You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@couchdb.apache.org by Nikolai Teofilov <n....@gmail.com> on 2010/07/15 22:47:50 UTC
Replication hangs on streaming inline document attachments
The problem apear to exist onliy on replication 0.11 to 1.0
How to reproduce the bug:
On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
{
"_id" : "test",
"_attachments":
{
"foo.txt":
{
"content_type":"text\/plain",
"data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
}
}
}
replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
Cheers
Nikolai
Re: Replication hangs on streaming inline document attachments
Posted by Filipe David Manana <fd...@apache.org>.
On Fri, Jul 16, 2010 at 7:27 PM, Nikolai Teofilov <n....@gmail.com> wrote:
> Hi Filipe,
>
> Actually I swear my wife not to go online ;) in my vacation. I would love to test it but it will be not possible for me in the next two weeks.
I understand, I do have the same kind of requests from mine all the time :)
>
> The first what I will do after the vacation is to test your patch.
Great. It's committed in trunk and branch 1.0.x.
>
> Best Regards from Sofia
> Nikolai
Best regards from Lisbon to Bulgaria then, and enjoy your holidays.
cheers
>
> Sent from my iPhone
>
> On Jul 16, 2010, at 13:02, Filipe David Manana <fd...@apache.org> wrote:
>
>> On Fri, Jul 16, 2010 at 5:41 AM, Nikolai Teofilov <n....@gmail.com> wrote:
>>> I am sorry but I was not able to create the ticket yesterday and today I am on in vacantion ...
>>> No internet i next 15 days :(
>>
>> Hi Nikolai,
>> The ticket was created :)
>>
>> I think the solution in the meanwhile would be to upgrade your 0.11 to
>> 1.0 (or to 0.11.1 and do a push replication to 1.0). Another solution
>> would be for you to compile the branch 1.0.x (after the fix is
>> commited).
>>
>> cheers
>>
>>>
>>> Cheers
>>> Nikolai
>>>
>>>
>>> On 16.07.2010, at 02:03, Filipe David Manana wrote:
>>>
>>>> On Thu, Jul 15, 2010 at 10:52 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>>>> Filipe,
>>>>>
>>>>> Yes, it happens only on replication from 0.11 (windows port) to 1.0.0 CouchDBX
>>>>> 1.0->1.0 no problem!
>>>>>
>>>>> Shell i create a ticket?
>>>>
>>>> Yes, please.
>>>> I was able to reproduce the problem. It happens when, from a CouchDB
>>>> 1.0 server, we do a pull replication with the source being a URL
>>>> pointing to a Couch 0.11 and the target is a local DB (1.0).
>>>>
>>>> I have a temporary solution now, but will add it later.
>>>>
>>>> cheers
>>>>
>>>>>
>>>>> cheers
>>>>> Nikolai
>>>>>
>>>>> On 15.07.2010, at 23:43, Filipe David Manana wrote:
>>>>>
>>>>>> So, it only happens on push replication, from a 0.11 Couch to a 1.0.0 Couch?
>>>>>>
>>>>>> I think I know where the issue is. Expect a fix soon (if I'm able to
>>>>>> reproduce the issue).
>>>>>>
>>>>>> cheers
>>>>>>
>>>>>> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>>>>>>
>>>>>>> The problem apear to exist onliy on replication 0.11 to 1.0
>>>>>>>
>>>>>>> How to reproduce the bug:
>>>>>>>
>>>>>>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>>>>>>
>>>>>>>
>>>>>>> {
>>>>>>> "_id" : "test",
>>>>>>> "_attachments":
>>>>>>> {
>>>>>>> "foo.txt":
>>>>>>> {
>>>>>>> "content_type":"text\/plain",
>>>>>>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>>>>>>> }
>>>>>>> }
>>>>>>> }
>>>>>>>
>>>>>>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>>>>>>
>>>>>>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>>>
>>>>>>> Cheers
>>>>>>> Nikolai
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Filipe David Manana,
>>>>>> fdmanana@apache.org
>>>>>>
>>>>>> "Reasonable men adapt themselves to the world.
>>>>>> Unreasonable men adapt the world to themselves.
>>>>>> That's why all progress depends on unreasonable men."
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Filipe David Manana,
>>>> fdmanana@apache.org
>>>>
>>>> "Reasonable men adapt themselves to the world.
>>>> Unreasonable men adapt the world to themselves.
>>>> That's why all progress depends on unreasonable men."
