All of lore.kernel.org
 help / color / mirror / Atom feed
* Newstore Read ?
@ 2015-05-05 22:11 kernel neophyte
  2015-05-05 22:48 ` Mark Nelson
  2015-05-06  5:01 ` Haomai Wang
  0 siblings, 2 replies; 6+ messages in thread
From: kernel neophyte @ 2015-05-05 22:11 UTC (permalink / raw)
  To: ceph-devel@vger.kernel.org

Hi,

I am unable to get any reads out of Newstore. The log is filled with
assert failures.

> 2015-05-05 14:49:46.654113 7f4236387700 -1 os/newstore/NewStore.cc: In function 'virtual int NewStore::fiemap(coll_t, const ghobject_t&, uint64_t, size_t, ceph::bufferlist&)' thread 7f4236387700 time 2015-05-05 14:49:46.582991

os/newstore/NewStore.cc: 1371: FAILED assert(0)


 ceph version 0.93-1098-g66e555d (66e555d4a6e2fd21efd9da0d1ebd3feb024b041e)

 1: (ceph::__ceph_assert_fail(char const*, char const*, int, char
const*)+0x8b) [0xc9bdeb]

 2: ceph-osd() [0x9ea293]

 3: (ReplicatedPG::do_osd_ops(ReplicatedPG::OpContext*,
std::vector<OSDOp, std::allocator<OSDOp> >&)+0x708c) [0x8d8ccc]

 4: (ReplicatedPG::prepare_transaction(ReplicatedPG::OpContext*)+0x97)
[0x8e6427]

 5: (ReplicatedPG::execute_ctx(ReplicatedPG::OpContext*)+0x5e5) [0x8e6c85]

 6: (ReplicatedPG::do_op(std::tr1::shared_ptr<OpRequest>&)+0x2907) [0x8eac87]

 7: (ReplicatedPG::do_request(std::tr1::shared_ptr<OpRequest>&,
ThreadPool::TPHandle&)+0x66a) [0x877bea]

 8: (OSD::dequeue_op(boost::intrusive_ptr<PG>,
std::tr1::shared_ptr<OpRequest>, ThreadPool::TPHandle&)+0x3bd)
[0x6d94fd]

 9: (OSD::ShardedOpWQ::_process(unsigned int,
ceph::heartbeat_handle_d*)+0x331) [0x6d9a21]

 10: (ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x875)
[0xc77905]

 11: (ShardedThreadPool::WorkThreadSharded::entry()+0x10) [0xc79a20]

 12: (()+0x8182) [0x7f427fe7c182]

 13: (clone()+0x6d) [0x7f427de8e47d]

 NOTE: a copy of the executable, or `objdump -rdS <executable>` is
needed to interpret this.

Is anyone else seeing this ?

-Neo

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Newstore Read ?
  2015-05-05 22:11 Newstore Read ? kernel neophyte
@ 2015-05-05 22:48 ` Mark Nelson
  2015-05-06  0:55   ` Chen, Xiaoxi
  2015-05-06  5:01 ` Haomai Wang
  1 sibling, 1 reply; 6+ messages in thread
From: Mark Nelson @ 2015-05-05 22:48 UTC (permalink / raw)
  To: kernel neophyte, ceph-devel@vger.kernel.org

