Index Problem in Infocube

Hi BW lovers
I am having problem with the index in one of our cubes. This is the story:
1. I did a lot of compress on a cube with 41 000 000 records in. After that run DB statistics (ended up green).
2. After that I checked index in the performance tab and it was yellow.
3. I tried to repair index, and it did not help (still yellow)
4. Delete and created index again (still yellow)
5. I run a check in RSRV and there was one yellow saying "Index /BIC/FZHISTORY~900 type is BITMAP. Type NORMAL was expected". I tried to repair it but still the same problem.
6. I tried to run ABAP program SAP_INFOCUBE_INDEXES_REPAIR (still yellow)
7. Did all things in note 401242 (check DB02 etc)
Still yellow.... Anyone have a clue what to do?
Best Regards
/Pontus

I'm assuming this is an Oracle DB because of refrence to bitmap index.
The 900 index is an index that is added to the F fact table for InfoCubes wher you have specified they are to be partition (from the Extras menu option). This is an index on characteristic that was specified for partitioning (0CALMONTH or 0FISCPER). According to Note 217397, the index is supposed to be a bitmap index.
Partitioned Infocubes
There is an additional index for F facttables of partitioned infocubes in BW 3.x systems. Thereby it does not matter whether the infocube is standard or transactional. There is an additional bitmap index on the column of the F facttable that corresponds to the partitioning column of the E facttable This index's name is "900", i.e. "/BIC/F...~900" on the database.
So how do we reconcile your error msg and this Note? Msg saying it expected a Normal index and this Note saying it is supposed ot be bitmap?
This isn't a transactional cube is it?
Note 1024018 makes reference an earlier Note that indicates the 900 index was not included as part to of the rebuild index process at some previosu point, but the note it referenced didn't see, to clarify when the change was made. I would check and see if the 900 index actually gets rebuilt when you perform the repair.
Maybe you could have your DBA rebuild it if the Repair function is not doing it. Seems like some more OSS Note digging is in your future. Note 401242 - Problems with InfoCube or aggregate indexes - runs thru some diagnostic efforts you can pursue.
Note 1003360 - BW fact tables: Deleting index from process chain terminates - talks about a failure in PCs because the 900 index gets deleted twice.
Note sure what SP you're on - but since it looks like the 900 index has been a problem off and on, you'll have to see what Notes might be applicalable to your SP.

Similar Messages

Index construction for InfoCube went wrong due to SQL Error

Hello,
I searched the topic before posting most of them suggested to delete index manually and create it and check in RSRV and SE38 prg to SAP_INFOCUBE_INDEXES_REPAIR. this steps and all i followed but its not working fine.
The problem is Index construction for InfoCube went wrong due to SQL Error.
Job started
Step 001 started (program RSINDEX1, variant &0000000000203, user ID )
SQL: 16.07.2010 11:56:08 user
DROP INDEX "/BIC/FTTCSCH003~02"
SQL Error: [IBM][CLI Driver][DB2] SQL0904N Unsuccessful exec
SQL-END: 16.07.2010 12:06:10 00:10:02
System error: RSDU_INFOCUBE_INDEXES_DROP/DD_DROP_INDEX /BIC/FTTCSCH003 020 DB2
Report RSINDEX1 ended with errors
Job canceled after system exception ERROR_MESSAGE
Regards,
Ranga

as no one is my answering my question i am going to close this thread, admin pl delete this thread.
Thanks,
Ranga

Z-index problems in IE6

i managed to get IE6 to use z-indexes and layer properly..
everything that is except form elements specifically combo boxes
appear above my dropdown menus...??