>>>
>>>
>>
>>
>>
>> --
>> Filipe David Manana,
>> fdmanana@apache.org
>>
>> "Reasonable men adapt themselves to the world.
>> Unreasonable men adapt the world to themselves.
>> That's why all progress depends on unreasonable men."
>
--
Filipe David Manana,
fdmanana@apache.org
"Reasonable men adapt themselves to the world.
Unreasonable men adapt the world to themselves.
That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Nikolai Teofilov <n....@gmail.com>.
Hi Filipe,
Actually I swear my wife not to go online ;) in my vacation. I would love to test it but it will be not possible for me in the next two weeks.
The first what I will do after the vacation is to test your patch.
Best Regards from Sofia
Nikolai
Sent from my iPhone
On Jul 16, 2010, at 13:02, Filipe David Manana <fd...@apache.org> wrote:
> On Fri, Jul 16, 2010 at 5:41 AM, Nikolai Teofilov <n....@gmail.com> wrote:
>> I am sorry but I was not able to create the ticket yesterday and today I am on in vacantion ...
>> No internet i next 15 days :(
>
> Hi Nikolai,
> The ticket was created :)
>
> I think the solution in the meanwhile would be to upgrade your 0.11 to
> 1.0 (or to 0.11.1 and do a push replication to 1.0). Another solution
> would be for you to compile the branch 1.0.x (after the fix is
> commited).
>
> cheers
>
>>
>> Cheers
>> Nikolai
>>
>>
>> On 16.07.2010, at 02:03, Filipe David Manana wrote:
>>
>>> On Thu, Jul 15, 2010 at 10:52 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>>> Filipe,
>>>>
>>>> Yes, it happens only on replication from 0.11 (windows port) to 1.0.0 CouchDBX
>>>> 1.0->1.0 no problem!
>>>>
>>>> Shell i create a ticket?
>>>
>>> Yes, please.
>>> I was able to reproduce the problem. It happens when, from a CouchDB
>>> 1.0 server, we do a pull replication with the source being a URL
>>> pointing to a Couch 0.11 and the target is a local DB (1.0).
>>>
>>> I have a temporary solution now, but will add it later.
>>>
>>> cheers
>>>
>>>>
>>>> cheers
>>>> Nikolai
>>>>
>>>> On 15.07.2010, at 23:43, Filipe David Manana wrote:
>>>>
>>>>> So, it only happens on push replication, from a 0.11 Couch to a 1.0.0 Couch?
>>>>>
>>>>> I think I know where the issue is. Expect a fix soon (if I'm able to
>>>>> reproduce the issue).
>>>>>
>>>>> cheers
>>>>>
>>>>> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>>>>>
>>>>>> The problem apear to exist onliy on replication 0.11 to 1.0
>>>>>>
>>>>>> How to reproduce the bug:
>>>>>>
>>>>>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>>>>>
>>>>>>
>>>>>> {
>>>>>> "_id" : "test",
>>>>>> "_attachments":
>>>>>> {
>>>>>> "foo.txt":
>>>>>> {
>>>>>> "content_type":"text\/plain",
>>>>>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>>>>>> }
>>>>>> }
>>>>>> }
>>>>>>
>>>>>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>>>>>
>>>>>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>>
>>>>>> Cheers
>>>>>> Nikolai
>>>>>
>>>>>
>>>>> --
>>>>> Filipe David Manana,
>>>>> fdmanana@apache.org
>>>>>
>>>>> "Reasonable men adapt themselves to the world.
>>>>> Unreasonable men adapt the world to themselves.
>>>>> That's why all progress depends on unreasonable men."
>>>>
>>>>
>>>
>>>
>>>
>>> --
>>> Filipe David Manana,
>>> fdmanana@apache.org
>>>
>>> "Reasonable men adapt themselves to the world.
>>> Unreasonable men adapt the world to themselves.
>>> That's why all progress depends on unreasonable men."
>>
>>
>
>
>
> --
> Filipe David Manana,
> fdmanana@apache.org
>
> "Reasonable men adapt themselves to the world.
> Unreasonable men adapt the world to themselves.
> That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Filipe David Manana <fd...@apache.org>.
On Fri, Jul 16, 2010 at 5:41 AM, Nikolai Teofilov <n....@gmail.com> wrote:
> I am sorry but I was not able to create the ticket yesterday and today I am on in vacantion ...
