Owner of cache group

Similar Questions

• Cache Group drop in timesten 11.2.1

  Hello
  
  I'm giving a group of asynchronous cache in timesten. I followed the steps below to do so:
  
  (a) I use the DSN, UID, PWD, OracleID, OraclePWD specified connection string
  (b) if the replication strategy is 'always', change it to "manual".
  (c) stop replication
  (d) drop the cache AWT Group (+ drop cache group cachegroupname; +)
  (e) create the AWT updated the
  (f) initiate replication by
  (g) set the replication policy to "always."
  
  
  After step (d), I get the following error:
  
  Order > drop cache group cachegroupname;
  + 5219: temporary Oracle in OCIServerAttach() connection failure error: ORA-12541: listener TNS:no rc = - 1 +.
  + 5131: unable to connect to the back-end database: OracleNetServiceName = "servicename", uid = "inputuid", pwd is hidden, TNS_ADMIN = "/ opt/TT/linux/info", ORACLE_HOME = "/ opt/TT/linux/ttoracle_home/instantclient_11_1 +.
  + 5109: General error cache Connect: BDB connection not open. +
  The command failed.
  Command >
  
  If the error suggests that the cache connect has a problem? Should I restart the timesten daemon and try again? Please let me know what is the real problem.
  
  Let me know if you need information.
  
  
  Thank you
  V

  SQL * more problem is simply because you have all the appropriate directories listed in LD_LIBRARY_PATH. It is likely that your .profile (or equivalenbt) file was affecting those based on ORACLE_HOME and if it is now disabled could be Roble. Make sure LD_LIBRARY_PATH is set correctly and this problem will disappear.

  Character set issues is may be more problematic. It is mandatory that the database of game used by TimesTen exactly character corresponds to Oracle DB when TT is used as a cache. If the character sets are really different, then it is very serious and you need to rectify many things will not succeed otherwise. You need to pass Oracle DB either US7ASCII (it's probably a lot of work) or you need to change the character of TT value WE8MSWIN1252.

  To accomplish the latter you would:

  1. make a backup of the TT data store using ttBackup (just for security).

  2. for any non-cache tables (i.e. one TT), unload data from flat files using ttBulkCp - o...

  3. save the schema for the data store by using ttSchema.

  4. stop replication and cache agents.

  5. ensure this data store is unloaded from memory and then destroy the data store (ttDestroy)

  6 edit sys.odbc.ini to change the character set of data store.

  7. to connect to the data administrator of instance store (to create the data store). Create all necessary users and to grant required privileges.

  8. set the cahce ID username/password (call ttCacheUidPwdSSet (...,...))

  9 start the cache agent.

  10. run the SQL script generated by ttSchema of re - create all database objects (tables and hiding groups etc.)

  11 fill again all memory tables not cache from flat files using ttBulkCp-i

  12 re-load all groups using GROUP of CACHE of LOADING cache...

  13 restart the replication agent.

  That's about all (I hope I missed out any vital step).

  Chris

• owner of the Group evil datafile

  I need to change the primary group of oinstall DBA after the installation of oracle. As a result, my database (data files) has been created with the owner of the group as 'oinstall' instead of "DBA". Any problem if I stop the DB, change the property or the data files to 'DBA '? I followed the recommendation to install Oracle and the oracle OS user belonged to groups both oinstall and dba.
  
  Thank you!

  There should be no effect as long as the owners are the same... because permission has only read/write for the owner and nothing else. As long as your landlord remains the even there should be no problem.

  Concerning
  Tim

• WRITETHROUGH ASYNCHRONOUS CACHE GROUP

  Hello
  I have a small question... is it possible create a GROUP of CACHE WRITETHROUGH ASYNCHRONOUS with AUTOREFRESH option?
  
