[Gluster-devel] can't make dht work

[Dan Parsons]

Yes, I put a "ls /glusterfs" in my mount script. I tried using the  
very latest fuse (2.7.4) and the problem was still there.


Dan Parsons


On Jan 13, 2009, at 11:57 PM, Anand Avati wrote:

> Dan,
> This is because in some cases fuse does not lookup the "/" of the
> filesystem before proceeding to access the entries. A simple
> workaround is to stat or ls the / as soon as you  mount. Most of the
> application access results in a lookup and everything works fine, but
> sometimes not so.
>
> avati
>
> 2009/1/14 Dan Parsons <dparsons at nyip.net>:
>> I think I have possibly found some more useful information on this  
>> issue.
>>
>> It appears that the problem only happens if my first write to dht  
>> is with
>> 'dd'. If I copy a normal file to the system first, everything works  
>> fine
>> from that point forward.
>>
>> However, if I unmount gluster and re-mount it, even though my  
>> original files
>> come back, I still have to copy a regular file again before dd will  
>> work.
>> Obviously dd is not important to my production use but it is sort  
>> of a
>> disconcerting bug, I have to say.
>>
>> Here's some log output:
>>
>> This happens when I try to use dd first:
>> 2009-01-13 10:45:54 E [dht-helper.c:216:dht_subvol_get_hashed]  
>> dht0: layout
>> missing path=/ddtest.txt parent=1
>> 2009-01-13 10:45:54 E [dht.c:623:dht_lookup] dht0: no subvolume in  
>> layout
>> for path=/ddtest.txt
>> 2009-01-13 10:45:54 E [fuse-bridge.c:406:fuse_entry_cbk] glusterfs- 
>> fuse: 2:
>> LOOKUP() /ddtest.txt => -1 (Invalid argument)
>>
>> This happens on the first copy of a regular file to gluster:
>> 2009-01-13 10:45:59 W [dht-layout.c:452:dht_layout_normalize] dht0:
>> directory / looked up first time
>> 2009-01-13 10:45:59 W [dht.c:132:dht_lookup_dir_cbk] dht0: fixing  
>> assignment
>> on /
>>
>> From that point onwards, everything works correctly, until I  
>> unmount and
>> remount, in which case I have to copy a regular file again.
>>
>>
>> Dan Parsons
>>
>>
>> On Jan 13, 2009, at 10:08 AM, Dan Parsons wrote:
>>
>>> I have tried that and get the same problem every time.
>>>
>>>
>>> Dan Parsons
>>>
>>>
>>> On Jan 13, 2009, at 3:37 AM, Krishna Srinivas wrote:
>>>
>>>> Dan,
>>>> Can you removie the backend directories and re-create it and try  
>>>> this
>>>> again?
>>>> Krishna
>>>>
>>>> On Tue, Jan 13, 2009 at 6:31 AM, Dan Parsons <dparsons at nyip.net>  
>>>> wrote:
>>>>>
>>>>> I'm unable to make dht work. I simply can't copy files to it  
>>>>> without
>>>>> getting
>>>>> the below error. If I make a subdirectory and put files inside  
>>>>> of that,
>>>>> then
>>>>> the problem goes away, but it seems like I must be doing something
>>>>> wrong.
>>>>> ???
>>>>>
>>>>> Client error message:
>>>>> 2009-01-12 16:54:03 E [dht-helper.c:216:dht_subvol_get_hashed]  
>>>>> dht0:
>>>>> layout
>>>>> missing path=/test3.txt parent=1
>>>>> 2009-01-12 16:54:03 E [dht.c:623:dht_lookup] dht0: no subvolume in
>>>>> layout
>>>>> for path=/test3.txt
>>>>> 2009-01-12 16:54:03 E [fuse-bridge.c:406:fuse_entry_cbk]  
>>>>> glusterfs-fuse:
>>>>> 2:
>>>>> LOOKUP() /test3.txt => -1 (Invalid argument)
>>>>>
>>>>> Command I ran:
>>>>> dd if=/dev/zero of=/glustertmp/test3.txt bs=1MB count=10
>>>>>
>>>>> Server config:
>>>>>
>>>>> ############################################################################
>>>>> # UNIFY
>>>>> volume posix-unify-brick
>>>>>    type storage/posix
>>>>>    option directory /distfs-storage-space/glusterfs/unify
>>>>> end-volume
>>>>>
>>>>> volume server
>>>>>    type protocol/server
>>>>>    option transport-type tcp
>>>>>    option auth.addr.posix-unify-brick.allow 10.8.101.*
>>>>>    subvolumes posix-unify-brick
>>>>> end-volume
>>>>>
>>>>> Client config:
>>>>> #################################################################
>>>>> # CLIENT-UNIFY-1
>>>>> volume distfs01-unify-brick
>>>>> type protocol/client
>>>>> option transport-type tcp
>>>>> option remote-host 10.8.101.51
>>>>> option remote-subvolume posix-unify-brick
>>>>> end-volume
>>>>>
>>>>> #################################################################
>>>>> # CLIENT-UNIFY-2
>>>>> volume distfs02-unify-brick
>>>>> type protocol/client
>>>>> option transport-type tcp
>>>>> option remote-host 10.8.101.52
>>>>> option remote-subvolume posix-unify-brick
>>>>> end-volume
>>>>>
>>>>> #################################################################
>>>>> # CLIENT-UNIFY-3
>>>>> volume distfs03-unify-brick
>>>>> type protocol/client
>>>>> option transport-type tcp
>>>>> option remote-host 10.8.101.53
>>>>> option remote-subvolume posix-unify-brick
>>>>> end-volume
>>>>>
>>>>> #################################################################
>>>>> # CLIENT-UNIFY-4
>>>>>
>>>>> volume distfs04-unify-brick
>>>>> type protocol/client
>>>>> option transport-type tcp
>>>>> option remote-host 10.8.101.54
>>>>> option remote-subvolume posix-unify-brick
>>>>> end-volume
>>>>>
>>>>> # END OF CLIENT-UNIFY
>>>>> ################################################################
>>>>>
>>>>> volume dht0
>>>>>    type cluster/dht
>>>>>    subvolumes distfs01-unify-brick distfs02-unify-brick
>>>>> distfs03-unify-brick distfs04-unify-brick
>>>>> end-volume
>>>>>
>>>>>
>>>>> Dan Parsons
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Gluster-devel mailing list
>>>>> Gluster-devel at nongnu.org
>>>>> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>>>>>
>>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Gluster-devel mailing list
>>> Gluster-devel at nongnu.org
>>> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>>>
>>
>>
>>
>> _______________________________________________
>> Gluster-devel mailing list
>> Gluster-devel at nongnu.org
>> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>>
>