[Lustre-discuss] More failover issues

[Robert LeBlanc]

My MDTs and MGS are separate since I have two MDTs, I didn¹t want the MGS
tied to just one so I separated it. It seems that you should be able to
specify more than one MGS for the MDT because there is really no other way
to tell it what the MGS failover partner is.

Robert


On 11/12/07 3:23 PM, "Wojciech Turek" <wjt27 at cam.ac.uk> wrote:

> Yes but in given example in section 2.2.2.1 two mgsnodes are specified for
> --ost and you are specifying it for --mdt maybe that is the problem? Do you
> have combined mgs with mdt ? Do you have one file system or more?
> 
> Wojciech
> On 12 Nov 2007, at 21:56, Robert LeBlanc wrote:
> 
>>  Yes only one MGS per site, but you should be able to specify multiple MGS
>> nodes. We have done it before with 1.6.0. See
>> http://manual.lustre.org/manual/LustreManual16_HTML/DynamicHTML-05-1.html
>> section 2.2.2.1.
>>  
>>  Robert
>>  
>>  
>>  On 11/12/07 2:48 PM, "Wojciech Turek" <wjt27 at cam.ac.uk> wrote:
>>  
>>  
>>> Hi,
>>>  
>>>  I think this is because there can be only one MGS per lustre installation
>>> (this is what manual says).
>>>  
>>>  Wojciech Turek
>>>  On 12 Nov 2007, at 21:18, Robert LeBlanc wrote:
>>>  
>>>  
>>>>  
>>>>  
>>>>  This is what I'm getting:
>>>>   
>>>>   head2-2:~# mkfs.lustre --mkfsoptions="-O dir_index" --reformat --mdt
>>>> --fsname=home --mgsnode=192.168.1.252 at o2ib --mgsnode=192.168.1.253 at o2ib
>>>> --failnode=192.168.1.252 at o2ib /dev/mapper/ldiskd-part1
>>>>   
>>>>      Permanent disk data:
>>>>   Target:     home-MDTffff
>>>>   Index:      unassigned
>>>>   Lustre FS:  home
>>>>   Mount type: ldiskfs
>>>>   Flags:      0x71
>>>>                 (MDT needs_index first_time update )
>>>>   Persistent mount opts: errors=remount-ro,iopen_nopriv,user_xattr
>>>>   Parameters:  mgsnode=192.168.1.253 at o2ib failover.node=192.168.1.252 at o2ib
>>>> mdt.group_upcall=/usr/sbin/l_getgroups
>>>>   
>>>>   device size = 972MB
>>>>   formatting backing filesystem ldiskfs on /dev/mapper/ldiskd-part1
>>>>           target name  home-MDTffff
>>>>           4k blocks     0
>>>>           options       -O dir_index -i 4096 -I 512 -q -F
>>>>   mkfs_cmd = mkfs.ext2 -j -b 4096 -L home-MDTffff -O dir_index -i 4096 -I
>>>> 512 -q -F /dev/mapper/ldiskd-part1
>>>>   Writing CONFIGS/mountdata
>>>>   
>>>>   
>>>>   For some reason, only the last --mgsnode option is being kept.
>>>>   
>>>>   Robert
>>>>   
>>>>   
>>>>   -----Original Message-----
>>>>   From: Nathan Rutman [mailto:Nathan.Rutman at Sun.COM]
>>>>   Sent: Mon 11/12/2007 1:51 PM
>>>>   To: Robert LeBlanc
>>>>   Cc: lustre
>>>>   Subject: Re: [Lustre-discuss] More failover issues
>>>>   
>>>>   Robert LeBlanc wrote:
>>>>   > In 1.6.0, when creating a MDT, you could specify multiple --mgsnode
>>>> options
>>>>   > and it would failover between them. 1.6.3 only seems to take the last
>>>> one
>>>>   > and --mgsnode=192.168.1.252 at o2ib:192.168.1.253 at o2ib doesn't seem to
>>>> failover
>>>>   > to the other node. Any ideas how to get around this?
>>>>   >  
>>>>   Multiple --mgsnode parameters should work:
>>>>   mkfs.lustre --mkfsoptions="-O dir_index" --reformat --mdt
>>>>   --mgsnode=192.168.1.253 at o2ib --mgsnode=1 at elan --device-size=10000
>>>> /tmp/foo
>>>>   
>>>>      Permanent disk data:
>>>>   Target:     lustre-MDTffff
>>>>   Index:      unassigned
>>>>   Lustre FS:  lustre
>>>>   Mount type: ldiskfs
>>>>   Flags:      0x71
>>>>                 (MDT needs_index first_time update )
>>>>   Persistent mount opts: errors=remount-ro,iopen_nopriv,user_xattr
>>>>   Parameters: mgsnode=192.168.1.253 at o2ib mgsnode=1 at elan
>>>>   
>>>>   > Robert
>>>>   > 
>>>>   > Robert LeBlanc
>>>>   > College of Life Sciences Computer Support
>>>>   > Brigham Young University
>>>>   > leblanc at byu.edu
>>>>   > (801)422-1882
>>>>   >
>>>>   >
>>>>   > _______________________________________________
>>>>   > Lustre-discuss mailing list
>>>>   > Lustre-discuss at clusterfs.com
>>>>   > https://mail.clusterfs.com/mailman/listinfo/lustre-discuss
>>>>   >  
>>>>   
>>>>   
>>>>    
>>>>  
>>>>  _______________________________________________
>>>>  Lustre-discuss mailing list
>>>>  Lustre-discuss at clusterfs.com
>>>>  https://mail.clusterfs.com/mailman/listinfo/lustre-discuss
>>>>   
>>>>  
>>>>   
>>>>  Mr Wojciech Turek
>>>>  Assistant System Manager
>>>>  University of Cambridge
>>>>  High Performance Computing service 
>>>>  email: wjt27 at cam.ac.uk
>>>>  tel. +441223763517
>>>>  
>>>>  
>>>>   
>>>>  
>>>>  
>>>>  
>>>> 
>>>>   
>>>>  Robert LeBlanc
>>>>  College of Life Sciences Computer Support
>>>>  Brigham Young University
>>>>  leblanc at byu.edu
>>>>  (801)422-1882
>>>>  
>>>>     
>>>> 
>>>>  
>>>> Mr Wojciech Turek
>>>> Assistant System Manager
>>>> University of Cambridge
>>>> High Performance Computing service 
>>>> email: wjt27 at cam.ac.uk
>>>> tel. +441223763517
>>>> 
>>>> 
>>>>  
>>>> 
>>>> 
>> 
>>  
>> Robert LeBlanc
>> College of Life Sciences Computer Support
>> Brigham Young University
>> leblanc at byu.edu
>> (801)422-1882
>> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20071112/e5f4da60/attachment.htm>