gluster-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gluster-devel] rm -rf sometimes failing in 2.0.0rc1


From: Anand Avati
Subject: Re: [Gluster-devel] rm -rf sometimes failing in 2.0.0rc1
Date: Tue, 10 Feb 2009 19:42:28 +0530

>> rm -rf /mnt/u1
>> rm: cannot remove directory `/mnt/u1/share/doc': Directory not empty
>> rm: cannot remove directory
>> `/mnt/u1/src/linux-headers-2.6.24-23-server/include/config': Directory not
>> empty
>>
>> I did similar commands on 7 other clients, and almost all of them failed to
>> delete share/doc (one succeeded).  After the failure, ls of the directory
>> shows empty, but an ls of the individual bricks shows that there are still
>> many subdirectories and some files still present.

>>
>> The two directories that gave problems have a higher than typical number of
>> entries.

Can you give us some more details about how the data was populated?
was it rsync'ed in? was it used for a kernel compile? were some nodes
down while any operation was performed on it?

Thanks,

Avati




reply via email to

[Prev in Thread] Current Thread [Next in Thread]