[lustre-discuss] slow mounting

Michael Di Domenico mdidomenico4 at gmail.com
Fri Sep 8 10:57:16 PDT 2017


thanks.  the issue seemed to clear up over time.  i'm not sure i've
tracked it down enough though.  it might have been related to a mount
storm when i rebooted a large cluster this morning.

On Fri, Sep 8, 2017 at 12:31 PM, Mannthey, Keith
<keith.mannthey at intel.com> wrote:
> You might check that you have the MGS as the first nid listed on the mount ha command.  Some slow mounts are due to the mount trying the MDT first and going through a timeout before mount.
>
> Thanks,
>  Keith
>
>
> -----Original Message-----
> From: lustre-discuss [mailto:lustre-discuss-bounces at lists.lustre.org] On Behalf Of Michael Di Domenico
> Sent: Friday, September 8, 2017 5:43 AM
> To: lustre-discuss <lustre-discuss at lists.lustre.org>
> Subject: [lustre-discuss] slow mounting
>
> i just installed 2.10.0-1 on top of rhel 7.4.  i started using all the new systemd and lnetctl stuff and abandoned some of my custom scripts.
> however, now when i try to mount lustre, it takes a real long time.  i believe it will eventually mount, but the log is showing
>
> request sent has timeout for slow reply:
>
> the storage is a sonexion unit with lnet routers in between.  i believe i setup the lnet nid and lnet routes correctly, but i'm having trouble determining why/where the mount is actually going so slow or failing
>
> i can ping everything and lctl ping everything so the network is "up"
> as far as i can tell
> _______________________________________________
> lustre-discuss mailing list
> lustre-discuss at lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org


More information about the lustre-discuss mailing list