* radosgw authentication problems
@ 2012-06-07 7:03 John Axel Eriksson
2012-06-07 7:14 ` Yehuda Sadeh
0 siblings, 1 reply; 8+ messages in thread
From: John Axel Eriksson @ 2012-06-07 7:03 UTC (permalink / raw)
To: ceph-devel
Hi I'm new to the list.
We've been looking at Ceph as a possible replacement for our current
distributed storage system. In particular
we're interested in the object storage, so I started researching the
radosgw. It did take me some time to get setup
and the docs/wiki is missing lots of information (not pointing any
fingers, just describing my experience).
Anyway, I can't seem to actually do anything with the radosgw using
any S3-client I've tried. The one I tried first
was Fog (used in Chef for example). Fog is the "The Ruby cloud
services library", available here https://github.com/fog/fog.
Fog has support for many different cloud service providers.
Anyway with Fog I get this in RadosGW logs:
##############################
2012-06-07 06:22:33.642827 7f638bfff700 2 req 4:0.000108:s3:GET /::getting op
2012-06-07 06:22:33.642831 7f638bfff700 2 req 4:0.000113:s3:GET
/:list_buckets:authorizing
2012-06-07 06:22:33.642854 7f638bfff700 20 get_obj_state:
rctx=0x7f637c0051e0 obj=.users:A state=0x7f637c005298
s->prefetch_data=0
2012-06-07 06:22:33.642863 7f638bfff700 10 moving .users+A to cache LRU end
2012-06-07 06:22:33.642866 7f638bfff700 10 cache get: name=.users+A : hit
2012-06-07 06:22:33.642872 7f638bfff700 20 get_obj_state: s->obj_tag
was set empty
2012-06-07 06:22:33.642877 7f638bfff700 10 moving .users+A to cache LRU end
2012-06-07 06:22:33.642879 7f638bfff700 10 cache get: name=.users+A : hit
2012-06-07 06:22:33.642920 7f638bfff700 10 get_canon_resource(): dest=/
2012-06-07 06:22:33.642923 7f638bfff700 10 auth_hdr:
GET
Thu, 07 Jun 2012 06:22:33 +0000
/
2012-06-07 06:22:33.642942 7f638bfff700 15 b64=zeK7YoQr63ajV6BRW6Z/UB1c4nM=
2012-06-07 06:22:33.642944 7f638bfff700 15
auth_sign=I9+abkuIjC89me95FoxmqApdrPs=
2012-06-07 06:22:33.642946 7f638bfff700 15 compare=-49
2012-06-07 06:22:33.642949 7f638bfff700 10 failed to authorize request
2012-06-07 06:22:33.642954 7f638bfff700 10 --> Status: 403
So for some reason it fails to authorize the request.
###############################
Then I moved to trying RightAws which is a different Cloud Services
Library in Ruby. Here I get a little further (because I first get a
200 response):
2012-06-07 06:24:46.006867 7fde21d66700 10 cache get: name=.users+A : miss
2012-06-07 06:24:46.008245 7fde21d66700 10 cache put: name=.users+A
2012-06-07 06:24:46.008266 7fde21d66700 10 adding .users+A to cache LRU end
2012-06-07 06:24:46.008276 7fde21d66700 20 get_obj_state: s->obj_tag
was set empty
2012-06-07 06:24:46.008283 7fde21d66700 10 moving .users+A to cache LRU end
2012-06-07 06:24:46.008291 7fde21d66700 10 cache get: name=.users+A :
type miss (requested=1, cached=6)
2012-06-07 06:24:46.008301 7fde21d66700 20 get_obj_state:
rctx=0x7fde040055b0 obj=.users:A state=0x7fde040068b8
s->prefetch_data=0
2012-06-07 06:24:46.008308 7fde21d66700 10 moving .users+A to cache LRU end
2012-06-07 06:24:46.008311 7fde21d66700 10 cache get: name=.users+A : hit
2012-06-07 06:24:46.008315 7fde21d66700 20 get_obj_state: s->obj_tag
was set empty
2012-06-07 06:24:46.008320 7fde21d66700 20 get_obj_state:
rctx=0x7fde040055b0 obj=.users:A state=0x7fde040068b8
s->prefetch_data=0
2012-06-07 06:24:46.008324 7fde21d66700 20 state for obj=.users:A is
not atomic, not appending atomic test
2012-06-07 06:24:46.008327 7fde21d66700 20 rados->read obj-ofs=0
read_ofs=0 read_len=16384
2012-06-07 06:24:46.008864 7fde21d66700 20 rados->read r=0 bl.length=126
2012-06-07 06:24:46.008898 7fde21d66700 10 cache put: name=.users+A
2012-06-07 06:24:46.008901 7fde21d66700 10 moving .users+A to cache LRU end
2012-06-07 06:24:46.008964 7fde21d66700 10 get_canon_resource(): dest=/
2012-06-07 06:24:46.008968 7fde21d66700 10 auth_hdr:
GET
Thu, 07 Jun 2012 06:24:46 GMT
/
2012-06-07 06:24:46.009059 7fde21d66700 15 b64=jpZmnXpUiWZq8IHBmsTKwKz4+Z4=
2012-06-07 06:24:46.009063 7fde21d66700 15
auth_sign=jpZmnXpUiWZq8IHBmsTKwKz4+Z4=
2012-06-07 06:24:46.009065 7fde21d66700 15 compare=0
2012-06-07 06:24:46.009069 7fde21d66700 2 req 1:0.002478:s3:GET
/:list_buckets:reading permissions
2012-06-07 06:24:46.009087 7fde21d66700 2 req 1:0.002496:s3:GET
/:list_buckets:verifying op permissions
2012-06-07 06:24:46.009099 7fde21d66700 2 req 1:0.002509:s3:GET
/:list_buckets:verifying op params
2012-06-07 06:24:46.009103 7fde21d66700 2 req 1:0.002513:s3:GET
/:list_buckets:executing
2012-06-07 06:24:46.009750 7fde21d66700 10 --> Status: 200
2012-06-07 06:24:46.009799 7fde21d66700 10 --> Content-Length: 318
2012-06-07 06:24:46.009803 7fde21d66700 10 --> Accept-Ranges: bytes
2012-06-07 06:24:46.009807 7fde21d66700 10 --> Content-type: application/xml
2012-06-07 06:24:46.009825 7fde21d66700 5 nothing to log for operation
2012-06-07 06:24:46.009832 7fde21d66700 2 req 1:0.003241:s3:GET
/:list_buckets:http status=200
And now I do a PUT:
2012-06-07 06:24:46.018316 7fde097fa700 2 req 2:0.000171:s3:PUT
/johnkey::getting op
2012-06-07 06:24:46.018322 7fde097fa700 2 req 2:0.000177:s3:PUT
/johnkey:create_bucket:authorizing
2012-06-07 06:24:46.018346 7fde097fa700 20 get_obj_state:
rctx=0x7fddfc0058a0 obj=.users:A state=0x7fddfc005958
s->prefetch_data=0
2012-06-07 06:24:46.018356 7fde097fa700 10 moving .users+A to cache LRU end
2012-06-07 06:24:46.018360 7fde097fa700 10 cache get: name=.users+A :
type miss (requested=6, cached=3)
2012-06-07 06:24:46.019922 7fde097fa700 10 cache put: name=.users+A
2012-06-07 06:24:46.019931 7fde097fa700 10 moving .users+A to cache LRU end
2012-06-07 06:24:46.019940 7fde097fa700 20 get_obj_state: s->obj_tag
was set empty
2012-06-07 06:24:46.019948 7fde097fa700 10 moving .users+A to cache LRU end
2012-06-07 06:24:46.019951 7fde097fa700 10 cache get: name=.users+A : hit
2012-06-07 06:24:46.019999 7fde097fa700 10 get_canon_resource(): dest=/johnkey
2012-06-07 06:24:46.020003 7fde097fa700 10 auth_hdr:
PUT
Thu, 07 Jun 2012 06:24:46 GMT
/johnkey
2012-06-07 06:24:46.020026 7fde097fa700 15 b64=RPIY36txOP127RgLYseHC8Ua9kc=
2012-06-07 06:24:46.020028 7fde097fa700 15
auth_sign=h020Ewm8NbfLYebWsXWkWCB6cDI=
2012-06-07 06:24:46.020030 7fde097fa700 15 compare=22
2012-06-07 06:24:46.020035 7fde097fa700 10 failed to authorize request
2012-06-07 06:24:46.020041 7fde097fa700 10 --> Status: 403
#################################
I've also tried s3cmd on the commandline which didn't work either. As
far as I've understood radosgw is supposed to be compatible with S3.