> No internet i next 15 days :(
Hi Nikolai,
The ticket was created :)
I think the solution in the meanwhile would be to upgrade your 0.11 to
1.0 (or to 0.11.1 and do a push replication to 1.0). Another solution
would be for you to compile the branch 1.0.x (after the fix is
commited).
cheers
>
> Cheers
> Nikolai
>
>
> On 16.07.2010, at 02:03, Filipe David Manana wrote:
>
>> On Thu, Jul 15, 2010 at 10:52 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>> Filipe,
>>>
>>> Yes, it happens only on replication from 0.11 (windows port) to 1.0.0 CouchDBX
>>> 1.0->1.0 no problem!
>>>
>>> Shell i create a ticket?
>>
>> Yes, please.
>> I was able to reproduce the problem. It happens when, from a CouchDB
>> 1.0 server, we do a pull replication with the source being a URL
>> pointing to a Couch 0.11 and the target is a local DB (1.0).
>>
>> I have a temporary solution now, but will add it later.
>>
>> cheers
>>
>>>
>>> cheers
>>> Nikolai
>>>
>>> On 15.07.2010, at 23:43, Filipe David Manana wrote:
>>>
>>>> So, it only happens on push replication, from a 0.11 Couch to a 1.0.0 Couch?
>>>>
>>>> I think I know where the issue is. Expect a fix soon (if I'm able to
>>>> reproduce the issue).
>>>>
>>>> cheers
>>>>
>>>> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>>>>
>>>>> The problem apear to exist onliy on replication 0.11 to 1.0
>>>>>
>>>>> How to reproduce the bug:
>>>>>
>>>>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>>>>
>>>>>
>>>>> {
>>>>> "_id" : "test",
>>>>> "_attachments":
>>>>> {
>>>>> "foo.txt":
>>>>> {
>>>>> "content_type":"text\/plain",
>>>>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>>>>> }
>>>>> }
>>>>> }
>>>>>
>>>>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>>>>
>>>>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>>
>>>>> Cheers
>>>>> Nikolai
>>>>
>>>>
>>>> --
>>>> Filipe David Manana,
>>>> fdmanana@apache.org
>>>>
>>>> "Reasonable men adapt themselves to the world.
>>>> Unreasonable men adapt the world to themselves.
>>>> That's why all progress depends on unreasonable men."
>>>
>>>
>>
>>
>>
>> --
>> Filipe David Manana,
>> fdmanana@apache.org
>>
>> "Reasonable men adapt themselves to the world.
>> Unreasonable men adapt the world to themselves.
>> That's why all progress depends on unreasonable men."
>
>
--
Filipe David Manana,
fdmanana@apache.org
"Reasonable men adapt themselves to the world.
Unreasonable men adapt the world to themselves.
That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Nikolai Teofilov <n....@gmail.com>.
I am sorry but I was not able to create the ticket yesterday and today I am on in vacantion ...
No internet i next 15 days :(
Cheers
Nikolai
On 16.07.2010, at 02:03, Filipe David Manana wrote:
> On Thu, Jul 15, 2010 at 10:52 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>> Filipe,
>>
>> Yes, it happens only on replication from 0.11 (windows port) to 1.0.0 CouchDBX
>> 1.0->1.0 no problem!
>>
>> Shell i create a ticket?
>
> Yes, please.
> I was able to reproduce the problem. It happens when, from a CouchDB
> 1.0 server, we do a pull replication with the source being a URL
> pointing to a Couch 0.11 and the target is a local DB (1.0).
>
> I have a temporary solution now, but will add it later.
>
> cheers
>
>>
>> cheers
>> Nikolai
>>
>> On 15.07.2010, at 23:43, Filipe David Manana wrote:
>>
>>> So, it only happens on push replication, from a 0.11 Couch to a 1.0.0 Couch?
>>>
>>> I think I know where the issue is. Expect a fix soon (if I'm able to
>>> reproduce the issue).
>>>
>>> cheers
>>>
>>> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>>>
>>>> The problem apear to exist onliy on replication 0.11 to 1.0
>>>>
>>>> How to reproduce the bug:
>>>>
>>>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>>>
>>>>
>>>> {
>>>> "_id" : "test",
>>>> "_attachments":
>>>> {
>>>> "foo.txt":
>>>> {
>>>> "content_type":"text\/plain",
>>>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>>>> }
>>>> }
>>>> }
>>>>
>>>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>>>
>>>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>>
>>>> Cheers
>>>> Nikolai
>>>
>>>
>>> --
>>> Filipe David Manana,
>>> fdmanana@apache.org
>>>
>>> "Reasonable men adapt themselves to the world.