On 05/05/2015 05:11 PM, kernel neophyte wrote:
> Hi,
>
> I am unable to get any reads out of Newstore. The log is filled with
> assert failures.
>
>> 2015-05-05 14:49:46.654113 7f4236387700 -1 os/newstore/NewStore.cc: In function 'virtual int NewStore::fiemap(coll_t, const ghobject_t&, uint64_t, size_t, ceph::bufferlist&)' thread 7f4236387700 time 2015-05-05 14:49:46.582991
>
> os/newstore/NewStore.cc: 1371: FAILED assert(0)
>
>
>   ceph version 0.93-1098-g66e555d (66e555d4a6e2fd21efd9da0d1ebd3feb024b041e)
>
>   1: (ceph::__ceph_assert_fail(char const*, char const*, int, char
> const*)+0x8b) [0xc9bdeb]
>
>   2: ceph-osd() [0x9ea293]
>
>   3: (ReplicatedPG::do_osd_ops(ReplicatedPG::OpContext*,
> std::vector<OSDOp, std::allocator<OSDOp> >&)+0x708c) [0x8d8ccc]
>
>   4: (ReplicatedPG::prepare_transaction(ReplicatedPG::OpContext*)+0x97)
> [0x8e6427]
>
>   5: (ReplicatedPG::execute_ctx(ReplicatedPG::OpContext*)+0x5e5) [0x8e6c85]
>
>   6: (ReplicatedPG::do_op(std::tr1::shared_ptr<OpRequest>&)+0x2907) [0x8eac87]
>
>   7: (ReplicatedPG::do_request(std::tr1::shared_ptr<OpRequest>&,
> ThreadPool::TPHandle&)+0x66a) [0x877bea]
>
>   8: (OSD::dequeue_op(boost::intrusive_ptr<PG>,
> std::tr1::shared_ptr<OpRequest>, ThreadPool::TPHandle&)+0x3bd)
> [0x6d94fd]
>
>   9: (OSD::ShardedOpWQ::_process(unsigned int,
> ceph::heartbeat_handle_d*)+0x331) [0x6d9a21]
>
>   10: (ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x875)
> [0xc77905]
>
>   11: (ShardedThreadPool::WorkThreadSharded::entry()+0x10) [0xc79a20]
>
>   12: (()+0x8182) [0x7f427fe7c182]
>
>   13: (clone()+0x6d) [0x7f427de8e47d]
>
>   NOTE: a copy of the executable, or `objdump -rdS <executable>` is
> needed to interpret this.
>
> Is anyone else seeing this ?

I've mostly been focused on write performance.  Reads used to work, but 
it's possible something broke recently.  I'll try to reproduce.

>
> -Neo
> --
> 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] 6+ messages in thread

* RE: Newstore Read ?
  2015-05-05 22:48 ` Mark Nelson
@ 2015-05-06  0:55   ` Chen, Xiaoxi
  0 siblings, 0 replies; 6+ messages in thread
From: Chen, Xiaoxi @ 2015-05-06  0:55 UTC (permalink / raw)
  To: Mark Nelson, kernel neophyte, ceph-devel@vger.kernel.org

Hi Neo,

       The newstore::fie_map isn't implemented yet, so CEPH_OSD_OP_SPARSE_READ will get an assert(0). I will try to make it done today:)
								Xiaoxi

> -----Original Message-----
> From: ceph-devel-owner@vger.kernel.org [mailto:ceph-devel-
> owner@vger.kernel.org] On Behalf Of Mark Nelson
> Sent: Wednesday, May 6, 2015 6:49 AM
> To: kernel neophyte; ceph-devel@vger.kernel.org
> Subject: Re: Newstore Read ?
> 
> On 05/05/2015 05:11 PM, kernel neophyte wrote:
> > Hi,
> >
> > I am unable to get any reads out of Newstore. The log is filled with
> > assert failures.
> >
> >> 2015-05-05 14:49:46.654113 7f4236387700 -1 os/newstore/NewStore.cc:
> >> In function 'virtual int NewStore::fiemap(coll_t, const ghobject_t&,
> >> uint64_t, size_t, ceph::bufferlist&)' thread 7f4236387700 time
> >> 2015-05-05 14:49:46.582991
> >
> > os/newstore/NewStore.cc: 1371: FAILED assert(0)
> >
> >
> >   ceph version 0.93-1098-g66e555d
> > (66e555d4a6e2fd21efd9da0d1ebd3feb024b041e)
> >
> >   1: (ceph::__ceph_assert_fail(char const*, char const*, int, char
> > const*)+0x8b) [0xc9bdeb]
> >
> >   2: ceph-osd() [0x9ea293]
> >
> >   3: (ReplicatedPG::do_osd_ops(ReplicatedPG::OpContext*,
> > std::vector<OSDOp, std::allocator<OSDOp> >&)+0x708c) [0x8d8ccc]
> >
> >   4:
> > (ReplicatedPG::prepare_transaction(ReplicatedPG::OpContext*)+0x97)
> > [0x8e6427]
> >
> >   5: (ReplicatedPG::execute_ctx(ReplicatedPG::OpContext*)+0x5e5)
> > [0x8e6c85]
> >
> >   6: (ReplicatedPG::do_op(std::tr1::shared_ptr<OpRequest>&)+0x2907)
> > [0x8eac87]
> >
> >   7: (ReplicatedPG::do_request(std::tr1::shared_ptr<OpRequest>&,
> > ThreadPool::TPHandle&)+0x66a) [0x877bea]
> >
> >   8: (OSD::dequeue_op(boost::intrusive_ptr<PG>,
> > std::tr1::shared_ptr<OpRequest>, ThreadPool::TPHandle&)+0x3bd)
> > [0x6d94fd]
> >
> >   9: (OSD::ShardedOpWQ::_process(unsigned int,
> > ceph::heartbeat_handle_d*)+0x331) [0x6d9a21]
> >
> >   10: (ShardedThreadPool::shardedthreadpool_worker(unsigned
> > int)+0x875) [0xc77905]
> >
> >   11: (ShardedThreadPool::WorkThreadSharded::entry()+0x10) [0xc79a20]
> >
> >   12: (()+0x8182) [0x7f427fe7c182]
> >
> >   13: (clone()+0x6d) [0x7f427de8e47d]
> >
> >   NOTE: a copy of the executable, or `objdump -rdS <executable>` is
> > needed to interpret this.
> >
> > Is anyone else seeing this ?
> 
> I've mostly been focused on write performance.  Reads used to work, but it's
> possible something broke recently.  I'll try to reproduce.
> 
> >
> > -Neo
> > --
> > 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
> >
> --
> 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] 6+ messages in thread

