|
POST
|
The format of the 'file=' file is documented with both 'shp2sde' and 'sde2shp', though it's unlikely the full range of features documented for 10.0 online were available at 9.1. You'll need to look in your local documentation for the features avaliable. Given that you are working with a view, it makes more sense to construct the view that will cause the least trouble at export (including limiting the column names to 10 characters). What does 'sdetable -o describe' report on your view? - V
... View more
06-02-2011
04:13 AM
|
0
|
0
|
1485
|
|
POST
|
You'd need to provide more details on the table and index allocation, and probably the contents of the DBTUNE keyword associated with the table, but it seems likely that indexes on the table are distributed across multiple tablespaces -- Inserts to a table generate index inserts as well. Note: Oracle 10.2.0.1 is not supported with Esri applications - the minimum supported release is 10.2.0.3. - V
... View more
06-02-2011
04:06 AM
|
0
|
0
|
755
|
|
POST
|
It's difficult to blame RAM or CPU when two identical hosts are provided for testing. We spent three weeks in the test lab and, based on the results that we had to present before an engineering review board, threw out the load-balanced 8 virtual server architecture in favor of a single physical server with a virtual failover host, and eventually threw out the failover host because it was unreliable with the needed fibrechannel drivers. By definition, virtualization has a cost associated with it (I call it the "virtualization tax"). One of the prime benefits of virtualization is that it lets you steal CPU cycles from unbusy systems in order to reduce the physical host footprint. Yet when the systems are NOT unbusy, and especially when the busyness is I/O related, there is a greater cost. GIS applications easily have the highest I/O demand in the business marketplace. Virtualizing production GIS servers causes them to have the highest taxes. You won't notice when the systems aren't under heavy load, but when they are, you can't help but notice it. I had VM administrators pleading with me to speed up physical deployment of an ArcSDE server because the four virtual servers it was replacing were increasing the data center's demand numbers to the point where it was endangering their contract performance metrics (when deployed, the ArcSDE connect time [with scores of concurrent users and thousands of feature classes] dropped from 40-120 minutes to 5 minutes). There was nothing wrong with the RAM or CPUs involved, and only two of the servers were really busy. That same physical server now services 3-4 times the client load of the four virtual hosts, and still has a similar connection time (with 30% more tables). - V
... View more
06-02-2011
03:58 AM
|
0
|
0
|
1917
|
|
POST
|
You need to speak with your database adminstrator about storage allocation in the database. It's quite possible that this doesn't have anything to do with SP2 application. - V
... View more
06-01-2011
05:39 AM
|
0
|
0
|
4026
|
|
POST
|
I worked on a project where we had to evaluate the relative performance of a cluster of VMs in a similar configuration with a single equivalent physical server. The exact details were under non-disclosure, but we found that a physical server out-performed a cluster of SOCs running on the same host as VMs, generating 40-120% more maps per minute over a 10+ minute test period. The details in our test report were sufficiently glaring that there was no objection when we requested variance from the "all systems shall be VMs" facility configuration requirement. We made use of dozens of VMs on that project, but not where production use of ArcGIS Server or ArcSDE was involved. - V
... View more
06-01-2011
02:44 AM
|
0
|
0
|
4260
|
|
POST
|
I once corrupted a file geodatabase when I did a Copy from ArcCatalog 10.0, but didn't wait until the job had completed before copying the .gdb folder to a network share (I have since put the Python window in the foreground). - V
... View more
05-31-2011
01:03 PM
|
0
|
0
|
2497
|
|
POST
|
Oracle Express is supported with ArcSDE 10 for non-production use, but you've got a couple of issues to work out first... ArcSDE has its own connection protocol (in place of OLE DB), so an OLE DB connection doesn't really have anything to do with ArcSDE (OLE DB connections from ArcGIS are not geometry type aware). ArcSDE uses Oracle OCI, so you need a robust TNS client (not Instant Client) for Direct Connect or application server connection to Oracle. Try an Oracle site for help in listener configuration -- you really don't want to consider ArcSDE until after you've got listener.ora (server) and tnsnames.ora (client) set up so that the 'tnsping' and 'sqlplus' utilities can successfully access your Oracle server from the client host. - V
... View more
05-31-2011
12:11 PM
|
0
|
0
|
3389
|
|
POST
|
I suggested you get comfortable with the SQL first because you need to *really* understand the query before using the ArcGIS GUI, to intuit what needs to be placed where. The ArcSDE API is quite clear on SE_QUERYINFO object components, but the Layer Properties form has three tabs that the components of the query reside -- Joins & Relates, Fields, and Definition Query. The task is to place the additional table in Joins, the join constraint in Definition Query, and restrict the result to just one spatial column in fields, but it doesn't look like GUI will give you the flexibility to use a SQL join constraint (ArcObjects might). A view is your best bet, but with that off the table, Definition Query is probably next best. It may be challenging to write your query so that the join will perform well when the spatial constraint is applied for rendering. Keep in mind that a "*" column list is pure poison when the result can only contain one geometry column -- explicitly referring to the columns with table alias will help document your procedure. - V
