manifoldcf-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Karl Wright <daddy...@gmail.com>
Subject Re: Documentum - unable to index metadata
Date Thu, 31 Mar 2016 05:44:58 GMT
Hi Radek,

A fix for the UI, at least, can be downloaded from the ticket
CONNECTORS-1293.  I can find no definitive mechanism for why this would
lead to no attributes being collected, but it's worth applying the patch,
updating your jobs, and giving it a try nonetheless.  Please let me know
what happens.

Thanks,
Karl


On Wed, Mar 30, 2016 at 5:21 PM, Karl Wright <daddywri@gmail.com> wrote:

> Hi Radek,
>
> The code that reads attribute values from Documentum DFC persistent
> objects does use the attribute name, as follows:
>
> >>>>>>
>   /** Get all the values that an attribute has, including multiple ones if
> present */
>   public String[] getAttributeValues(String attribute)
>     throws DocumentumException, RemoteException
>   {
>     try
>     {
>       int valueCount = object.getValueCount(attribute);
>       String[] values = new String[valueCount];
>       int y = 0;
>       while (y < valueCount)
>       {
>         // Fetch the attribute.
>         // It's supposed to work for all attribute types...
>         String value = object.getRepeatingString(attribute,y);
>         values[y++] = value;
>       }
>       return values;
>     }
>     catch (DfAuthenticationException ex)
>     {
>       throw new DocumentumException("Bad credentials:
> "+ex.getMessage(),DocumentumException.TYPE_BADCREDENTIALS);
>     }
>     catch (DfIdentityException ex)
>     {
>       throw new DocumentumException("Bad docbase name:
> "+ex.getMessage(),DocumentumException.TYPE_BADCONNECTIONPARAMS);
>     }
>     catch (DfDocbaseUnreachableException e)
>     {
>       throw new DocumentumException("Docbase unreachable:
> "+e.getMessage(),DocumentumException.TYPE_SERVICEINTERRUPTION);
>     }
>     catch (DfIOException e)
>     {
>       throw new DocumentumException("Docbase io exception:
> "+e.getMessage(),DocumentumException.TYPE_SERVICEINTERRUPTION);
>     }
>     catch (DfException e)
>     {
>       throw new DocumentumException("Documentum error: "+e.getMessage());
>     }
>   }
> <<<<<<
>
> This is how the DFC IDfPersistentObject API is structured.  So it doesn't
> look like multiple language values are supported in DFC.  So I don't know
> why you wouldn't get attribute values unless the UI issue is causing there
> to be no specified attributes for whatever type matches the document.  I'll
> have to dig into that code next.
>
> Karl
>
>
> On Wed, Mar 30, 2016 at 9:58 AM, Karl Wright <daddywri@gmail.com> wrote:
>
>> Hi Radek,
>>
>> I will have to check how the connector uses attribute names and get back
>> to you.  But I am pretty certain that the connector specifies attributes in
>> its dql queries by means of the attribute name, not the r_object_id.  If
>> that's the problem, it also implies that there can be a different attribute
>> value for each language, which might be why you aren't seeing the
>> attributes you are expecting.
>>
>> This is not an easy problem to address, however.
>>
>> Can you confirm whether or not documents can have different attribute
>> values for each language in Documentum?
>>
>> Thanks,
>> Karl
>>
>>
>>
>>
>> On Wed, Mar 30, 2016 at 9:41 AM, Radek Sklenicka <
>> radek.sklenicka@gmail.com> wrote:
>>
>>> Hi Karl,
>>>
>>>
>>>
>>> We discovered that we get metadata names in triplicate because there are
>>> 3 languages installed in Documentum.
>>>
>>> Multiple attribute records have each the same attr_name and type_name
>>> but unique r_object_id and different nls_key (en, es, pt).
>>>
>>>
>>>
>>> Could this be the reason why metadata doesn’t make it through the
>>> pipeline and we can’t get any metadata during crawling?
>>>
>>> Are unique attr_names required in Documentum connector?
>>>
>>>
>>>
>>> Any suggestions would be greatly appreciated.
>>>
>>>
>>>
>>> Thank you,
>>>
>>>
>>> Radek
>>>
>>> On 23 March 2016 at 18:28, Radek Sklenicka <radek.sklenicka@gmail.com>
>>> wrote:
>>>
>>>> Thanks for verification, Karl.