Thanks for your help!
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: radosgw authentication problems
2012-06-07 7:03 radosgw authentication problems John Axel Eriksson
@ 2012-06-07 7:14 ` Yehuda Sadeh
2012-06-07 10:16 ` John Axel Eriksson
0 siblings, 1 reply; 8+ messages in thread
From: Yehuda Sadeh @ 2012-06-07 7:14 UTC (permalink / raw)
To: John Axel Eriksson; +Cc: ceph-devel
On Thu, Jun 7, 2012 at 12:03 AM, John Axel Eriksson <john@insane.se> wrote:
> Hi I'm new to the list.
>
> We've been looking at Ceph as a possible replacement for our current
> distributed storage system. In particular
> we're interested in the object storage, so I started researching the
> radosgw. It did take me some time to get setup
> and the docs/wiki is missing lots of information (not pointing any
> fingers, just describing my experience).
>
> Anyway, I can't seem to actually do anything with the radosgw using
> any S3-client I've tried. The one I tried first
> was Fog (used in Chef for example). Fog is the "The Ruby cloud
> services library", available here https://github.com/fog/fog.
> Fog has support for many different cloud service providers.
>
> Anyway with Fog I get this in RadosGW logs:
>
> ##############################
>
> 2012-06-07 06:22:33.642827 7f638bfff700 2 req 4:0.000108:s3:GET /::getting op
> 2012-06-07 06:22:33.642831 7f638bfff700 2 req 4:0.000113:s3:GET
> /:list_buckets:authorizing
> 2012-06-07 06:22:33.642854 7f638bfff700 20 get_obj_state:
> rctx=0x7f637c0051e0 obj=.users:A state=0x7f637c005298
> s->prefetch_data=0
> 2012-06-07 06:22:33.642863 7f638bfff700 10 moving .users+A to cache LRU end
> 2012-06-07 06:22:33.642866 7f638bfff700 10 cache get: name=.users+A : hit
> 2012-06-07 06:22:33.642872 7f638bfff700 20 get_obj_state: s->obj_tag
> was set empty
> 2012-06-07 06:22:33.642877 7f638bfff700 10 moving .users+A to cache LRU end
> 2012-06-07 06:22:33.642879 7f638bfff700 10 cache get: name=.users+A : hit
> 2012-06-07 06:22:33.642920 7f638bfff700 10 get_canon_resource(): dest=/
> 2012-06-07 06:22:33.642923 7f638bfff700 10 auth_hdr:
> GET
>
>
> Thu, 07 Jun 2012 06:22:33 +0000
> /
> 2012-06-07 06:22:33.642942 7f638bfff700 15 b64=zeK7YoQr63ajV6BRW6Z/UB1c4nM=
> 2012-06-07 06:22:33.642944 7f638bfff700 15
> auth_sign=I9+abkuIjC89me95FoxmqApdrPs=
> 2012-06-07 06:22:33.642946 7f638bfff700 15 compare=-49
> 2012-06-07 06:22:33.642949 7f638bfff700 10 failed to authorize request
> 2012-06-07 06:22:33.642954 7f638bfff700 10 --> Status: 403
>
> So for some reason it fails to authorize the request.
