While testing file system performance of a NFS mounted directory with bonnie++, it exited with a "Delete files in sequential order...Bonnie: drastic I/O error (rmdir): Directory not empty" error.<br />
<br />
It appears that readdir() on a CentOS 6.2 NFS client (both x86 and x64, connecting to either a 6.2 or 5.7 NFS server) will fail to return all entries in an NFS directory. This is different than how the NFS client on CentOS 5.7 or Oracle 6.2 behaves.<br />
<br />
In the examples below, /home/greg/mnt/tmp is an NFS mounted directory. I created 20,480 0-length files by hand. nfsrm is included below.<br />
<br />
CentOS 5.7 (expected behavior):<br />
Linux 4eg5fre 2.6.18-274.17.1.el5 <a href="http://bugs.centos.org/view.php?id=1">0000001</a> SMP Tue Jan 10 17:25:58 EST 2012 x86_64 x86_64 x86_64 GNU/Linux<br />
[greg@4eg5fre ~]$ ls /home/greg/mnt/tmp | wc -l<br />
20480<br />
[greg@4eg5fre ~]$ ./nfsrm /home/greg/mnt/tmp<br />
[greg@4eg5fre ~]$ ls /home/greg/mnt/tmp | wc -l<br />
0<br />
<br />
Oracle 6.2 (expected behavior):<br />
Linux 8gpxkf1 2.6.32-300.3.1.el6uek.x86_64 <a href="http://bugs.centos.org/view.php?id=1">0000001</a> SMP Fri Dec 9 18:57:35 EST 2011 x86_64 x86_64 x86_64 GNU/Linux<br />
[greg@8gpxkf1 ~]$ ls /home/greg/mnt/tmp | wc -l<br />
20480<br />
[greg@8gpxkf1 ~]$ ./nfsrm /home/greg/mnt/tmp<br />
[greg@8gpxkf1 ~]$ ls /home/greg/mnt/tmp | wc -l<br />
0<br />
<br />
CentOS 6.2 client (unexpected behavior):<br />
Linux 9yf6091 2.6.32-220.4.1.el6.i686 <a href="http://bugs.centos.org/view.php?id=1">0000001</a> SMP Mon Jan 23 22:37:12 GMT 2012 i686 i686 i386 GNU/Linux<br />
[greg@9yf6091 ~]$ ls /home/greg/mnt/tmp | wc -l<br />
20480<br />
[greg@9yf6091 ~]$ ./nfsrm /home/greg/mnt/tmp<br />
[greg@9yf6091 ~]$ ls /home/greg/mnt/tmp | wc -l<br />
19316<br />
<br />
readdir() works as expected on 6.2 when using a local filesystem.
↧