>>>>
>>>> -Radek
>>>>
>>>> On 23 March 2016 at 14:01, Karl Wright <daddywri@gmail.com> wrote:
>>>>
>>>>> Hi Radek,
>>>>>
>>>>> This log output comes from RMI, apparently, and is not something I've
>>>>> ever seen before.  But it does look like it's a complete list of what's
>>>>> being returned for a request for the list of attributes (the first entry),
>>>>> and for a specific object (the second entry).
>>>>>
>>>>> Karl
>>>>>
>>>>> On Wed, Mar 23, 2016 at 8:41 AM, Radek Sklenicka <
>>>>> radek.sklenicka@gmail.com> wrote:
>>>>>
>>>>>> Hi Karl,
>>>>>>
>>>>>> "select attr_name FROM dmi_dd_attr_info" really returns duplicates
-
>>>>>> we're looking into that.
>>>>>>
>>>>>> Is there also a DQL query (or function) used by ManifoldCF that we
>>>>>> can try to check what/if attributes are being returned for a particular
>>>>>> record?
>>>>>>
>>>>>> We have trace logs from DFC and it looks like the attributes are
>>>>>> being returned from the content server.
>>>>>> Could you please help us decode the logs - where to look/verify if
>>>>>> attributes are handed over to ManifoldCF?
>>>>>> Can we deduce from the logs attached below that the attributes are
>>>>>> transferred from DFC to ManifoldCF?
>>>>>>
>>>>>> Many thanks,
>>>>>> Radek
>>>>>>
>>>>>>
>>>>>> 2016-03-22 13:29:26.008 <USER_DTESTER|s9(21.0)|SM@14660772>
 [RMI
>>>>>> TCP Connection(1823)-127.0.0.1] [EXIT]
>>>>>>  .com.documentum.fc.client.DfTypedObject@36b9ba.getLiteType ==>
>>>>>> AspectedLiteType@110eb5e{name=do_domep_project_hse, typeVersion=0,
>>>>>> cacheVStamp=178498, attributes={asp_herencia.atr_isnew,
>>>>>> asp_herencia.atr_niveles, asp_herencia.atr_tipo, asp_herencia.i_partition},
>>>>>> superType=LiteType@2045f2{name=do_domep_project_hse, typeVersion=32,
>>>>>> cacheVStamp=178498, attributes={atr_audit_type, atr_speciality,
>>>>>> atr_emergency_related}, superType=LiteType@a67471{name=do_domep_project,
>>>>>> typeVersion=32, cacheVStamp=178486, attributes={atr_uwi, atr_well_name,
>>>>>> atr_usi, atr_survey_name}, superType=LiteType@f74077{name=do_domep_base,
>>>>>> typeVersion=27, cacheVStamp=178438, attributes={atr_confidential_level,
>>>>>> atr_owner_area, atr_logical_code, atr_original_reference_id, atr_revision,
>>>>>> atr_entity, atr_author, atr_doc_type, atr_category_doc, atr_subcat_doc,
>>>>>> atr_discipline, atr_subdiscipline, atr_language, atr_physical_document,
>>>>>> atr_physical_code, atr_warehouse, atr_retention, atr_digital_media,
>>>>>> atr_internal, atr_country, atr_basin, atr_environment, atr_acreage,
>>>>>> atr_abstract, atr_doc_creation_date, atr_title, atr_collection,
>>>>>> atr_is_collection, atr_is_principal, atr_is_anexo, atr_id_collection,
>>>>>> atr_is_relation, atr_field, atr_original_revision, atr_remarks,
>>>>>> atr_keywords, atr_principal_folder_id, atr_original_version, atr_be_name,
>>>>>> atr_be_ref, atr_be_short_name, atr_issued_for_code,
>>>>>> atr_issued_for_description, atr_subbasin, atr_be_type_id, atr_comment,
>>>>>> atr_status, atr_prepared_by, atr_preparation_date, atr_verified_by,
>>>>>> atr_verification_date, atr_approved_by, atr_approval_date, atr_workflow},