>>> Unreasonable men adapt the world to themselves.
>>> That's why all progress depends on unreasonable men."
>>
>>
>
>
>
> --
> Filipe David Manana,
> fdmanana@apache.org
>
> "Reasonable men adapt themselves to the world.
> Unreasonable men adapt the world to themselves.
> That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Filipe David Manana <fd...@apache.org>.
On Thu, Jul 15, 2010 at 10:52 PM, Nikolai Teofilov <n....@gmail.com> wrote:
> Filipe,
>
> Yes, it happens only on replication from 0.11 (windows port) to 1.0.0 CouchDBX
> 1.0->1.0 no problem!
>
> Shell i create a ticket?
Yes, please.
I was able to reproduce the problem. It happens when, from a CouchDB
1.0 server, we do a pull replication with the source being a URL
pointing to a Couch 0.11 and the target is a local DB (1.0).
I have a temporary solution now, but will add it later.
cheers
>
> cheers
> Nikolai
>
> On 15.07.2010, at 23:43, Filipe David Manana wrote:
>
>> So, it only happens on push replication, from a 0.11 Couch to a 1.0.0 Couch?
>>
>> I think I know where the issue is. Expect a fix soon (if I'm able to
>> reproduce the issue).
>>
>> cheers
>>
>> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>>
>>> The problem apear to exist onliy on replication 0.11 to 1.0
>>>
>>> How to reproduce the bug:
>>>
>>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>>
>>>
>>> {
>>> "_id" : "test",
>>> "_attachments":
>>> {
>>> "foo.txt":
>>> {
>>> "content_type":"text\/plain",
>>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>>> }
>>> }
>>> }
>>>
>>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>>
>>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>>
>>> Cheers
>>> Nikolai
>>
>>
>> --
>> Filipe David Manana,
>> fdmanana@apache.org
>>
>> "Reasonable men adapt themselves to the world.
>> Unreasonable men adapt the world to themselves.
>> That's why all progress depends on unreasonable men."
>
>
--
Filipe David Manana,
fdmanana@apache.org
"Reasonable men adapt themselves to the world.
Unreasonable men adapt the world to themselves.
That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Nikolai Teofilov <n....@gmail.com>.
Filipe,
Yes, it happens only on replication from 0.11 (windows port) to 1.0.0 CouchDBX
1.0->1.0 no problem!
Shell i create a ticket?
cheers
Nikolai
On 15.07.2010, at 23:43, Filipe David Manana wrote:
> So, it only happens on push replication, from a 0.11 Couch to a 1.0.0 Couch?
>
> I think I know where the issue is. Expect a fix soon (if I'm able to
> reproduce the issue).
>
> cheers
>
> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>>
>> The problem apear to exist onliy on replication 0.11 to 1.0
>>
>> How to reproduce the bug:
>>
>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>
>>
>> {
>> "_id" : "test",
>> "_attachments":
>> {
>> "foo.txt":
>> {
>> "content_type":"text\/plain",
>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>> }
>> }
>> }
>>
>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>
>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>
>> Cheers
>> Nikolai
>
>
> --
> Filipe David Manana,
> fdmanana@apache.org
>
> "Reasonable men adapt themselves to the world.
> Unreasonable men adapt the world to themselves.
> That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Filipe David Manana <fd...@apache.org>.
So, it only happens on push replication, from a 0.11 Couch to a 1.0.0 Couch?
I think I know where the issue is. Expect a fix soon (if I'm able to
reproduce the issue).
cheers
On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>
> The problem apear to exist onliy on replication 0.11 to 1.0
>
> How to reproduce the bug:
>
> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>
>
> {
> "_id" : "test",
> "_attachments":
> {
> "foo.txt":
> {
> "content_type":"text\/plain",
> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
> }
> }
> }
>
> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>
> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>
> Cheers
> Nikolai
--
Filipe David Manana,
fdmanana@apache.org
"Reasonable men adapt themselves to the world.
Unreasonable men adapt the world to themselves.
That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Nikolai Teofilov <n....@gmail.com>.
Fails as well.
I triggered the push and pull replications only from 1.0
1.0 -> 0.11 fails
0.11->1.0 fails
On 15.07.2010, at 23:56, Randall Leeds wrote:
> On Thu, Jul 15, 2010 at 14:55, Nikolai Teofilov <n....@gmail.com> wrote:
>> pull replication?