>
> ###############################
>
> Then I moved to trying RightAws which is a different Cloud Services
> Library in Ruby. Here I get a little further (because I first get a
> 200 response):
>
> 2012-06-07 06:24:46.006867 7fde21d66700 10 cache get: name=.users+A : miss
> 2012-06-07 06:24:46.008245 7fde21d66700 10 cache put: name=.users+A
> 2012-06-07 06:24:46.008266 7fde21d66700 10 adding .users+A to cache LRU end
> 2012-06-07 06:24:46.008276 7fde21d66700 20 get_obj_state: s->obj_tag
> was set empty
> 2012-06-07 06:24:46.008283 7fde21d66700 10 moving .users+A to cache LRU end
> 2012-06-07 06:24:46.008291 7fde21d66700 10 cache get: name=.users+A :
> type miss (requested=1, cached=6)
> 2012-06-07 06:24:46.008301 7fde21d66700 20 get_obj_state:
> rctx=0x7fde040055b0 obj=.users:A state=0x7fde040068b8
> s->prefetch_data=0
> 2012-06-07 06:24:46.008308 7fde21d66700 10 moving .users+A to cache LRU end
> 2012-06-07 06:24:46.008311 7fde21d66700 10 cache get: name=.users+A : hit
> 2012-06-07 06:24:46.008315 7fde21d66700 20 get_obj_state: s->obj_tag
> was set empty
> 2012-06-07 06:24:46.008320 7fde21d66700 20 get_obj_state:
> rctx=0x7fde040055b0 obj=.users:A state=0x7fde040068b8
> s->prefetch_data=0
> 2012-06-07 06:24:46.008324 7fde21d66700 20 state for obj=.users:A is
> not atomic, not appending atomic test
> 2012-06-07 06:24:46.008327 7fde21d66700 20 rados->read obj-ofs=0
> read_ofs=0 read_len=16384
> 2012-06-07 06:24:46.008864 7fde21d66700 20 rados->read r=0 bl.length=126
> 2012-06-07 06:24:46.008898 7fde21d66700 10 cache put: name=.users+A
> 2012-06-07 06:24:46.008901 7fde21d66700 10 moving .users+A to cache LRU end
> 2012-06-07 06:24:46.008964 7fde21d66700 10 get_canon_resource(): dest=/
> 2012-06-07 06:24:46.008968 7fde21d66700 10 auth_hdr:
> GET
>
>
> Thu, 07 Jun 2012 06:24:46 GMT
> /
> 2012-06-07 06:24:46.009059 7fde21d66700 15 b64=jpZmnXpUiWZq8IHBmsTKwKz4+Z4=
> 2012-06-07 06:24:46.009063 7fde21d66700 15
> auth_sign=jpZmnXpUiWZq8IHBmsTKwKz4+Z4=
> 2012-06-07 06:24:46.009065 7fde21d66700 15 compare=0
> 2012-06-07 06:24:46.009069 7fde21d66700 2 req 1:0.002478:s3:GET
> /:list_buckets:reading permissions
> 2012-06-07 06:24:46.009087 7fde21d66700 2 req 1:0.002496:s3:GET
> /:list_buckets:verifying op permissions
> 2012-06-07 06:24:46.009099 7fde21d66700 2 req 1:0.002509:s3:GET
> /:list_buckets:verifying op params
> 2012-06-07 06:24:46.009103 7fde21d66700 2 req 1:0.002513:s3:GET
> /:list_buckets:executing
> 2012-06-07 06:24:46.009750 7fde21d66700 10 --> Status: 200
>
> 2012-06-07 06:24:46.009799 7fde21d66700 10 --> Content-Length: 318
>
> 2012-06-07 06:24:46.009803 7fde21d66700 10 --> Accept-Ranges: bytes
>
> 2012-06-07 06:24:46.009807 7fde21d66700 10 --> Content-type: application/xml
>
> 2012-06-07 06:24:46.009825 7fde21d66700 5 nothing to log for operation
> 2012-06-07 06:24:46.009832 7fde21d66700 2 req 1:0.003241:s3:GET
> /:list_buckets:http status=200
>
> And now I do a PUT:
>
> 2012-06-07 06:24:46.018316 7fde097fa700 2 req 2:0.000171:s3:PUT
> /johnkey::getting op
> 2012-06-07 06:24:46.018322 7fde097fa700 2 req 2:0.000177:s3:PUT
> /johnkey:create_bucket:authorizing
> 2012-06-07 06:24:46.018346 7fde097fa700 20 get_obj_state:
> rctx=0x7fddfc0058a0 obj=.users:A state=0x7fddfc005958
> s->prefetch_data=0
> 2012-06-07 06:24:46.018356 7fde097fa700 10 moving .users+A to cache LRU end
> 2012-06-07 06:24:46.018360 7fde097fa700 10 cache get: name=.users+A :
> type miss (requested=6, cached=3)
> 2012-06-07 06:24:46.019922 7fde097fa700 10 cache put: name=.users+A
> 2012-06-07 06:24:46.019931 7fde097fa700 10 moving .users+A to cache LRU end
> 2012-06-07 06:24:46.019940 7fde097fa700 20 get_obj_state: s->obj_tag
> was set empty
> 2012-06-07 06:24:46.019948 7fde097fa700 10 moving .users+A to cache LRU end
> 2012-06-07 06:24:46.019951 7fde097fa700 10 cache get: name=.users+A : hit
> 2012-06-07 06:24:46.019999 7fde097fa700 10 get_canon_resource(): dest=/johnkey
> 2012-06-07 06:24:46.020003 7fde097fa700 10 auth_hdr:
> PUT
>
>
> Thu, 07 Jun 2012 06:24:46 GMT
> /johnkey
> 2012-06-07 06:24:46.020026 7fde097fa700 15 b64=RPIY36txOP127RgLYseHC8Ua9kc=
> 2012-06-07 06:24:46.020028 7fde097fa700 15
> auth_sign=h020Ewm8NbfLYebWsXWkWCB6cDI=
> 2012-06-07 06:24:46.020030 7fde097fa700 15 compare=22
> 2012-06-07 06:24:46.020035 7fde097fa700 10 failed to authorize request
> 2012-06-07 06:24:46.020041 7fde097fa700 10 --> Status: 403
>
> #################################
>
> I've also tried s3cmd on the commandline which didn't work either. As
> far as I've understood radosgw is supposed to be compatible with S3.
