Matthew Seaman wrote:

The filesystem you're writing to doesn't have sufficient inodes
available to create all of the files from the tarball.  Effectively
you need an inode for each file you create. Inodes are created at the
time the filesystem is generated: the newfs(8) command has an option
to set the number of bytes-per-inode: generally the defaults are fine,
but the bytes-per-inode setting should be set to no more than the
expected average size of files on the partition, and preferably rather
less than that.  Running out of inodes before you run out of disk
space is embarrassing.  Worse, it requires backing up the whole
partition, rebuilding the filesystem and then recovering the data from
backup in order to fix.

to add the information about inode usage to the df output. As a rule
of thumb, the %iused value should always be less than the percentage
capacity used.

hivemind# df -iH
Filesystem    Size   Used  Avail Capacity iused   ifree %iused  Mounted on
/dev/da0s1a   132M    46M    76M    38%    1404   14850    9%   /
/dev/da0s1g   2.7G   917M   1.6G    37%  118127  210191   36%   /usr
/dev/da0s1e   103M    15M    80M    16%    1316   11482   10%   /var
/dev/da0s1f    52M   387K    47M     1%      11    6387    0%   /tmp
/dev/da0s1h   320M   197M    97M    67%   35094    4072   90%   /files
/dev/da0s1d   415M    77M   305M    20%    1534   49536    3%   /db
/dev/ccd0c     54G    11G    39G    22%   33684 6551914    1%   /storage
procfs        4.1K   4.1K     0B   100%      61     983    6%   /proc

Well I believe that explains it. I guess with CVS files, mailing list archive, and webserver files, it became
an out of the ordinary partition.
Thanks for the heads up..

[EMAIL PROTECTED] mailing list
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Reply via email to