  This error pops up so that the time of creation,
  
  concerning
  
  lewismm
  
  
  CREATE ASYNCHRONOUS, WRITETHROUGH TEST

  AUTOREFRESH INTERVAL INCREMENTAL MODE 5 SECONDS CACHE GROUP
  PAUSED STATE
  Of
  DBADMIN. () TEST_CACHE
  OBJECT_NAME VARCHAR2 (128 BYTE),
  SUBOBJECT_NAME VARCHAR2 (30 BYTE),
  OBJECT_ID TT_INTEGER NOT NULL,
  VARCHAR2 (7-BYTE) STATUS,
  VARCHAR2 (1 BYTE) TEMPORARY,
  DATA_OBJECT_ID TT_INTEGER,
  OBJECT_TYPE VARCHAR2 (BYTE 19),
  DATE OF CREATION,
  VARCHAR2 (1 BYTE) SECONDARY SCHOOL,
  DATE OF LAST_DDL_TIME,
  TIMESTAMP VARCHAR2 (BYTE 19),
  VARCHAR2 (7-BYTE) STATUS,
  VARCHAR2 (1 BYTE) TEMPORARY,
  VARCHAR2 (1 BYTE) GENERATED,
  VARCHAR2 (1 BYTE) SECONDARY SCHOOL,
  KEY (OBJECT_ID) ELEMENTARY SCHOOL
  );

  7025: illegal use of the key word for 'FROM', the character position: 114
  ... DIFFERENTIAL 5 SECONDS INTERVAL DBADMIN SUSPENDED STATE. TEST_CAC...
  ^^^^
  The command failed.
  Command >

  No, this is not possible. At present, the only way to get
  two-way synchronization in the same group of cache is to use a USERMANAGED
  cache and to designate all tables as SPREAD (_synchronous_ group with AUTOREFRESH
  WriteThrough). However, it is potentially dangerous (it is a form of)
  multimaster replication but with no conflict detection/resolution) and
  relatively slow (because it is synchronous).

  There are some neat features coming in the next version of TimesTen
  (11 GR 2), which is just entering beta, which will greatly expand the
  features of the cache connect feature. Watch this space...

  Chris

• Multiple table cache group

  Hello
  
  For example, I want to cache the 3 tables, a child table with two FKs:
  PARENT1 <-CHILD-> PARENT2
  With TT 11.2 and restriction of Group of cache (only one table of root), I need to create two groups of AWT cache:
  PARENT1 <-CHILD and PARENT2
  In this case, the FK of PARENT2 is not implemented in TT.
  An update on CHILDREN may succeed in TT and fail when propagated to Oracle, for example, if the parent in PARENT2 row has been deleted by another session in TT.
  
  Is there a way to cache these 3 tables in TT with all FKs applied?

  Sorry, but no.

  Chris

• Of the group as the owner of another group - error LDAP 50

  I installed some groups within the IOM and for some of these members of the group, I would like to be able to manage membership in another group.
  
  I have:
  
  Group1 with owner User1, User2, user3 and user4, User5, Utilisateur6 members
  With owner User1 and User2 Group1 group2.
  
  In the bosom of the DAS should user4, User5 and Utilisateur6 now be able to assign members to Group2 right?
  
  User4 becomes a "LDAP: error code 50 - insufficient access rights" error when you try to assign members within Group 2.
  
  I completely understand you how spread the rights or is something wrong?
  
  Thank you.

  You must be an administrator of the group in order to manage a group.

• insert into the table of GLOBAL of AWT DYNAMICS with passthrough CACHE GROUP 3

  Hi all
  
  I have AWT DYNAMIC GLOBAL GROUP CACHE table on the customer table
  
  
  Relay 3:
  insert into customer values (5, 'US', 'Dublin', 'KkROAD');
  3340: Delete/update/insert/truncate Passthrough of the grid cache tables are not allowed
  
  
  but when I insert it to PL/SQL passthrough block 3:
  
  
  Start
  insert into customer values (5, 'US', 'Dublin', 'KkROAD');
  end;
  
  /
  
  PL/SQL procedure successfully completed.
  
  
  record inserted, but I
  
  1. Select * from customer (passthrough 0)
  can not see the record inserted in Timesten
  
  2 see the recoerd inserted in oracle
  
  3. can not insert gives same record in TT agiain .it
  342: introduced key already exists on Oracle or remote node
  
  
  is that corect behavior, can tell me

  I need to confirm, but I think that the fact that the Pl/SQL based insert works is a bug; He must also fail with error 3340. I will check into it and get back to you.

