>>    Original Message
>> From: ToddAndMargo
>> Sent: Sunday, May 15, 2016 04:16
>> To: [log in to unmask]
>> Subject: Re: What is this xfadump error?
>>
>>>> Original Message
>>>> From: ToddAndMargo
>>>> Sent: Saturday, May 14, 2016 20:37
>>>> To: [log in to unmask]
>>>> Subject: What is this xfadump error?
>>>>
>>>> Hi All,
>>>>
>>>> $ rpm -qa \*xfsdump\*
>>>> xfsdump-3.1.4-1.el7.x86_64
>>>>
>>>> What is the meaning of this xfsdump error. There is
>>>> no /dev/sda on this system at the moment.
>>>>
>>>> # /usr/sbin/xfsdump -v verbose -M 1 -L 1 -l 0 -f
>>>> /lin-bak/2016-05-14_15-49-55_homeXfsDump /home
>>>>
>>>> /usr/sbin/xfsdump: dumping non-directory files
>>>> WARNING: Your hard drive is failing
>>>> Device: /dev/sda [SAT], unable to open device
>>>> /usr/sbin/xfsdump: ending media file
>>>> /usr/sbin/xfsdump: media file size 453262590208 bytes
>>>> /usr/sbin/xfsdump: dump size (non-dir files) : 453181559704 bytes
>>>> /usr/sbin/xfsdump: dump complete: 5404 seconds elapsed
>>>> /usr/sbin/xfsdump: Dump Summary:
>>>> /usr/sbin/xfsdump: stream 0 /lin-bak/2016-05-14_15-49-55_homeXfsDump
>>>> OK (success)
>>>> /usr/sbin/xfsdump: Dump Status: SUCCESS
>>>>
>>>>
>>>> And it succeeded!
>>>>
>>>> /home is on /dev/sdb3
>>>> /lin-bak is /dev/sdc1
>>>>
>>>> Really!
>>>>
>>>> # fdisk -l /dev/sda
>>>> fdisk: cannot open /dev/sda: No such file or directory
>>>>
>>>> Any idea?
>>>>
>>>> Many thanks,
>>>> -T
>>
>>
>> On 05/14/2016 09:58 PM, [log in to unmask] wrote:
>>> Sounds like your hard drive or the controller are having issues‎.
>>> By any chance is/var‎ on sda. XFSdump keeps a catalog in /var/lib/xfs which is used for incremental backups.
>>> I know this because about 3 or 4 years ago I had to file a bug ticket with Red Hat for RHEL 6 about improper selinux contexts on that directory.
>>>
>>>
>>
>> Hi Rrmarino,
>>
>> It is not an incremental dump. But .... It is two removable
>> backups in a row. Due to this bug
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=1257018
>>
>> The backup drive mounts on /dev/sda the first time,
>> gets removed and the second drive mounts on /dev/sdc.
>> This error occurred on the second drive.
>>
>> Maybe it is using this catalog?
>>


On 05/15/2016 08:01 AM, [log in to unmask] wrote:
> It always uses the catalog, even though you are doing full dumps it updates the catalog so you can do incremental in the future.
> It also knows where the backups were stored and where to restore them by default.
>

I think we have found the source of the error.  Now we wait ...
https://bugzilla.redhat.com/show_bug.cgi?id=1336160