>>>>>> superType=LiteType@19390bd{name=do_general, typeVersion=6,
>>>>>> cacheVStamp=167968, attributes={negocio, attr_is_gdcom},
>>>>>> superType=LiteType@16658e8{name=dm_document, typeVersion=2,
>>>>>> cacheVStamp=52034, attributes={}, superType=LiteType@6aac49{name=dm_sysobject,
>>>>>> typeVersion=3, cacheVStamp=0, attributes={object_name, r_object_type,
>>>>>> title, subject, authors, keywords, a_application_type, a_status,
>>>>>> r_creation_date, r_modify_date, r_modifier, r_access_date, a_is_hidden,
>>>>>> i_is_deleted, a_retention_date, a_archive, a_compound_architecture,
>>>>>> a_link_resolved, i_reference_cnt, i_has_folder, i_folder_id,
>>>>>> r_composite_id, r_composite_label, r_component_label, r_order_no,
>>>>>> r_link_cnt, r_link_high_cnt, r_assembled_from_id, r_frzn_assembly_cnt,
>>>>>> r_has_frzn_assembly, resolution_label, r_is_virtual_doc, i_contents_id,
>>>>>> a_content_type, r_page_cnt, r_content_size, a_full_text, a_storage_type,
>>>>>> i_cabinet_id, owner_name, owner_permit, group_name, group_permit,
>>>>>> world_permit, i_antecedent_id, i_chronicle_id, i_latest_flag, r_lock_owner,
>>>>>> r_lock_date, r_lock_machine, log_entry, r_version_label, i_branch_cnt,
>>>>>> i_direct_dsc, r_immutable_flag, r_frozen_flag, r_has_events, acl_domain,
>>>>>> acl_name, a_special_app, i_is_reference, r_creator_name, r_is_public,
>>>>>> r_policy_id, r_resume_state, r_current_state, r_alias_set_id,
>>>>>> a_effective_date, a_expiration_date, a_publish_formats, a_effective_label,
>>>>>> a_effective_flag, a_category, language_code, a_is_template,
>>>>>> a_controlling_app, r_full_content_size, a_extended_properties, a_is_signed,
>>>>>> a_last_review_date, i_retain_until, r_aspect_name, i_retainer_id,
>>>>>> i_partition, i_is_replica, i_vstamp}}}}}}}}
>>>>>>
>>>>>>
>>>>>>
>>>>>> 2016-03-22 13:29:26.010 <USER_DTESTER|s5(17.0)|SM@16366401>
 [RMI
>>>>>> TCP Connection(1820)-127.0.0.1] [RPC_EXIT]  ......RPC: applyForObject
==>
>>>>>> TypedData@144302b[id=098c1b38809921b1,
>>>>>> type=do_domep_project_well_wd, readOnly=false, autoFill=true,
>>>>>> fetchTimestamp=0, values=[object_name=DSC00683.JPG,
>>>>>> r_object_type=do_domep_project_well_wd, title=, subject=, authors=[],
>>>>>> keywords=[], a_application_type=, a_status=, r_creation_date=2/11/2016
>>>>>> 8:35:45 AM, r_modify_date=3/7/2016 11:17:37 AM, r_modifier=admdcmt,
>>>>>> r_access_date=3/16/2016 12:23:59 PM, a_is_hidden=F, i_is_deleted=F,
>>>>>> a_retention_date=nulldate, a_archive=F, a_compound_architecture=,
>>>>>> a_link_resolved=F, i_reference_cnt=1, i_has_folder=T,
>>>>>> i_folder_id=[0b8c1b3880991ad3], r_composite_id=[], r_composite_label=[],
>>>>>> r_component_label=[], r_order_no=[], r_link_cnt=0, r_link_high_cnt=0,
>>>>>> r_assembled_from_id=0000000000000000, r_frzn_assembly_cnt=0,
>>>>>> r_has_frzn_assembly=F, resolution_label=, r_is_virtual_doc=0,
>>>>>> i_contents_id=068c1b388064051a, a_content_type=jpeg, r_page_cnt=1,
>>>>>> r_content_size=949228, a_full_text=T, a_storage_type=repo,
>>>>>> i_cabinet_id=0c8c1b38806aee30, owner_name=Domep controlador 00010,
