  SOURCE
rpool                            refquota  none   default
rpool/ROOT                       refquota  none   default
rpool/ROOT/solaris               refquota  none   default
rpool/ROOT/solaris@install       refquota  -      -
rpool/ROOT/solaris/var           refquota  none   default
rpool/ROOT/solaris/var@install   refquota  -      -
rpool/VARSHARE                   refquota  none   default
rpool/VARSHARE/pkg               refquota  none   default
rpool/VARSHARE/pkg/repositories  refquota  none   default
rpool/VARSHARE/zones             refquota  none   default
rpool/dump                       refquota  -      -
rpool/export                     refquota  none   default
rpool/export/home                refquota  none   default
rpool/swap                       refquota  -      -

====
df -h rpool
Filesystem             Size   Used  Available Capacity  Mounted on
rpool                  457G   4.5M       428G     1%    /rpool

Strange. ZFS behaves like any other filesystem in many regards, so in this case something is weird. Try to identify the problematic file, and copy the others?

Hi,

Also, was that the full text of the error message ? Usually, out-of-space errors during package installation will mention the full path of whatever directory or file that was trying to be written to disk at the time. The full path might give you some clues as to where the problem lies, possibly. Was anything else showing an error at the console or in the kernel log ?

Have you used /tmp as a destination of the copy ?

On Solaris systems /tmp filesystem is free memory + swap, so you perhaps used that filesystem, filling it up.

Regards
Peasant.

hi,

the reason is found. it is midnight commander(shell link) on linux server. rsync & scp work correct.

well, even well known dog may be found mad.

cheers