[lustre-discuss] Mount lustre client with MDS/MGS backup

[Pardo Diaz, Alfonso]

Hello Richard,

I still having the same problem in my system. My clients is stucked in the primary MDS, that it's down, and It doesn’t use the backup (service MDS), but only when try to connect there first time.
As I said in previous messages, the client connected when the primary was ok it can use the service MDS without problems.

Any suggestion?


> El 15 sept 2016, a las 6:46, Mohr Jr, Richard Frank (Rick Mohr) <rmohr at utk.edu> escribió:
> 
> Alfonso,
> 
> Are you still having problems with this, or were you able to get it resolved?
> 
> --
> Rick Mohr
> Senior HPC System Administrator
> National Institute for Computational Sciences
> http://www.nics.tennessee.edu
> 
> 
>> On Sep 1, 2016, at 12:43 PM, Pardo Diaz, Alfonso <alfonso.pardo at ciemat.es> wrote:
>> 
>> Hi!
>> 
>> I am using a combined MDS/MGS. This is my config:
>> 
>> Checking for existing Lustre data: found
>> Reading CONFIGS/mountdata
>> 
>>  Read previous values:
>> Target:     fs-MDT0000
>> Index:      0
>> Lustre FS:  fs
>> Mount type: ldiskfs
>> Flags:      0x1005
>>             (MDT MGS no_primnode )
>> Persistent mount opts: user_xattr,errors=remount-ro
>> Parameters:  failover.node=192.168.8.9 at o2ib:192.168.8.10 at o2ib mdt.identity_upcall=NONE
>> 
>> 
>> 
>> 
>> Alfonso Pardo Diaz
>> System Administrator / Researcher
>> c/ Sola nº 1; 10200 Trujillo, ESPAÑA
>> Tel: +34 927 65 93 17 Fax: +34 927 32 32 37
>> 
>> 
>> 
>> ________________________________________
>> De: Ben Evans [bevans at cray.com]
>> Enviado el: jueves, 01 de septiembre de 2016 15:25
>> Para: Pardo Diaz, Alfonso; Mohr Jr, Richard Frank (Rick Mohr)
>> Cc: lustre-discuss at lists.lustre.org
>> Asunto: Re: [lustre-discuss] Mount lustre client with MDS/MGS backup
>> 
>> where is the MGS mounted, and now is it configured?
>> 
>> -Ben Evans
>> 
>> On 9/1/16, 2:16 AM, "lustre-discuss on behalf of Pardo Diaz, Alfonso"
>> <lustre-discuss-bounces at lists.lustre.org on behalf of
>> alfonso.pardo at ciemat.es> wrote:
>> 
>>> Oppps, damm copy and paste!
>>> 
>>> I am writing the correct output with same result. If the MDT is mounted
>>> in the backup MDS (192.168.8.10) the mounted client work OK, but new
>>> clients throw the next error:
>>> 
>>> mount -v -t lustre 192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs /mnt/fs
>>> arg[0] = /sbin/mount.lustre
>>> arg[1] = -v
>>> arg[2] = -o
>>> arg[3] = rw
>>> arg[4] = 192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs
>>> arg[5] = /mnt/fs
>>> source = 192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs
>>> (192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs), target = /mnt/fs
>>> options = rw
>>> mounting device 192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs at /mnt/fs,
>>> flags=0x1000000 options=device=192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs
>>> mount.lustre: mount 192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs at /mnt/fs
>>> failed: Input/output error retries left: 0
>>> mount.lustre: mount 192.168.8.9 at o2ib:192.168.8.10 at o2ib:/fs at /mnt/fs
>>> failed: Input/output error
>>> Is the MGS running?
>>> 
>>> 
>>> 
>>> 
>>>> El 31 ago 2016, a las 15:32, Mohr Jr, Richard Frank (Rick Mohr)
>>>> <rmohr at utk.edu> escribió:
>>>> 
>>>> 
>>>>> On Aug 31, 2016, at 8:12 AM, Pardo Diaz, Alfonso
>>>>> <alfonso.pardo at ciemat.es> wrote:
>>>>> 
>>>>> I mount my clients: mount -t lustre mds1 at o2ib:mds2 at o2ib:/fs /mnt/fs
>>>>> 
>>>>> 1) When both MDS are OK I can mount without problems
>>>>> 2) If the MDS1 is down and my clients have lustre mounted, they use
>>>>> MDS2 without problems
>>>>> 3) If the MDS1 is down and I try to mount a new client, It can¹t mount
>>>>> lustre with the next error:
>>>>> 
>>>>> 
>>>> <snip>
>>>>> arg[4] = 192.168.8.9 at o2ib:192.168.8.9 at o2ib:/fs
>>>> 
>>>> The client is resolving both hostnames (mds1 and mds2) to the same IP
>>>> address.  I am guessing that this corresponds to mds1, so when it is
>>>> down, there is no second host for the client to try.  Try specifying IP
>>>> addresses instead of hostnames and see if that make a difference.
>>>> 
>>>> --
>>>> Rick Mohr
>>>> Senior HPC System Administrator
>>>> National Institute for Computational Sciences
>>>> http://www.nics.tennessee.edu
>>>> 
>>> 
>>> _______________________________________________
>>> lustre-discuss mailing list
>>> lustre-discuss at lists.lustre.org
>>> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
> 
>