>>>>>> owner_permit=7, group_name=, group_permit=1, world_permit=1,
>>>>>> i_antecedent_id=0000000000000000, i_chronicle_id=098c1b38809921b1,
>>>>>> i_latest_flag=T, r_lock_owner=, r_lock_date=nulldate, r_lock_machine=,
>>>>>> log_entry=, r_version_label=[1.0, CURRENT], i_branch_cnt=0, i_direct_dsc=F,
>>>>>> r_immutable_flag=F, r_frozen_flag=F, r_has_events=F, acl_domain=admdocum,
>>>>>> acl_name=domep_ac_en_02080, a_special_app=, i_is_reference=F,
>>>>>> r_creator_name=Domep controlador 00010, r_is_public=F,
>>>>>> r_policy_id=468c1b38809c6e47, r_resume_state=-1, r_current_state=0,
>>>>>> r_alias_set_id=0000000000000000, a_effective_date=[], a_expiration_date=[],
>>>>>> a_publish_formats=[], a_effective_label=[], a_effective_flag=[],
>>>>>> a_category=, language_code=, a_is_template=F, a_controlling_app=,
>>>>>> r_full_content_size=949228, a_extended_properties=[], a_is_signed=F,
>>>>>> a_last_review_date=nulldate, i_retain_until=nulldate,
>>>>>> r_aspect_name=[asp_herencia], i_retainer_id=[], i_partition=0,
>>>>>> i_is_replica=F, i_vstamp=4, negocio=E&P, attr_is_gdcom=F,
>>>>>> atr_confidential_level=Internal Use, atr_owner_area=OFICINA DE E,
>>>>>> atr_logical_code=IQEXPEOMKUR000WEL2016000063, atr_original_reference_id=[],
>>>>>> atr_revision=, atr_entity=[], atr_author=[Domep controlador 00010],
>>>>>> atr_doc_type=Reporting, atr_category_doc=Geology, atr_subcat_doc=Progress
>>>>>> Report, atr_discipline=GEOLOGY, atr_subdiscipline=[],
>>>>>> atr_language=[ENGLISH], atr_physical_document=F, atr_physical_code=[],
>>>>>> atr_warehouse=, atr_retention=YES, atr_digital_media=F, atr_internal=YES,
>>>>>> atr_country=[XYZ], atr_basin=[XYZZ], atr_environment=, atr_acreage=[],
>>>>>> atr_abstract=, atr_doc_creation_date=5/22/2006 8:31:43 AM, atr_title=WELL
>>>>>> BARAM 1 PHOTOS FIELD 06, atr_collection=[], atr_is_collection=F,
>>>>>> atr_is_principal=F, atr_is_anexo=F, atr_id_collection=[],
>>>>>> atr_is_relation=F, atr_field=[], atr_original_revision=, atr_remarks=,
>>>>>> atr_keywords=[], atr_principal_folder_id=0b8c1b3880991ad3,
>>>>>> atr_original_version=, atr_be_name=BARAM 1, atr_be_ref=IQWEL000008,
>>>>>> atr_be_short_name=BA 1, atr_issued_for_code=, atr_issued_for_description=,
>>>>>> atr_subbasin=[], atr_be_type_id=8, atr_comment=, atr_status=Draft,
>>>>>> atr_prepared_by=[], atr_preparation_date=nulldate, atr_verified_by=[],
>>>>>> atr_verification_date=nulldate, atr_approved_by=[],
>>>>>> atr_approval_date=nulldate, atr_workflow=, atr_well_name=BARAM 1,
>>>>>> atr_uwi=IQ010004432, atr_borehole_name=[BARAM 1], atr_ubhi=[IQ01000443200],
>>>>>> atr_borehole_alias=[], atr_borehole_short_name=[], atr_sample_type=[],
>>>>>> atr_analysis_type=[], asp_herencia.atr_isnew=F, asp_herencia.atr_niveles=0,
>>>>>> asp_herencia.atr_tipo=[], asp_herencia.i_partition=0,
>>>>>> r_object_id=098c1b38809921b1, _KEEP_LOCK_=F, _FREEZE_COMPONENTS_=F,
>>>>>> _THAW_COMPONENTS_=F, _CONTENTS_CHANGED_=F, _DIST_SAVE_AS_NEW_=F]]
>>>>>>
>>>>>>
>>>>>>
>>>>>> On 11 March 2016 at 15:33, Radek Sklenicka <radek.sklenicka@gmail.com
>>>>>> > wrote:
>>>>>>
>>>>>>> Thanks Karl, we'll verify that.
