[Lustre-discuss] index alread in use

[Mohr Jr, Richard Frank (Rick Mohr)]

Brian,

Did you add --writeconf option to mkfs.lustre when you reformatted your MDT?  See "Erasing a File System" in the Lustre manual:

http://build.whamcloud.com/job/lustre-manual/lastSuccessfulBuild/artifact/lustre_manual.xhtml#dbdoclet.50438194_70905

-- 
Rick Mohr
Senior HPC System Administrator
National Institute for Computational Sciences
http://www.nics.tennessee.edu


On Jun 2, 2014, at 1:08 PM, "Andrus, Brian Contractor" <bdandrus at nps.edu> wrote:

> All,
>  
> I am trying to setup another lustre filesystem.
>  
> I have one named WORK and the new one is named SCRATCH
>  
> Separate MGS from the 2 MDTs
> I have the MGS and WORK MDT mounted.
> When I tried mounting the new SCRATCH MDT, I get:
>  
>                 LustreError: 140-5: Server SCRATCH-MDT0000 requested index 0, but that index is already in use. Use --writeconf to force
>  
> This seems new/different to me (I had this setup running for some time, but have recently reformatted the SCRATCH filesystem).
>  
> I am aware of indicating indices for OSTs when formatting them, but not MDTs.
> Is this an index relative to MDTs mounted on the same server? What is an appropriate number to use and what is it relative to?
>  
> Thanks in advance,
>  
>  
> Brian Andrus
> ITACS/Research Computing
> Naval Postgraduate School
> Monterey, California
> voice: 831-656-6238
>  
>  
> _______________________________________________
> Lustre-discuss mailing list
> Lustre-discuss at lists.lustre.org
> http://lists.lustre.org/mailman/listinfo/lustre-discuss