This is not a z-index problem. It's due to the way that combo
elements are
drawn to the screen. There are only two solutions that I know
of -
1. Redesign the form so that there is no conflict, or
2. Hack the menu code so that the combo element is hidden
when the menu is
opened.
Murray --- ICQ 71997575
Adobe Community Expert
(If you *MUST* email me, don't LAUGH when you do so!)
==================
http://www.dreamweavermx-templates.com
- Template Triage!
http://www.projectseven.com/go
- DW FAQs, Tutorials & Resources
http://www.dwfaq.com - DW FAQs,
Tutorials & Resources
http://www.macromedia.com/support/search/
- Macromedia (MM) Technotes
==================
"Terrence Martineau" <[email protected]>
wrote in message
news:em7187$h79$[email protected]..
>i managed to get IE6 to use z-indexes and layer
properly.. everything that
>is except form elements specifically combo boxes appear
above my dropdown
>menus...??

Flash z-index problem

hi!
I have problem with z-index of my flash animation.
On Mac everything works great (on Safari and Firefox), but on Windows I have the same problem on all browsers.
Basically, the dropdown menu is displaying below my flash animation instead of above.
I'm surprised on mac everything is working perfectly and it doesn't depend on the browser...
is maybe due to the published javascript code generated by Flash ?
thanks

>I created some pull down menus using CSS. Directly below
the menu bar I
>had a
> Flash file that runs on load. The pull down menus look
corrupted in
> the
> browser (safari and FireFox) when they appear over the
running flash
> movie. Is
> this a z index problem? Any work arounds?
Safari and Firefox Mac have very broken and slightly broken
Flash
Players. There is no viable workaround unless you want to
eliminate all
"hover" effects. Adobe claims it's an Apple or a browser bug.
Apple
claims it's a failure on the part of Adobe to "cocoa-ize"
something or
other.
Al Sparber - PVII
http://www.projectseven.com
Popup Menus | Image Galleries | CSS Tutorials & Templates
Newsgroup: news://forums.projectseven.com/pviiwebdev/
CSS Newsgroup: news://forums.projectseven.com/css/
DW Newsgroup: news://forums.projectseven.com/dreamweaver/

What is the use of Indexes in ODS&Infocube??

What is the use of Indexes in ODS&Infocube??

An Index can improve the reading performance when data is searched for values of fileds contained in the index.
The fact whether the Index is used or not depends on the database optimizer, which decides it after taking the following under consideration:
- Size of the table
- Fields in the index compared to the fields in the statement
- The quality of the index (its clustering factor)
One backdraw of Index is that they decrease writing performance, as they are maintained during the write to the table.
Hope it helps you,
Gilad

SUNONE Indexing problem

have an indexing problem :
2 start points
1) myportal/portal/dt/ - main page robot can index it Oll Right
2) myportal/portal/dt/?provider=2DefTab&&2DefTabContaner.setSelected=Def98MainTableContainer&last=false&Def98TabContainer.setSelected=Def98STableContainer&tab=1&subdiv=01&node01=on&node5089=on&std_html=off&scroll_x=0&scroll_y=0 - dinamic page - robot can index only the first part of the page then
the interesting thing is however that teh first parts of the pages are identical and robot can index this part on main page and can not do it on the secons one
maybe some one faced this problem before? please help

Does "_type" really have a leading underscore? That must be a PITA to work with. Sorry my bad it does not have an underscore.
the queries always states the type of the PARTY.
I am worried that this PARTY table is kind of abstract and we are putting two types of information into the same table.Does this effect my index design?

Table attribute assignments / indexing problem

Appears that in the out of the box Ultra Search setup Oracle overrides the basic assignments of attributes. For example if I assign Author search attribute to table field auth , it gets overriden and indexed from what was in the meta data field of the incoming document. If pdf or other doc type it uses that author meta data field of doc instead of what attribute assignment I made. Is this just for this version and the next version will have the option? It appears that no matter what, the override happens. Any help or suggestions in this attribute assignment and indexing problem?

This is a bug in 1.0.3 and has been fixed in 9.0.2 and 9.2.

How to check for Active Indexes in an Infocube ?

Hello Gurus,
Is there a tcode to check/display the active indexes in an Infocube.
Thanks in advance.
Points to correct answers.