>
It looks like the clients you're using are trying to access buckets
via the virtual host api: it preprends the bucket name to the service
host name. This requires wildcard dns, and also to have the 'rgw dns
name' config option correctly set in your ceph.conf.
Yehuda
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: radosgw authentication problems
2012-06-07 7:14 ` Yehuda Sadeh
@ 2012-06-07 10:16 ` John Axel Eriksson
2012-06-07 16:33 ` Tommi Virtanen
0 siblings, 1 reply; 8+ messages in thread
From: John Axel Eriksson @ 2012-06-07 10:16 UTC (permalink / raw)
To: Yehuda Sadeh; +Cc: ceph-devel
Thanks! That seems to have been the issue.
In general I really like ceph but it does seem like it has a few rough
edges. We're currently using another distributed object storage that
has been slightly unreliable and we've lost data twice now. That's why
we're looking at alternatives and ceph is at the top of my list.
What I understand is that cephs object storage is pretty stable and is
used in production by some. We're only interested in the object
storage/rados part so what's a more informed opinion on this than my
speculation?
Thanks again for your help!
John
On Thu, Jun 7, 2012 at 9:14 AM, Yehuda Sadeh <yehuda@inktank.com> wrote:
> On Thu, Jun 7, 2012 at 12:03 AM, John Axel Eriksson <john@insane.se> wrote:
>> Hi I'm new to the list.
>>
>> We've been looking at Ceph as a possible replacement for our current
>> distributed storage system. In particular
>> we're interested in the object storage, so I started researching the
>> radosgw. It did take me some time to get setup
>> and the docs/wiki is missing lots of information (not pointing any
>> fingers, just describing my experience).
>>
>> Anyway, I can't seem to actually do anything with the radosgw using
>> any S3-client I've tried. The one I tried first
>> was Fog (used in Chef for example). Fog is the "The Ruby cloud
>> services library", available here https://github.com/fog/fog.
>> Fog has support for many different cloud service providers.
>>
>> Anyway with Fog I get this in RadosGW logs:
>>
>> ##############################
>>
>> 2012-06-07 06:22:33.642827 7f638bfff700 2 req 4:0.000108:s3:GET /::getting op
>> 2012-06-07 06:22:33.642831 7f638bfff700 2 req 4:0.000113:s3:GET
>> /:list_buckets:authorizing
>> 2012-06-07 06:22:33.642854 7f638bfff700 20 get_obj_state:
>> rctx=0x7f637c0051e0 obj=.users:A state=0x7f637c005298
>> s->prefetch_data=0
>> 2012-06-07 06:22:33.642863 7f638bfff700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:22:33.642866 7f638bfff700 10 cache get: name=.users+A : hit
>> 2012-06-07 06:22:33.642872 7f638bfff700 20 get_obj_state: s->obj_tag
>> was set empty
>> 2012-06-07 06:22:33.642877 7f638bfff700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:22:33.642879 7f638bfff700 10 cache get: name=.users+A : hit
>> 2012-06-07 06:22:33.642920 7f638bfff700 10 get_canon_resource(): dest=/
>> 2012-06-07 06:22:33.642923 7f638bfff700 10 auth_hdr:
>> GET
>>
>>
>> Thu, 07 Jun 2012 06:22:33 +0000
>> /
>> 2012-06-07 06:22:33.642942 7f638bfff700 15 b64=zeK7YoQr63ajV6BRW6Z/UB1c4nM=
>> 2012-06-07 06:22:33.642944 7f638bfff700 15
>> auth_sign=I9+abkuIjC89me95FoxmqApdrPs=
>> 2012-06-07 06:22:33.642946 7f638bfff700 15 compare=-49
>> 2012-06-07 06:22:33.642949 7f638bfff700 10 failed to authorize request
>> 2012-06-07 06:22:33.642954 7f638bfff700 10 --> Status: 403
>>
>> So for some reason it fails to authorize the request.
