Ocfs2 users - Mar 2007 - ocfs2_file_sendfile: 372 ERROR

Hi,

I'm using a 4 node cluster with ocfs2-1.2.4-2 on centos 4.4
(2.6.9-42.0.8.ELsmp) as a webservercluster running apache-2.2.x.

About 3 to 5 times an hour I can see error messages like:

kernel: (15315,0):ocfs2_file_sendfile:372 ERROR: status = -104
kernel: (15315,0):ocfs2_file_sendfile:372 ERROR: status = -32
kernel: (10622,0):ocfs2_file_sendfile:372 ERROR: status = -104
kernel: (10848,0):ocfs2_file_sendfile:372 ERROR: status = -32

How do I interpret the errors? What can I do? The system seems to work as
expected.

Stephan

#define ECONNRESET      104     /* Connection reset by peer */
#define EPIPE           32      /* Broken pipe */

Harmless. But do file a bug. oss.oracle.com/bugzilla. We should not be
printing the ERROR. It should be handled by the userspace.

Stephan Hendl wrote:
> Hi,
>
> I'm using a 4 node cluster with ocfs2-1.2.4-2 on centos 4.4
(2.6.9-42.0.8.ELsmp) as a webservercluster running apache-2.2.x.
>
> About 3 to 5 times an hour I can see error messages like:
>
> kernel: (15315,0):ocfs2_file_sendfile:372 ERROR: status = -104
> kernel: (15315,0):ocfs2_file_sendfile:372 ERROR: status = -32
> kernel: (10622,0):ocfs2_file_sendfile:372 ERROR: status = -104
> kernel: (10848,0):ocfs2_file_sendfile:372 ERROR: status = -32
>
> How do I interpret the errors? What can I do? The system seems to work as
expected.
>
> Stephan
>
>
> _______________________________________________
> Ocfs2-users mailing list
> Ocfs2-users@oss.oracle.com
> http://oss.oracle.com/mailman/listinfo/ocfs2-users
>

We are using an ipvs-based load balancer triggering 4 nodes running centos 4
with ocfs2 and apache. The home- as well as the webserver-directory are two
ocfs2-mount points. All the webservers see the same filesystem (ocfs2) and
deliver websites to the clients. In order to avoid heavy fs-locking we put the
logfiles on the local disks of every server so that write operations on the
ocfs2-filesystems are the minority.

hope that helps...
Stephan
>>> "Amr Elgbaly" <amr922@gmail.com> schrieb am
27.03.2007 um 11:37 in Nachricht
<b9356ee40703270237u6b4156fex43fa9f7a4e10d48d@mail.gmail.com>:
> hello !
> i want ask how u made webserver cluster with ocfs2
> i mean what u used ?
> 
> 
> Regards ,
> 
> On 3/26/07, Stephan Hendl <Stephan.Hendl@lds.brandenburg.de> wrote:
>>
>> Hi,
>>
>> I'm using a 4 node cluster with ocfs2-1.2.4-2 on centos 4.4 (
>> 2.6.9-42.0.8.ELsmp) as a webservercluster running apache-2.2.x.
>>
>> About 3 to 5 times an hour I can see error messages like:
>>
>> kernel: (15315,0):ocfs2_file_sendfile:372 ERROR: status = -104
>> kernel: (15315,0):ocfs2_file_sendfile:372 ERROR: status = -32
>> kernel: (10622,0):ocfs2_file_sendfile:372 ERROR: status = -104
>> kernel: (10848,0):ocfs2_file_sendfile:372 ERROR: status = -32
>>
>> How do I interpret the errors? What can I do? The system seems to work
as
>> expected.
>>
>> Stephan
>>
>>
>> _______________________________________________
>> Ocfs2-users mailing list
>> Ocfs2-users@oss.oracle.com 
>> http://oss.oracle.com/mailman/listinfo/ocfs2-users 
>>