  The reason that SELECT * FROM CUSTOMER doesn't nto show the record is this "dynamic load" (extraction of missing data from Oracle) works only when the query contains a key equality predicate on a primary key or foreign defined on one of the tables in the Group of cache (it's documenetd in the cache user guide).

  The reason for which the insertion fails later in TimesTen is that we are to check if the record exists in Oracle (who does) and rejecting the insert (behavior correct and expected).

  It is not clear to me why you would want to use PassThrough to insert a row in Oracle in this case. The idea is surely that you insert the TimesTen online and it gets proagated to Oracle via AWT...

  Chris

• Owner/group display on remote server

  I would like to see the user and the Group of all files on my server. On Dreamweaver CC is it possible to add a column that says this?

  I'm not aware anyway, you can view these attributes in DW file panel. If you right-click on a file name on the list of files from the remote server, you can see what the permissions are for owner, user and group, but it must be done one at a time.

• How full the refresh Group cache (incremental autorefresh) without interruption?

  Hello

  I have a group of cache read-only mode incremental autorefresh.

  Its works fine, but I want to make some time full a refresh of the cache group.

  The command 'REFRESH CACHE GROUP' do the job but after the "unload" and before the "charge" of the cache group, no additional data is available in cache group.

  I try with the option "commit all 0 rows" and autocommit = false, to force the command UPDATE to be performed in a single transaction, but the behavior remains the same.

  Is there a way to perform a full refresh without losing data? (I think yes because "FULL autorefresh" mode works like that).

  Thanks in advance.

  You can try the following:

  1. stop the Cache Agent

  2. change the AUTOREFRESH of the GROUP disabled CACHE STATE

  3 change the AUTOREFRESH of the GROUP CACHE STATE we

  4. start the Cache Agent

  The steps above should trigger a complete refresh for the mentioned cache group.
  Please make sure you have enough PERMSIZE available to keep 2 copies of the cache group when updating FULL.

  Kind regards

  Simon

• What is the difference between regular dynamic cache AWT and AWT group?

  Hi Chris, on the dynamic cache of AWT group, you said that:
  ------------------------------------------------------------------------------------
  A group of dynamic cache is where, for some types of access (key equality) in SELECT/UPDATE/DELETE and for certain types of iNSERT as a "cache miss" (no corresponding line found in TimesTen) will cause a dynamic extraction of the files required from Oracle to be loaded into the cache, and also returned to the application (in SOME cases).
  ------------------------------------------------------------------------------------
  So I guess that for the regular cache AWT group, it will not have any dynamic fetches files required from Oracle to be loaded into the cache
  because the cache AWT regular group, we load the data into RAM first, right?
  
  
  Thank you
  
  SuoNayi

  That is right. For a group of AWT regular cache it is load you explicitly any required data explicitly using LOAD CACHE GROUP. To cache DYNAMIC groups you can "Prime" group cache using CACHE of LOADING GROUP if you wish, but it will also make dynamic extraction for the types of the SQL operation supported. They each have their uses depending on your use case.

  Chris

  Published by: ChrisJenkins on November 28, 2012 13:15

• Automation of creating groups procedure cache

  Hi all
  
  I've got ~ 100 tables in Oracle DB and I want to cache.
  I don't want to define a group of 100 cache because it's a lot of time.
  Is there a method to automate the creation of cache groups?
  
  Thank you.

  Hi 928879,

  Unfortunately, there is no way to automate this process. You can write a script for the unloading of the descriptions of the tables but the definition of a cache groups (create a cache of read-only group), you must write by hand.

  Kind regards
  Gennady

• Group cache cleaning

  Hello
  
  How to clear cache groups without dropping if Group cache is type AWT or SWT?
  