Hi,
we can check indexs uisng T codes SE11 and DB02
Useful to run the ABAP reports SAP_UPDATE_DBDIFF and SAP_INFOCUBE_INDEXES_REPAIR.
relevent links:
http://help.sap.com/saphelp_nw70/helpdata/en/80/1a6473e07211d2acb80000e829fbfe/content.htm
Re: Checking Indexes using SE11 and DB02
Regards,
BBC
Edited by: BBC Achari on Jul 4, 2010 6:23 PM
Edited by: BBC Achari on Jul 4, 2010 6:25 PM

How I solved my Spotlight+ Mail indexing problem

OK, so I thought I would post my solution to a stubborn indexing problem, for people who may have the same issue (and it does seem many people do).
Problem: Spotlight index was unreliable. In particular, Spotlight was terrible at indexing mail. It could not find the messages by content that it would find by searching the "from" field, nor could it find those reliably. Unusable.
What DID NOT work:
-- adding and removing any folder whatsoever in the Privacy preferences of Spotlight. The index would rebuild, but with the same problems.
-- doing this via terminal (e.g, using sudo mdutil -E /Volumes/MyHardDisk)
-- using Onyx
-- using Spotless
-- simply reinstalling the system
-- manually modifying the plist files inside the .spotlight folder to add and remove the folders that would not get indexed.
What DID work:
-- first, I reinstalled the system. That's not such a big deal. Upgrade up to 10.4.6
-- then, I authorized the root user (that's probably not necessary) (this, you do by using Netinfo manager)
-- Then, I deactivated the index (via terminal:
sudo mdutil -i off /Volumes/MyHardDisk
-- now the tricky part. This concerns Mail, in particular.
I have a very complex mail folder (tens of thousands of messages). I noticed that inside each mail folder, I had two things that were not supposed to be there
1. a copy of each folder.mbox which was empty, but with the same name without the extension .mbox. This is probably something that was created while importing mail from Panther to Tiger, for what I can tell.
2. Most importantly, a file named ".index.ready". This also, apparently, was an old legacy coming from Panther. I suspect this was the main problem for the lack of indexing in Mail.
Ok, so... I deleted these folders and files. If you have a complex structure of subfolders, you can delete the files with an Unix command by first cd to your mail folder, and then use
find . -name \.index.r* -ok rm {} \;
which will ask you confirmation before deleting each file.
It was faster to remove the empty subfolders by hand, as they all had different names.
-- Ok, at this point, I erased entirely the .Spotlight-V100 directory which stays at the root of your computer. You can do it with a rmdir command from the terminal or, as I did, you can login as root (remember, we activated root), use a utility that allows you to see hidden files (e.g., Tinkertool), and move it manually to the trash folder.
Then I emptied the trash. For superstition, I restarted, and let the index be built again (that took about 5 hours in my computer, for an HD with about 60 gb occupied on it).
Now everything seems to work perfectly fine. I don't know whether this procedure can be shortened, but I do know that nothing else worked for me. I thought I would post it here as I spent sooo much time to solve this problem, and furthermore, I found that none of the suggested solutions would work.
OK, your responsibility to do it, if you want to do it. Hope nobody needs to!
Best to you all,
l.
Mac OS X (10.4.6)