>
> POST /_replicate to the target server (rather than the source).
Re: Replication hangs on streaming inline document attachments
Posted by Randall Leeds <ra...@gmail.com>.
On Thu, Jul 15, 2010 at 14:55, Nikolai Teofilov <n....@gmail.com> wrote:
> pull replication?
POST /_replicate to the target server (rather than the source).
Re: Replication hangs on streaming inline document attachments
Posted by Nikolai Teofilov <n....@gmail.com>.
pull replication?
On 15.07.2010, at 23:52, Filipe David Manana wrote:
> In the meanwhile, can you try with a pull replication? Source would be
> your 0.11 couch and the target your 1.0 couch.
>
> cheers
>
> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>> The problem apear to exist onliy on replication 0.11 to 1.0
>>
>> How to reproduce the bug:
>>
>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>
>>
>> {
>> "_id" : "test",
>> "_attachments":
>> {
>> "foo.txt":
>> {
>> "content_type":"text\/plain",
>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>> }
>> }
>> }
>>
>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>
>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>
>> Cheers
>> Nikolai
>
>
>
> --
> Filipe David Manana,
> fdmanana@apache.org
>
> "Reasonable men adapt themselves to the world.
> Unreasonable men adapt the world to themselves.
> That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Nikolai Teofilov <n....@gmail.com>.
replication triggered from 0.11:
1.0 -> 0.11 works (pull)
0.11 -> 1.0 fails with (push):
1> [info] [<0.103.0>] starting new replication "f9533f7a30bef0a5ca398b501c6d0b70" at <0.157.0>
1> [info] [<0.157.0>] replication terminating due to target_error: <<"bad_content_type">>
1> [info] [<0.103.0>] 127.0.0.1 - - 'POST' /_replicate 500
triggered from 1.0
1.0 -> 0.11 fails (push) without logs
0.11->1.0 fails (pull) with
>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
Cheers
Nikolai
On 15.07.2010, at 23:52, Filipe David Manana wrote:
> In the meanwhile, can you try with a pull replication? Source would be
> your 0.11 couch and the target your 1.0 couch.
>
> cheers
>
> On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
>> The problem apear to exist onliy on replication 0.11 to 1.0
>>
>> How to reproduce the bug:
>>
>> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>>
>>
>> {
>> "_id" : "test",
>> "_attachments":
>> {
>> "foo.txt":
>> {
>> "content_type":"text\/plain",
>> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
>> }
>> }
>> }
>>
>> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>>
>> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
>> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>>
>> Cheers
>> Nikolai
>
>
>
> --
> Filipe David Manana,
> fdmanana@apache.org
>
> "Reasonable men adapt themselves to the world.
> Unreasonable men adapt the world to themselves.
> That's why all progress depends on unreasonable men."
Re: Replication hangs on streaming inline document attachments
Posted by Filipe David Manana <fd...@apache.org>.
In the meanwhile, can you try with a pull replication? Source would be
your 0.11 couch and the target your 1.0 couch.
cheers
On Thu, Jul 15, 2010 at 9:47 PM, Nikolai Teofilov <n....@gmail.com> wrote:
> The problem apear to exist onliy on replication 0.11 to 1.0
>
> How to reproduce the bug:
>
> On couchdb 0.11 create a database with a document that contains inline base64 encoded attachment.
>
>
> {
> "_id" : "test",
> "_attachments":
> {
> "foo.txt":
> {
> "content_type":"text\/plain",
> "data": "VGhpcyBpcyBhIGJhc2U2NCBlbmNvZGVkIHRleHQ="
> }
> }
> }
>
> replication 0.11 -> (CouchDBX) 1.0 will hang with following error log:
>
> 1> [info] [<0.1028.0>] starting new replication "5427f9ab10e93ee8b02bd9a3e0050c58" at <0.1069.0>
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225570,362980}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225571,40192}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225574,323477}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,24131}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225575,688384}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
> 1> [error] [<0.1083.0>] streaming att. ended but more data requested {1279,225576,382178}
> 1> [info] [<0.1083.0>] request for "test/foo.txt" timed out, retrying in 0.5 seconds
>
> Cheers
> Nikolai
--
Filipe David Manana,
fdmanana@apache.org
"Reasonable men adapt themselves to the world.
Unreasonable men adapt the world to themselves.
That's why all progress depends on unreasonable men."