ERROR: rcModAccessControl failure status = -809000 CATALOG_ALREADY_HAS_ITEM_BY_THAT_NAME
Shaofeng Liu
When I run ichmod -r read username coll_name,
I got this error. any input?
Best,
Leo
Reagan Moore
Is the "coll-name" an existing collection?
Reagan Moore
Shaofeng Liu
sliu@starrycluster:~/program/iRODS> iadmin
iadmin>lg
rodsadmin
CinegridA
iadmin>quit
sliu@starrycluster:~/program/iRODS> ils
/tempZone/CinegridContent:
C- /tempZone/CinegridContent/24flowers
C- /tempZone/CinegridContent/7bridges
C- /tempZone/CinegridContent/Flight-Patterns
C- /tempZone/CinegridContent/Jpeg2kEncoded
C- /tempZone/CinegridContent/cruiseship
C- /tempZone/CinegridContent/dalsa-prague
C- /tempZone/CinegridContent/eralanotte
C- /tempZone/CinegridContent/julia4d
sliu@starrycluster:~/program/iRODS> ichmod -r read CinegridA 24flowers
ERROR: rcModAccessControl failure status = -809000 CATALOG_ALREADY_HAS_ITEM_BY_THAT_NAME
sliu@starrycluster:~/program/iRODS>
best,
Leo
Is the "username" a valid user within your data grid?
Is the "coll-name" an existing collection?
Reagan Moore
Wayne Schroeder
What's in the server log (server/log/rodsLog*)? It may have some
additional information about the error.
I'm unable to reproduce the problem on my system.
- Wayne -
Date: Tue, 8 Jul 2008 12:25:21 -0700
From: "Shaofeng Liu" <liush...@gmail.com>
Shaofeng Liu
Here is the log...
Jul 8 12:24:21 pid:29695 NOTICE: rsAuthCheck user rods
Jul 8 12:24:21 pid:29695 NOTICE: rsAuthResponse set proxy and client authFlag to 5, user:rods proxy:rods client:rods
Jul 8 12:24:22 pid:29695 NOTICE: rsSimpleQuery get stat=0
Jul 8 12:24:26 pid:29695 NOTICE: agentMain: received disconnect msg from client
Jul 8 12:24:26 pid:29695 NOTICE: Agent exiting with status = 0
Jul 8 12:24:29 pid:28122 NOTICE: Agent proccess 29698 started for for puser=rods and cuser=rods from 137.110.119.127
Jul 8 12:24:29 pid:29698 NOTICE: rsAuthCheck user rods
Jul 8 12:24:29 pid:29698 NOTICE: rsAuthResponse set proxy and client authFlag to 5, user:rods proxy:rods client:rods
Jul 8 12:24:29 pid:29698 NOTICE: agentMain: received disconnect msg from client
Jul 8 12:24:29 pid:29698 NOTICE: Agent exiting with status = 0
Jul 8 12:24:35 pid:28122 NOTICE: Agent proccess 29701 started for for puser=rods and cuser=rods from 137.110.119.127
Jul 8 12:24:35 pid:29701 NOTICE: rsAuthCheck user rods
Jul 8 12:24:35 pid:29701 NOTICE: rsAuthResponse set proxy and client authFlag to 5, user:rods proxy:rods client:rods
Jul 8 12:24:35 pid:29701 NOTICE: recursiveFlag 1
Jul 8 12:24:35 pid:29701 NOTICE: bindVar[1]=18373
Jul 8 12:24:35 pid:29701 NOTICE: bindVar[2]=read object
Jul 8 12:24:35 pid:29701 NOTICE: bindVar[3]=01215545075
Jul 8 12:24:35 pid:29701 NOTICE: bindVar[4]=01215545075
Jul 8 12:24:35 pid:29701 NOTICE: bindVar[5]=/tempZone/CinegridContent/24flowers
Jul 8 12:24:35 pid:29701 NOTICE: bindVar[6]=36
Jul 8 12:24:35 pid:29701 NOTICE: bindVar[7]=/tempZone/CinegridContent/24flowers/
Jul 8 12:24:35 pid:29701 NOTICE: cllExecSqlNoResult: SQLExecDirect error: -1 sql:insert into r_objt_access (object_id, user_id, access_type_id, create_ts, modify_ts) (select data_id, ?, (select token_id from R_TOKN_MAIN where token_namespace = 'access_type' and token_name = ?), ?, ? from r_data_main where coll_id in (select coll_id from r_coll_main where coll_name = ? or substr(coll_name,1,?) = ?))
Jul 8 12:24:35 pid:29701 NOTICE: SQLSTATE: 01000
Jul 8 12:24:35 pid:29701 NOTICE: SQLCODE: 4294967295
Jul 8 12:24:35 pid:29701 NOTICE: SQL Error message: [unixODBC]Error while executing the query (non-fatal);
ERROR: duplicate key violates unique constraint "idx_objt_access1"
Jul 8 12:24:35 pid:29701 NOTICE: rsModAccessControl: rcModAccessControl failed
Jul 8 12:24:35 pid:29701 NOTICE: agentMain: received disconnect msg from client
Jul 8 12:24:35 pid:29701 NOTICE: Agent exiting with status = 0
Wayne Schroeder
Are you using 1.1, or are you still using a pre-release version? I
fixed some problems a little before the release that could possibly be
related (altho I really don't think so).
The error indicates that adding the new access information (row) to
the table (r_objt_access) would violate the contraint that each
user-obj item in the table must be unique. But the ICAT code should
be deleting all user-obj rows just before that. So it shouldn't
happen.
We'll have to do some more digging to see what's happening.
I've recently added a Troubleshooting Guide page to our web site:
https://www.irods.org/index.php/Troubleshooting_Guide
Toward the end it mentions spLogSql. If you can set that and restart
your server and run the test again, the log should tell us even more.
Thanks,
- Wayne -
Date: Tue, 8 Jul 2008 13:22:50 -0700
Shaofeng Liu
The weird thing is:
it happens to some dirs, not to others.
Best,
Leo
Shaofeng Liu
Best,
Leo
Wayne Schroeder
row somehow got into the table somehow, but even if that happened, the
code should clean it up before trying the new insert.
So let's try to track it down some more, tomorrow. And let's just
email each other for now (drop the cc to the irods-chat).
If you could try it with spLogSql set that might tell us something.
Also, if the table is not too big, you might do 'psql ICAT' and
select * from r_objt_access;
to see what that says.
- Wayne -
Date: Tue, 8 Jul 2008 17:12:04 -0700
Shaofeng Liu
Best,
Leo
Wayne Schroeder
Leo and I looked into this some more and found that the error would
occur when doing a recursive ichmod ('ichmod -r') and when one or more
of the dataObjects in the collection had multiple copies, that is more
than one replica.
This was a bug in some ICAT code and has been fixed for the next
release. It's a one line change (extending some SQL in
chlModAccessControl). I can sent this change to you if you need it
(email me). The bug causes the 'ichmod -r' to fail, but otherwise
does no harm.
Leo has patched his system and it seems to be working fine.
- Wayne -
Date: Tue, 8 Jul 2008 18:01:13 -0700
From: "Shaofeng Liu" <liush...@gmail.com>
Great. See you at 2:00PM.
Best,
Leo
[...]