Thank you, thank you, thank you, Lucabo! Using ideas from your post, I also (at least for now) solved my Mail indexing problem. One day (not even after an upgrade) Mail just quit finding messages by content, even when I had open a message with that content!
What I did:
-- In System Preferences, told Spotlight to index only Mail (I just use grep from a Terminal window for everything else!)
-- Ensured Mail app wasn't running
-- Backed up my Library/Mail folder to an external disk
-- Turned off indexing via Terminal (after sudo sh), with
mdutil -i off /Volumes/MyHardDisk
-- Deleted old files in my mailboxes (inside the whatever.mbox folders):
.index.ready
mbox
tableofcontents
content_index
It was clear that mbox, tableofcontents, and content_index were left over from a previous version of mail - they hadn't been accessed since I upgraded.
Note that I didn't delete the whatever.mbox folders (just the plain mbox file inside)!
-- Deleted (in Terminal) the .Spotlight-V100 in the root directory
-- Restarted my Mac. The indexing started itself. Since I had it set up to index only Mail, it took about 3-ish hours to finish (I have 100s of 1000s of emails in thousands of folders).
After the indexing finished, all the test searches that I tried worked. From now on I will always be suspicious if search doesn't find anything!
Also, I just checked - using the Spotlight field on the menu bar now works better too. When the Mail search problem showed up, I tried using the Spotlight field and the Spotlight window, but they would just flash up a few items, remove the list, over and over, sometimes with the beachball, then finally settle to an empty list. Now the Spotlight field seems to work.
I infer that the Spotlight index gets corrupted, and that the Spotlight code doesn't notice that it's reading junk from the index...
BTW, about the same time that search quit working, some emails in my Inboxes got highlighted with a dark brown background color. Have no idea why. They aren't marked as Junk (the junk messages have the text in the usual lighter brown). But, when I select a message with the dark brown highlight, mark it as junk, then mark it as not junk, the dark brown highlight goes away. On some occasions, just quitting Mail and restarting makes the dark brown highlights go away - but not always. Sigh... at least it's not a show stopper, even if it is ugly!
Thanks again, Lucabo, for your post!
Mac OS X (10.4.6)

Index problem during the creation of a new partition

We have a range partitioned table, with a local spatial index on each partition. While trying to use the alter table command to add a new partition we get the following errors.
ORA-29855: error occurred in the execution of ODCIINDEXCREATE routine
ORA-13249: internal error in Spatial index: [mdidxrbd]
ORA-13249: Error in Spatial index: index build failed
ORA-13249: Error in R-tree: [mdrcritbl]
ORA-13231: failed to create index table [MDRT_D789CC$] during R-tree
creation
ORA-29400: data cartridge error
ORA-01031: insufficient privileges
ORA-06512: at "MDSYS.SDO_INDEX_METHOD_9I", line 7
ORA-06512: at line 1
ORA-06512: at "LPDAACECS_PART.UPDATE_METADATA", line 1937
ORA-06512: at "LPDAACECS_PART.UPDATE_METADATA", line 3625
ORA-06512: at line 1

I just wanted to expand on this for the sake of others who may need a bit more detail, having just resolved my similar problem.
In Oracle Enterprise Manager, in the left-hand tree view, expand Security then Users inside the relevant Databases entry.
Then select the name of the user/schema which needs to be able to perform the required task (in my case, create a spatial index from within a stored procedure).
On the System tab in the right-hand pane, highlight
Create Any Table
Create Any Sequence
Create Any Index
(depending on the task that needs to be performed)
Click the arrow to move these items into the "Granted" area. Click apply and your prayers have been answered. Mine were, anyway!
Regards
Stuart

Spotlight Indexing Problem

I've been scouring the internet looking in forum after forum for the answer to this problem for several hours now. I've tried all the usual fixes (which I will detail below) and none of them seem to be working.
I recently installed Leopard from Tiger on my MacBook. Since then I've been having a problem with Spotlight hanging while it's indexing. It gets about 80% of the way through before it starts hanging. I have no external hard drives connected (and therefore have not begun using Time Machine).
I Repaired Disk Permissions and booted up from the install disc and Repaired Disk.
I opened Terminal and entered +sudo mdutil -E /+ which restarted Spotlight indexing, but the problem arose again.
I added my entire hard drive to the Privacy list within Spotlight options. Spotlight stopped indexing (as would be expected). When I removed my hard drive from the Privacy list, the problem arose again.
Now, I have told Spotlight not to index any of the categories within Spotlight preferences. I will add each category in, one-by-one. This is currently taking place, and I will add posts as progress persists.
Can anyone help? I would really appreciate any kind of assistance in this matter.