>>
>> ###############################
>>
>> Then I moved to trying RightAws which is a different Cloud Services
>> Library in Ruby. Here I get a little further (because I first get a
>> 200 response):
>>
>> 2012-06-07 06:24:46.006867 7fde21d66700 10 cache get: name=.users+A : miss
>> 2012-06-07 06:24:46.008245 7fde21d66700 10 cache put: name=.users+A
>> 2012-06-07 06:24:46.008266 7fde21d66700 10 adding .users+A to cache LRU end
>> 2012-06-07 06:24:46.008276 7fde21d66700 20 get_obj_state: s->obj_tag
>> was set empty
>> 2012-06-07 06:24:46.008283 7fde21d66700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:24:46.008291 7fde21d66700 10 cache get: name=.users+A :
>> type miss (requested=1, cached=6)
>> 2012-06-07 06:24:46.008301 7fde21d66700 20 get_obj_state:
>> rctx=0x7fde040055b0 obj=.users:A state=0x7fde040068b8
>> s->prefetch_data=0
>> 2012-06-07 06:24:46.008308 7fde21d66700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:24:46.008311 7fde21d66700 10 cache get: name=.users+A : hit
>> 2012-06-07 06:24:46.008315 7fde21d66700 20 get_obj_state: s->obj_tag
>> was set empty
>> 2012-06-07 06:24:46.008320 7fde21d66700 20 get_obj_state:
>> rctx=0x7fde040055b0 obj=.users:A state=0x7fde040068b8
>> s->prefetch_data=0
>> 2012-06-07 06:24:46.008324 7fde21d66700 20 state for obj=.users:A is
>> not atomic, not appending atomic test
>> 2012-06-07 06:24:46.008327 7fde21d66700 20 rados->read obj-ofs=0
>> read_ofs=0 read_len=16384
>> 2012-06-07 06:24:46.008864 7fde21d66700 20 rados->read r=0 bl.length=126
>> 2012-06-07 06:24:46.008898 7fde21d66700 10 cache put: name=.users+A
>> 2012-06-07 06:24:46.008901 7fde21d66700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:24:46.008964 7fde21d66700 10 get_canon_resource(): dest=/
>> 2012-06-07 06:24:46.008968 7fde21d66700 10 auth_hdr:
>> GET
>>
>>
>> Thu, 07 Jun 2012 06:24:46 GMT
>> /
>> 2012-06-07 06:24:46.009059 7fde21d66700 15 b64=jpZmnXpUiWZq8IHBmsTKwKz4+Z4=
>> 2012-06-07 06:24:46.009063 7fde21d66700 15
>> auth_sign=jpZmnXpUiWZq8IHBmsTKwKz4+Z4=
>> 2012-06-07 06:24:46.009065 7fde21d66700 15 compare=0
>> 2012-06-07 06:24:46.009069 7fde21d66700 2 req 1:0.002478:s3:GET
>> /:list_buckets:reading permissions
>> 2012-06-07 06:24:46.009087 7fde21d66700 2 req 1:0.002496:s3:GET
>> /:list_buckets:verifying op permissions
>> 2012-06-07 06:24:46.009099 7fde21d66700 2 req 1:0.002509:s3:GET
>> /:list_buckets:verifying op params
>> 2012-06-07 06:24:46.009103 7fde21d66700 2 req 1:0.002513:s3:GET
>> /:list_buckets:executing
>> 2012-06-07 06:24:46.009750 7fde21d66700 10 --> Status: 200
>>
>> 2012-06-07 06:24:46.009799 7fde21d66700 10 --> Content-Length: 318
>>
>> 2012-06-07 06:24:46.009803 7fde21d66700 10 --> Accept-Ranges: bytes
>>
>> 2012-06-07 06:24:46.009807 7fde21d66700 10 --> Content-type: application/xml
>>
>> 2012-06-07 06:24:46.009825 7fde21d66700 5 nothing to log for operation
>> 2012-06-07 06:24:46.009832 7fde21d66700 2 req 1:0.003241:s3:GET
>> /:list_buckets:http status=200
>>
>> And now I do a PUT:
>>
>> 2012-06-07 06:24:46.018316 7fde097fa700 2 req 2:0.000171:s3:PUT
>> /johnkey::getting op
>> 2012-06-07 06:24:46.018322 7fde097fa700 2 req 2:0.000177:s3:PUT
>> /johnkey:create_bucket:authorizing
>> 2012-06-07 06:24:46.018346 7fde097fa700 20 get_obj_state:
>> rctx=0x7fddfc0058a0 obj=.users:A state=0x7fddfc005958
>> s->prefetch_data=0
>> 2012-06-07 06:24:46.018356 7fde097fa700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:24:46.018360 7fde097fa700 10 cache get: name=.users+A :
>> type miss (requested=6, cached=3)
>> 2012-06-07 06:24:46.019922 7fde097fa700 10 cache put: name=.users+A
>> 2012-06-07 06:24:46.019931 7fde097fa700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:24:46.019940 7fde097fa700 20 get_obj_state: s->obj_tag
>> was set empty
>> 2012-06-07 06:24:46.019948 7fde097fa700 10 moving .users+A to cache LRU end
>> 2012-06-07 06:24:46.019951 7fde097fa700 10 cache get: name=.users+A : hit
>> 2012-06-07 06:24:46.019999 7fde097fa700 10 get_canon_resource(): dest=/johnkey
>> 2012-06-07 06:24:46.020003 7fde097fa700 10 auth_hdr:
>> PUT
>>
>>
>> Thu, 07 Jun 2012 06:24:46 GMT
>> /johnkey
>> 2012-06-07 06:24:46.020026 7fde097fa700 15 b64=RPIY36txOP127RgLYseHC8Ua9kc=
>> 2012-06-07 06:24:46.020028 7fde097fa700 15
>> auth_sign=h020Ewm8NbfLYebWsXWkWCB6cDI=
>> 2012-06-07 06:24:46.020030 7fde097fa700 15 compare=22
>> 2012-06-07 06:24:46.020035 7fde097fa700 10 failed to authorize request
>> 2012-06-07 06:24:46.020041 7fde097fa700 10 --> Status: 403
>>
>> #################################
>>
>> I've also tried s3cmd on the commandline which didn't work either. As
>> far as I've understood radosgw is supposed to be compatible with S3.
>>
>
> It looks like the clients you're using are trying to access buckets
> via the virtual host api: it preprends the bucket name to the service
> host name. This requires wildcard dns, and also to have the 'rgw dns
> name' config option correctly set in your ceph.conf.
>
> Yehuda
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: radosgw authentication problems
2012-06-07 10:16 ` John Axel Eriksson
@ 2012-06-07 16:33 ` Tommi Virtanen
2012-06-07 18:52 ` John Axel Eriksson
2012-06-11 0:03 ` Barry Pederson
0 siblings, 2 replies; 8+ messages in thread
From: Tommi Virtanen @ 2012-06-07 16:33 UTC (permalink / raw)
To: John Axel Eriksson; +Cc: Yehuda Sadeh, ceph-devel
On Thu, Jun 7, 2012 at 3:16 AM, John Axel Eriksson <john@insane.se> wrote:
> In general I really like ceph but it does seem like it has a few rough
> edges.
We're putting a significant amount of effort in the documentation,
installability and manageability aspects of Ceph, and hope that more
of the fruit of that labor will become useful to the larger community
in the coming months.
> What I understand is that cephs object storage is pretty stable and is
> used in production by some. We're only interested in the object
> storage/rados part so what's a more informed opinion on this than my
> speculation?
RADOS, RBD and radosgw are considered production ready, when
configured correctly. Inktank is actively providing commercial support
for those components.
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: radosgw authentication problems
2012-06-07 16:33 ` Tommi Virtanen
@ 2012-06-07 18:52 ` John Axel Eriksson
2012-06-07 19:44 ` Tommi Virtanen
2012-06-11 0:03 ` Barry Pederson
1 sibling, 1 reply; 8+ messages in thread
From: John Axel Eriksson @ 2012-06-07 18:52 UTC (permalink / raw)
To: Tommi Virtanen; +Cc: Yehuda Sadeh, ceph-devel
Thanks Tommi!