  SHD

  I guess you want to just take all or part of the data in the tables of cache group?

  UNLOAD the CACHE GROUP CacheGroupName [WHERE...];

  Chris

• Managed user group cache

  Hello
  
  I created the Group of managed users cache as follows:
  
  create cache group writewherecache usermanaged
  AUTOREFRESH
  DIFFERENTIAL MODE
  INTERVAL OF 30 SECONDS
  STATE
  of interchange.writewhere
  (PK NUMBER NOT NULL primary key,
  VARCHAR2 (40), PROPAGATE) ATTR
  where (interchange.writewhere.pk between ' 105 'and ' 106');
  
  
  Oracle have 5 rows in the table, but now TT ' select * from interchange.writewhere ' statement doesnot show any result.
  
  
  What is the problem?
  
  Published by: user11969173 on November 4, 2009 02:30

  Yes, AUTOREFRESH State must be "suspended" to run a GROUP of CACHE of LOADING. By default, State is set on a break when the cache is created but set you it explicitly on IT in your create statement.

  When you run the LOAD CACHE GROUP statement, what was the response? It is said normally ' n instances cache loading "." He said in your case? When have you under inert atmosphere the 5 rows in oracle, you post the transaction? If you connect to Oracle using a new SQL * more (or SQLDeveloper) session you see lines in Oracle?

  Chris

• Immediate refresh of the cache read-only with autorefresh incremental on group

  I work with a group of read-only cache that was created like this:
  
  Create group readonly TT_SOME_TABLE cache
  AutoRefresh
  differential mode
  interval milliseconds 300000
  State on
  Of
  () UNE_TABLE
  SOME_COL_1 NUMBER (11),
  SOME_COL_2 CHAR (1),
  SOME_COL_3 varchar (8) inline non-null,.
  PRIMARY KEY (SOME_COL_1));
  
  In normal use, the auto-refresh every 300000 milliseconds (5 minutes) is good - but in a particular scenario I want to manually force an immediate update of the Group cache from the command line.
  
  The following two commands in ttIsql do the trick:
  
  command > alter cache group tt_some_table set autorefresh State suspended;
  command > refresh cache group tt_some_table commit all 256 rows.
  
  However, the Refresh command issues a response message "21 cache of the authorities concerned.". This is the total number of lines, oracle holds for une_table, whereas before the update I've only updated 1 row. This makes me think that my manual refresh is not progressive. Am I wrong? On the plus side, the Refresh command switches the autorefresh return State on 'on' which is useful.
  
  Is there a better way to make an immediate update on a group of cache that is normall auto-actualise?. Can be done more?

  Manul LOAD / REFRESH is always a complete refresh rather than a differential (they work by different mechanisms). There is no way that I know to force an immediate refresh of incremental. You can try setting the AUTOREFRESH State pause and back to on. That could force an immediate refresh, but I have not tried so I do not know whether there will be.

  Chris

• Index for the Group of readonly cache table

  Hello
  
  Suppose I have a table in Oracle which is cached in TT as a readonly cache group.
  The oracle table has a primary key, based on columns (c1, c2, c3).
  
  Now, the oracle table also has a unique index (for example defined on columns c4, c5) in support of some queries.
  
  (1) is there a way to propagate this unique index for the cache group? (so that we can perform queries effectively TimesTen)
  
  (2) is it correct to assume that the group create cache statement MUST have the columns (c1, c2, c3) for the primary key? (that is, exactly the same as the pharmacokinetics of the oracle)
  
  
  Thank you!

  (1) after you have created the cache in TimesTen group, simply create the index needed on the tables of cache in TimesTen (using CREATE [UNIQUE] INDEX) as you would for any table.

  (2), it depends. A caching table must have a primary key defined in TimesTen. This PK must map to either a real PK defined in Oracle (same column names, order of the columns, data types, nullability, etc..) or a unique index defined on a set of columns not null (column names of the soul, order of the columns, types of data etc.).

  Chris