I've been seeing the exact same problem and have posted queries with no solution. Currently my status is that I have switched off Spotlight indexing with an app called Spotless and am using EasyFind as my search tool. This is far from satisfactory though, as Spotlight is embedded inside many Apple apps as the search tool (Mail, for instance) and EasyFind doesn't replicate the function of searching inside apps. Since I rely on searching my mail more often than anything else on my drives, I'm not happy.
Spotlight has failed on both my my MacBook Pro and also my dual quad MacPro, despite my having done complete clean installs on both these machines. I've tried a process of elimination to attempt to find the file/files that are the culprit for the indexing hangs but it is a tedious and indefinite process, principally because it's impossible to know exactly when Spotlight gets hung up (I basically just have to listen for my fans to spin up watch the Activity Monitor for mds taking over the CPU). The console logs are almost no help in trying to find out what's going on, as they seem only to register the problem after I have quit Console and relaunched it (ie, they are not seeing the process hang as it happens).
One thing that may help you (although it didn't help me) is that Spotless allows you to delete old indexes created under earlier systems, that, as I understand it, may hang around through the install of 10.5 and interfere with new indexing.

Significant slowdown of Time Machine after upgrade to Lion; indexing problem

After upgrading to Lion over the weekend I notice a significant slowdown of Time Machine processing. I assumed the first backup or two would take some time but it's been a couple of days. The process always hangs up at the "Indexing backup" stage. It takes a couple of hours to finish.
I found several threads relating to slowed down backups on Lion, most seem to hang at the "Preparing" stage (not the same symptom) but I did find some good diagnostic pointers. I installed the widget to see what kind of messages were being produced and found a few dozen "Waiting for index to be ready (100)" messages at the end of the stream. Further research - this is not a good thing :-) It seems a couple of such messages can be ok, but this many usually indicates a problem. I'm going to discount the possibility of a hardware failure because everything was perfectly fine up to the point I made the upgrade. Plus the prepare and copy seem to proceed at the expected speed (I watched the most recent cycle just to be sure).
Here is the full message log (it's still running at the moment):
Starting standard backup
Attempting to mount network destination URL: afp://Paul%[email protected]/Paul%20DiMarzio's%20Tim e%20Capsul
Mounted network destination at mountpoint: /Volumes/Paul DiMarzio's Time Capsul using URL: afp://Paul%[email protected]/Paul%20DiMarzio's%20Tim e%20Capsul
QUICKCHECK ONLY; FILESYSTEM CLEAN
Disk image /Volumes/Paul DiMarzio's Time Capsul/Paul DiMarzio’s MacBook Pro.sparsebundle mounted at: /Volumes/Time Machine Backups
Backing up to: /Volumes/Time Machine Backups/Backups.backupdb
12.15 GB required (including padding), 490.69 GB available
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
Waiting for index to be ready (100)
I further found the advice and instructions on how to run the disk utility on the sparsebundle. I did a verify and repair, and the results came back that the drive appears to be ok (it did not say that it explicitly repaired anything).
But the problem remains.
I use a 1TB Time Capsule with plenty of free capacity (only half full). The TC and my MacBook are both hardwired into the home network, networking speed has never been an issue.
I'd sure hate to delete the backup file and start from scratch, there's over a year's worth of data there and I have, occassionally, wanted to go in there and pick up a downlevel or subsequently deleted file.
Any advice on other diagnostics or repair options to take? Thanks!
Paul

After posting I decided to keep trying to slog my way through the troubleshooting guide. Some of the steps are beyond my understanding but I did follow the advice in #D2 to delete the .Spotlight-V100 file from the sparsebundle. That action does seem to have done the trick, the backups are now completing in normal time. I ran a couple just to be sure. I've verified the backups with BackupLoupe and entered the Time Machine, all appears to be in order and as expected. The only odd side effect is that the backup volume is no longer ejected after completion but that might just be an artifact of all my mounting/de-mounting and playing with Finder. Anyway a very small price to pay for what looks to be a working system again.
Hopefully the good behavior will last through a few days and a few reboots :-)
ps - if you're reading this and don't know about the troublshooting guide it's here: http://web.me.com/pondini/Time_Machine/Troubleshooting.html (thanks pondini)