* Re: Newstore Read ?
  2015-05-05 22:11 Newstore Read ? kernel neophyte
  2015-05-05 22:48 ` Mark Nelson
@ 2015-05-06  5:01 ` Haomai Wang
  2015-05-07  3:01   ` kernel neophyte
  1 sibling, 1 reply; 6+ messages in thread
From: Haomai Wang @ 2015-05-06  5:01 UTC (permalink / raw)
  To: kernel neophyte; +Cc: ceph-devel@vger.kernel.org

I think you may need to disable sparse op in client side. NewStore
doesn't support fiemap interface now.

On Wed, May 6, 2015 at 6:11 AM, kernel neophyte
<neophyte.hacker001@gmail.com> wrote:
> Hi,
>
> I am unable to get any reads out of Newstore. The log is filled with
> assert failures.
>
>> 2015-05-05 14:49:46.654113 7f4236387700 -1 os/newstore/NewStore.cc: In function 'virtual int NewStore::fiemap(coll_t, const ghobject_t&, uint64_t, size_t, ceph::bufferlist&)' thread 7f4236387700 time 2015-05-05 14:49:46.582991
>
> os/newstore/NewStore.cc: 1371: FAILED assert(0)
>
>
>  ceph version 0.93-1098-g66e555d (66e555d4a6e2fd21efd9da0d1ebd3feb024b041e)
>
>  1: (ceph::__ceph_assert_fail(char const*, char const*, int, char
> const*)+0x8b) [0xc9bdeb]
>
>  2: ceph-osd() [0x9ea293]
>
>  3: (ReplicatedPG::do_osd_ops(ReplicatedPG::OpContext*,
> std::vector<OSDOp, std::allocator<OSDOp> >&)+0x708c) [0x8d8ccc]
>
>  4: (ReplicatedPG::prepare_transaction(ReplicatedPG::OpContext*)+0x97)
> [0x8e6427]
>
>  5: (ReplicatedPG::execute_ctx(ReplicatedPG::OpContext*)+0x5e5) [0x8e6c85]
>
>  6: (ReplicatedPG::do_op(std::tr1::shared_ptr<OpRequest>&)+0x2907) [0x8eac87]
>
>  7: (ReplicatedPG::do_request(std::tr1::shared_ptr<OpRequest>&,
> ThreadPool::TPHandle&)+0x66a) [0x877bea]
>
>  8: (OSD::dequeue_op(boost::intrusive_ptr<PG>,
> std::tr1::shared_ptr<OpRequest>, ThreadPool::TPHandle&)+0x3bd)
> [0x6d94fd]
>
>  9: (OSD::ShardedOpWQ::_process(unsigned int,
> ceph::heartbeat_handle_d*)+0x331) [0x6d9a21]
>
>  10: (ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x875)
> [0xc77905]
>
>  11: (ShardedThreadPool::WorkThreadSharded::entry()+0x10) [0xc79a20]
>
>  12: (()+0x8182) [0x7f427fe7c182]
>
>  13: (clone()+0x6d) [0x7f427de8e47d]
>
>  NOTE: a copy of the executable, or `objdump -rdS <executable>` is
> needed to interpret this.
>
> Is anyone else seeing this ?
>
> -Neo
> --
> 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