>>>>>>>
>>>>>>> -Radek
>>>>>>>
>>>>>>> On 11 March 2016 at 14:21, Karl Wright <daddywri@gmail.com>
wrote:
>>>>>>>
>>>>>>>> Hi Radek,
>>>>>>>>
>>>>>>>> This is the DQL query that is run:
>>>>>>>>
>>>>>>>>       String strDQL = "select attr_name FROM dmi_dd_attr_info
where
>>>>>>>> type_name = '" + docType + "' order by attr_name asc";
>>>>>>>>
>>>>>>>> Karl
>>>>>>>>
>>>>>>>>
>>>>>>>> On Fri, Mar 11, 2016 at 8:19 AM, Karl Wright <daddywri@gmail.com>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> Hi Radek,
>>>>>>>>>
>>>>>>>>> The Document Types page runs a DQL query to populate
the document
>>>>>>>>> types.  The fact that you get duplicates means that something
may be
>>>>>>>>> corrupt with your Document instance.  It's possible that
for some reason
>>>>>>>>> the instance is set up with multiple records that each
have the same name
>>>>>>>>> but different key values.
>>>>>>>>>
>>>>>>>>> Documentum used to have a little web app that allowed
you to
>>>>>>>>> execute DQL queries.  I'd experiment to see what was
leading to the
>>>>>>>>> duplication.  The fact that you can't get any metadata
during crawling is
>>>>>>>>> almost certainly related.
>>>>>>>>>
>>>>>>>>> Thanks,
>>>>>>>>> Karl
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Fri, Mar 11, 2016 at 8:10 AM, Radek Sklenicka <
>>>>>>>>> radek.sklenicka@gmail.com> wrote:
>>>>>>>>>
>>>>>>>>>> Hello,
>>>>>>>>>>
>>>>>>>>>> We are not able to pull metadata from one of our
Documentum
>>>>>>>>>> instances (it is 6.7)
>>>>>>>>>> Interestingly, on the Job > Document Types page
each metadata
>>>>>>>>>> field is displayed 3 times in the metadata boxes
- could this be an issue?
>>>>>>>>>> Screenshots:
>>>>>>>>>> http://take.ms/mJhPh
>>>>>>>>>> http://take.ms/AMZF0
>>>>>>>>>> We have quite a long list of document types and it
takes minutes
>>>>>>>>>> to load the Document Types page.
>>>>>>>>>>
>>>>>>>>>> Also, we can successfully pull metadata from our
testing
>>>>>>>>>> Documentum (it is 7.1), and I noticed that there
is a difference in
>>>>>>>>>> connector logs between the two:
>>>>>>>>>>
>>>>>>>>>> 1.) here we are able to pull metadata:
>>>>>>>>>>
>>>>>>>>>> DEBUG 2016-03-10 03:50:08,051 (Worker thread '3')
- DCTM:
>>>>>>>>>> Document 090007c28000569d has version label:
>>>>>>>>>> 11+authors+object_name+owner_name+owner_permit+r_creation_date+r_creator_name+r_modifier+r_modify_date+r_object_id+r_object_type+title++0+DEAD_AUTHORITY+1.0_0_
>>>>>>>>>> http://localhost/webtop/
>>>>>>>>>> DEBUG 2016-03-10 03:50:08,052 (Worker thread '3')
- DCTM: Inside
>>>>>>>>>> processDocuments
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> 2.) NOT able to pull metadata:
>>>>>>>>>>
>>>>>>>>>> DEBUG 2016-03-10 14:58:22,908 (Worker thread '22')
- DCTM:
>>>>>>>>>> Document 098c1b3880991f48 has version label: 0++0+DEAD_AUTHORITY+_4_
>>>>>>>>>> http://localhost/webtop
>>>>>>>>>> DEBUG 2016-03-10 14:58:22,908 (Worker thread '22')
- DCTM: Inside
>>>>>>>>>> processDocuments
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Any ideas will be appreciated.
>>>>>>>>>>
>>>>>>>>>> Thank you,
>>>>>>>>>>
>>>>>>>>>> Radek
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>

Mime
View raw message