Do you recommend btrfs or perhaps xfs for osds etc?
On Thu, Jun 7, 2012 at 6:33 PM, Tommi Virtanen <tv@inktank.com> wrote:
> On Thu, Jun 7, 2012 at 3:16 AM, John Axel Eriksson <john@insane.se> wrote:
>> In general I really like ceph but it does seem like it has a few rough
>> edges.
>
> We're putting a significant amount of effort in the documentation,
> installability and manageability aspects of Ceph, and hope that more
> of the fruit of that labor will become useful to the larger community
> in the coming months.
>
>> What I understand is that cephs object storage is pretty stable and is
>> used in production by some. We're only interested in the object
>> storage/rados part so what's a more informed opinion on this than my
>> speculation?
>
> RADOS, RBD and radosgw are considered production ready, when
> configured correctly. Inktank is actively providing commercial support
> for those components.
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: radosgw authentication problems
2012-06-07 18:52 ` John Axel Eriksson
@ 2012-06-07 19:44 ` Tommi Virtanen
2012-06-11 0:30 ` Mark Nelson
0 siblings, 1 reply; 8+ messages in thread
From: Tommi Virtanen @ 2012-06-07 19:44 UTC (permalink / raw)
To: John Axel Eriksson; +Cc: ceph-devel
On Thu, Jun 7, 2012 at 11:52 AM, John Axel Eriksson <john@insane.se> wrote:
> Do you recommend btrfs or perhaps xfs for osds etc?
Right now I think the preference order is xfs btrfs ext4, but we keep
seeing performance & reliability issues fairly evenly these days.
Btrfs apparently still has a tendency to get slower as the filesystem
ages.
Mark, feel free to correct me here ;)
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: radosgw authentication problems
2012-06-07 19:44 ` Tommi Virtanen
@ 2012-06-11 0:30 ` Mark Nelson
0 siblings, 0 replies; 8+ messages in thread
From: Mark Nelson @ 2012-06-11 0:30 UTC (permalink / raw)
To: Tommi Virtanen; +Cc: John Axel Eriksson, ceph-devel
On 06/07/2012 02:44 PM, Tommi Virtanen wrote:
> On Thu, Jun 7, 2012 at 11:52 AM, John Axel Eriksson<john@insane.se> wrote:
>> Do you recommend btrfs or perhaps xfs for osds etc?
>
> Right now I think the preference order is xfs btrfs ext4, but we keep
> seeing performance& reliability issues fairly evenly these days.
> Btrfs apparently still has a tendency to get slower as the filesystem
> ages.
>
> Mark, feel free to correct me here ;)
I think Tommi's got it right. So far what we've seen is that XFS is
slower on a fresh file system but over time ends up being faster for
small IO and about on-par for large IO. We just started testing EXT4
again now that we have leveldb in place. I'm a little concerned about
data integrity after ext4 failed on some benchmark runs, but on the
upside performance looks pretty good so far. We'll see how well it ages
vs xfs/btrfs in the coming weeks.
Mark
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: radosgw authentication problems
2012-06-07 16:33 ` Tommi Virtanen
2012-06-07 18:52 ` John Axel Eriksson
@ 2012-06-11 0:03 ` Barry Pederson
1 sibling, 0 replies; 8+ messages in thread
From: Barry Pederson @ 2012-06-11 0:03 UTC (permalink / raw)
To: Tommi Virtanen, ceph-devel
On Jun 7, 2012, at 11:33 AM, Tommi Virtanen wrote:
>
> RADOS, RBD and radosgw are considered production ready, when
> configured correctly. Inktank is actively providing commercial support
> for those components.
Are there any particular minimum versions of the Linux kernel or qemu-kvm that would be considered production-ready for RBD usage as clients? I was wondering if Ubuntu 12.04's 3.2.0-24 Kernel or qemu-kvm 1.0 would make the grade.
Barry
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2012-06-11 0:30 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-06-07 7:03 radosgw authentication problems John Axel Eriksson
2012-06-07 7:14 ` Yehuda Sadeh
2012-06-07 10:16 ` John Axel Eriksson
2012-06-07 16:33 ` Tommi Virtanen
2012-06-07 18:52 ` John Axel Eriksson
2012-06-07 19:44 ` Tommi Virtanen
2012-06-11 0:30 ` Mark Nelson
2012-06-11 0:03 ` Barry Pederson
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.