-- 
Best Regards,

Wheat

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Newstore Read ?
  2015-05-06  5:01 ` Haomai Wang
@ 2015-05-07  3:01   ` kernel neophyte
  2015-05-07  4:08     ` Haomai Wang
  0 siblings, 1 reply; 6+ messages in thread
From: kernel neophyte @ 2015-05-07  3:01 UTC (permalink / raw)
  To: Haomai Wang; +Cc: ceph-devel@vger.kernel.org

On Tue, May 5, 2015 at 10:01 PM, Haomai Wang <haomaiwang@gmail.com> wrote:
> I think you may need to disable sparse op in client side. NewStore

Could you please explain more ?  I do not see any config option ?

-Neo


> doesn't support fiemap interface now.
>
> On Wed, May 6, 2015 at 6:11 AM, kernel neophyte
> <neophyte.hacker001@gmail.com> wrote:
>> Hi,
>>
>> I am unable to get any reads out of Newstore. The log is filled with
>> assert failures.
>>
>>> 2015-05-05 14:49:46.654113 7f4236387700 -1 os/newstore/NewStore.cc: In function 'virtual int NewStore::fiemap(coll_t, const ghobject_t&, uint64_t, size_t, ceph::bufferlist&)' thread 7f4236387700 time 2015-05-05 14:49:46.582991
>>
>> os/newstore/NewStore.cc: 1371: FAILED assert(0)
>>
>>
>>  ceph version 0.93-1098-g66e555d (66e555d4a6e2fd21efd9da0d1ebd3feb024b041e)
>>
>>  1: (ceph::__ceph_assert_fail(char const*, char const*, int, char
>> const*)+0x8b) [0xc9bdeb]
>>
>>  2: ceph-osd() [0x9ea293]
>>
>>  3: (ReplicatedPG::do_osd_ops(ReplicatedPG::OpContext*,
>> std::vector<OSDOp, std::allocator<OSDOp> >&)+0x708c) [0x8d8ccc]
>>
>>  4: (ReplicatedPG::prepare_transaction(ReplicatedPG::OpContext*)+0x97)
>> [0x8e6427]
>>
>>  5: (ReplicatedPG::execute_ctx(ReplicatedPG::OpContext*)+0x5e5) [0x8e6c85]
>>
>>  6: (ReplicatedPG::do_op(std::tr1::shared_ptr<OpRequest>&)+0x2907) [0x8eac87]
>>
>>  7: (ReplicatedPG::do_request(std::tr1::shared_ptr<OpRequest>&,
>> ThreadPool::TPHandle&)+0x66a) [0x877bea]
>>
>>  8: (OSD::dequeue_op(boost::intrusive_ptr<PG>,
>> std::tr1::shared_ptr<OpRequest>, ThreadPool::TPHandle&)+0x3bd)
>> [0x6d94fd]
>>
>>  9: (OSD::ShardedOpWQ::_process(unsigned int,
>> ceph::heartbeat_handle_d*)+0x331) [0x6d9a21]
>>
>>  10: (ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x875)
>> [0xc77905]
>>
>>  11: (ShardedThreadPool::WorkThreadSharded::entry()+0x10) [0xc79a20]
>>
>>  12: (()+0x8182) [0x7f427fe7c182]
>>
>>  13: (clone()+0x6d) [0x7f427de8e47d]
>>
>>  NOTE: a copy of the executable, or `objdump -rdS <executable>` is
>> needed to interpret this.
>>
>> Is anyone else seeing this ?
>>
>> -Neo
>> --
>> 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
>
>
>
> --
> Best Regards,
>
> Wheat

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: Newstore Read ?
  2015-05-07  3:01   ` kernel neophyte
@ 2015-05-07  4:08     ` Haomai Wang
  0 siblings, 0 replies; 6+ messages in thread
From: Haomai Wang @ 2015-05-07  4:08 UTC (permalink / raw)
  To: kernel neophyte; +Cc: ceph-devel@vger.kernel.org

Hmm, I looked at rbd configs that doesn't support disable sprase read.
So It looks like we can't do rbd read for NewStore now.