Spry dropdown z-index problems in IE

Hello
I'm having some trouble with the spry horizontal dropdown in
IE, and it's interaction with SmoothGallery. In every browser (with
the exception on IE 6&7 of course) the higher z-index values of
the menu cause it to overlay the JS image gallery below it. In IE
however it does not.
http://www.digitalrift.co.uk/projects/woods/
Works fine on every browser tested so I can't see how it'd be
a fault with the z-index values used. Anyone have a suggestion as
to a workaround for good ol' IE?
Thanks

All right, well I just figured out my problem, and hopefully this will help anybody else.
I was positioning my images using <position: absolute;> While this is definitely the easiest way to specifically place something on a page, Internet Explorer, it seems, assigns that z-value as the highest on the page, which is why my dropdown menus wouldn't appear on top of the image, but behind it.
The solution I found was changing the position attribute to "static" rather than "absolute". The "static" attribute can't use position qualities like "top", "bottom", "left", and "right", but has to use margins and float to be specifically place.
While this isn't as easy as absolute positioning, using the static value will allow the Spry dropdown menu to appear over top of the images instead of behind them.

Index problem in a spatial query

I try to run following query in SQL Plus:
SELECT A.objectid FROM TABBDG A, TABREG B
WHERE A.objectid = 68 AND B.regionname = 'Hong2'
AND SDO_RELATE(A.geoloc, B.geoloc, 'mask=INSIDE querytype = WINDOW') = 'TRUE';
To find out weather or not objectid = 68 is inside of region (='Hong2"). I got following error messages:
ERROR at line 1:
ORA-13226: interface not supported without a spatial index
ORA-06512: at "MDSYS.MD", line 1723
ORA-06512: at "MDSYS.MDERR", line 8
ORA-06512: at "MDSYS.SDO_3GL", line 57
ORA-06512: at line 1
But when I try to add an index to table TABREG, I got following error:
ERROR at line 1:
ORA-29879: cannot create multiple domain indexes on a column list using same indextype
Seem to me that the index was already there. Can any one help me? Thanks.

I found out the reason of problem. I used an old version of EasyLoader (4.5?). When I get the latest EasyLoader V6.5, there is no problem any more to run spatial query. You can get lates EasyLoader V6.5 from MapInfo or from me at: [email protected]
Thanks for all help.

Partial indexing problem

Hello all, I've recently developed a problem with Spotlight and it's ability to find files that I've tagged. I've been using Quicksilver to try and create on-the-fly smart folders for a specific tag, &project1, for instance. However, my first attempts were futile until I re-indexed Spotlight. After doing this, I was able to find some files with the &project1 tag, however, if I created new tags for files, Spotlight would not find them or even acknowledge that they existed unless I re-indexed again. This renders Spotlight almost useless for making smart folders quickly, as I have to constantly re-index my HD to find anything by tags. So my question is, does anyone have any idea what's going wrong? And more importantly, does anyone have any idea as to how to fix it? Please respond.
Dan

In Disk Utility, you need to run the Repair Disk function (not Repair Disk Permissions).
In order to do so on your main hard disk, you need to start the computer from the Tiger install DVD.
1. Insert DVD
2. Shut down computer
3. Turn on computer while holding 'C' key until you get to the grey spinner screen.
4. Select the language
5. Go to the Menubar- Utilities>DiskUtility
6. Run the Repair Disk routine.
-If it reports any errors, run it a second time to get the all OK
-If it reports errors that it can't fix, you may need an alternative application.
7. Then quit the installer, set your startup disk to the hard disk, and restart the computer.

Index Problem in Infocube

Similar Messages

Maybe you are looking for