... View more
05-31-2011
09:35 AM
|
0
|
0
|
6408
|
|
POST
|
You can't use Esri's ST_GEOMETRY operators on Oracle's SDO_GEOMETRY datatype. You need to either use SDE.ST_GEOMETRY storage (which is the usual default for ArcSDE 10) or use Oracle's operators. While it's possible to use query definition to establish this sort of query, it would be far easier to get the SQL right in a view first, then either register the view or port it to ArcGIS layer definition, or use Query Layers to access the view (or SQL). Any way you cook it, the SQL would have to test the result of the spatial operator function, like this SELECT I created for a functionality test at a client site with corrupted permissions:
SELECT objectid
FROM poly_stg t
WHERE sde.ST_Intersects(
sde.ST_GeomFromText('POLYGON ((0 0,5 0,5 5,0 5,0 0))',2),
t.shape) = 1;
In your case you'd need to refer to both tables in a single statement as in this documentation example:
SELECT sa.id SA_ID, hs.id HS_ID
FROM SENSITIVE_AREAS sa, HAZARDOUS_SITES hs
WHERE sde.st_intersects (sde.st_buffer (hs.location, .1), sa.zone) = 1
ORDER BY sa.id;
Note that both tables must share the same SRID, or the comparison operator will always fail silently. - V
... View more
05-31-2011
05:30 AM
|
0
|
0
|
6408
|
|
POST
|
If ArcGIS does not regognize the directory as a file geodatabase, then the copy process has corrupted the data. The safest way to transport file geodatabases is probably as a zip archive, which will make sure all the files are there and that they have passed checksum on restoration. In my experience, 7-zip is *much* faster at compressing and decompressing zip files than native Windows capabilities. - V
... View more
05-30-2011
06:00 AM
|
0
|
0
|
941
|
|
POST
|
While not supported, the easiest way is to create empty templates of the tables in the current database, use 'sdetable -o create_view' to build the spatial view, and then edit the view definition to point to the other databases. If the table uses GEOMETRY/GEOGRAPHY, then just define the view yourself, and register the result with 'sdelayer -o register'. - V
... View more
05-30-2011
05:31 AM
|
0
|
0
|
1485
|
|
POST
|
While avant-garde when initialy introduced with ArcView 1.0, shapefiles have not aged well. They are limited by dBase naming restrictions, poor date support, inability to store numeric NULLs, and a lack of international character support. I'd recommend using ASCII as a transfer mechanism, but I've systematically stripped pre-9.x support from 'se_toolkit' over the past year (8.3 was retired in Sept 2008). Would waiting for 10.x clients help motivate your organization to begin an ArcSDE upgrade process? A lack of overlap between modern computers and the software products available when 8.3 was released (which are all now unsupported by their manufacturers) is going to increase the upgrade difficulty (as will the inability of anything but 10.0 ArcGIS clients to read 10.0 geodatabases). It's normally recommended to upgrade the clients first, but across the chasm of 8.3->10.0, you probably need to upgrade your 8.3 ArcSDE to 9.3.1sp2, then upgrade the desktops, and then upgrade the server again to 10.x. - V
... View more
05-27-2011
11:49 AM
|
0
|
0
|
464
|
|
POST
|
Ahh! Two Oracle installs, but only one tnsnames.ora? That drives me batty at times, too. You also need LOCAL (or TWO_TASK, on Linux) instead of ORACLE_SID if the ORACLE_HOME is different from the database. - V
... View more
05-26-2011
04:53 PM
|
0
|
0
|
1806
|
|
POST
|
The first thing to try is bouncing the listener -- Stop the database instance Stop the listener Start the listener Start the database instance Then try 'tnsping gisdata' and the sqlplus command-line. Try explicitly setting ORACLE_HOME and moving the Oracle and ArcSDE directories to the front of the PATH set ORACLE_HOME=... set PATH=%ORACLE_HOME%\bin;%SDEHOME%\bin;%PATH% It's possible the %ORACLE_HOME%\network\admin\sqlnet.ora authentication setting is causing issues... Try setting it to NONE Do you have an Oracle DBA resource to help? - V
... View more
05-26-2011
11:21 AM
|
0
|
0
|
1806
|
|
POST
|
If "sqlplus sys/password@gisdata as sysdba" doesn't work, there's no hope that the GUI will work for you. Try using the manual process. It will break the steps down so you can see what's happening. In truth, there's no reason to trust the GUI to create tablespaces and users for you, especially when the commands are so basic. You also need hands-on access to configure ST_GEOMETRY for remote use (which is a good habit to get into). There's no reason to have Opatch in the PATH after the patch has been applied; extraneous PATH entries can only cause trouble in the long run. I'll often create the service *before* doing 'sdesetup', since it populates the dbinit.sde with an initial value (which needs to be altered if your server is remote). Tech Support has a lot more experience walking folks through installs than I do (I've done so many that I can't break down the individual steps anymore). - V
... View more
05-26-2011
09:48 AM
|
0
|
0
|
1806
|
| Title | Kudos | Posted |
|---|---|---|
| 1 | a week ago | |
| 2 | 06-08-2026 09:13 PM | |
| 1 | 05-29-2026 12:51 PM | |
| 1 | 06-01-2026 06:03 PM | |
| 2 | 05-29-2026 08:31 AM |