On Thu, May 7, 2015 at 11:01 AM, kernel neophyte
<neophyte.hacker001@gmail.com> wrote:
> On Tue, May 5, 2015 at 10:01 PM, Haomai Wang <haomaiwang@gmail.com> wrote:
>> I think you may need to disable sparse op in client side. NewStore
>
> Could you please explain more ?  I do not see any config option ?
>
> -Neo
>
>
>> doesn't support fiemap interface now.
>>
>> On Wed, May 6, 2015 at 6:11 AM, kernel neophyte
>> <neophyte.hacker001@gmail.com> wrote:
>>> Hi,
>>>
>>> I am unable to get any reads out of Newstore. The log is filled with
>>> assert failures.
>>>
>>>> 2015-05-05 14:49:46.654113 7f4236387700 -1 os/newstore/NewStore.cc: In function 'virtual int NewStore::fiemap(coll_t, const ghobject_t&, uint64_t, size_t, ceph::bufferlist&)' thread 7f4236387700 time 2015-05-05 14:49:46.582991
>>>
>>> os/newstore/NewStore.cc: 1371: FAILED assert(0)
>>>
>>>
>>>  ceph version 0.93-1098-g66e555d (66e555d4a6e2fd21efd9da0d1ebd3feb024b041e)
>>>
>>>  1: (ceph::__ceph_assert_fail(char const*, char const*, int, char
>>> const*)+0x8b) [0xc9bdeb]
>>>
>>>  2: ceph-osd() [0x9ea293]
>>>
>>>  3: (ReplicatedPG::do_osd_ops(ReplicatedPG::OpContext*,
>>> std::vector<OSDOp, std::allocator<OSDOp> >&)+0x708c) [0x8d8ccc]
>>>
>>>  4: (ReplicatedPG::prepare_transaction(ReplicatedPG::OpContext*)+0x97)
>>> [0x8e6427]
>>>
>>>  5: (ReplicatedPG::execute_ctx(ReplicatedPG::OpContext*)+0x5e5) [0x8e6c85]
>>>
>>>  6: (ReplicatedPG::do_op(std::tr1::shared_ptr<OpRequest>&)+0x2907) [0x8eac87]
>>>
>>>  7: (ReplicatedPG::do_request(std::tr1::shared_ptr<OpRequest>&,
>>> ThreadPool::TPHandle&)+0x66a) [0x877bea]
>>>
>>>  8: (OSD::dequeue_op(boost::intrusive_ptr<PG>,
>>> std::tr1::shared_ptr<OpRequest>, ThreadPool::TPHandle&)+0x3bd)
>>> [0x6d94fd]
>>>
>>>  9: (OSD::ShardedOpWQ::_process(unsigned int,
>>> ceph::heartbeat_handle_d*)+0x331) [0x6d9a21]
>>>
>>>  10: (ShardedThreadPool::shardedthreadpool_worker(unsigned int)+0x875)
>>> [0xc77905]
>>>
>>>  11: (ShardedThreadPool::WorkThreadSharded::entry()+0x10) [0xc79a20]
>>>
>>>  12: (()+0x8182) [0x7f427fe7c182]
>>>
>>>  13: (clone()+0x6d) [0x7f427de8e47d]
>>>
>>>  NOTE: a copy of the executable, or `objdump -rdS <executable>` is
>>> needed to interpret this.
>>>
>>> Is anyone else seeing this ?
>>>
>>> -Neo
>>> --
>>> 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
>>
>>
>>
>> --
>> Best Regards,
>>
>> Wheat



-- 
Best Regards,

Wheat

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2015-05-07  4:08 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-05-05 22:11 Newstore Read ? kernel neophyte
2015-05-05 22:48 ` Mark Nelson
2015-05-06  0:55   ` Chen, Xiaoxi
2015-05-06  5:01 ` Haomai Wang
2015-05-07  3:01   ` kernel neophyte
2015-05-07  4:08     ` Haomai Wang

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.