|
POST
|
Hi there It was never possible to locate the cause of this problem. Unfortunately for various reasons it was not possible to get hold of a backup from a point before it happened so with ESRI UK support, it was not possible to compare them and see what had changed underneath. So the call had to be reluctantly closed. Kind regards Tim
... View more
02-17-2013
11:22 PM
|
0
|
0
|
1962
|
|
POST
|
"Opening the selected ArcMap document failed: The data frames of the map document could not be read. The data frames of the map document could not be read due to the following error: ox80040111 (ClassFactory cannot supply requested class) The last successfull loaded component was: Unknown ProgID ({FFFF0000-0000-0000-79CC-56770840FF04}) The specified file does not contain a valid ArcMap Document." Just a quick update. With regard to the above error. I pin pointed that down to two basemap group layers. If either were included in my MXD document and I down saved it to a 9.3.1 version, it would not open in 9.3.1 or 10.1. However am able to down save to 9.2 and have that open in 9.3.1 or 10.1. I don't have 9.2 installed to test that. Therefore it's not related to my LOAD-ONLY mode problems. The load only mode issue has not been resolved yet as awaiting backups of the data prior to the problem to see if that holds any clue. Kind regards Tim
... View more
02-04-2013
02:20 AM
|
0
|
0
|
1962
|
|
POST
|
I have no update to this in case it is of help to anyone else. It may be the solution user vdundzys had come across, I don't know. Today I was tracing and I got the error message, a reshape line must only have one part in it. However I got it twice in succession, so I did the same tracing again and then zoomed in to see what was happening. See attached picture. What is happening is that there are two vertices very close together but not along the same line I wish to trace. When I trace ArcMap is choosing to trace both of them. However it doesn't go from one directly to the other. As you can see in the attached image, there is a gap between the two. It may be that the tracing line is doubling back on itself but I wouldn't have thought that would cause an error. Therefore it may simply be creating a spur line and then carrying on from the vertex the spur line came off, without actually going back to it before carrying on, if that makes sense. Now the attachment shows I have really zoomed in to see this and you may be wondering why I have two points so close together. Well they are actually different data sets. I am tracing off existing road data, which is a series of polygons. At the end of each road I trace, I need have it split from the road it joins. Although the roads are already split in the existing data, they do not follow the same convention I have to follow for the work I am doing, so sometimes I end up with points very close together. Hence the conditions for this error to occur then exist. When I deleted the vertex that I did not require at the end of the spur line just now, I was able to finish the sketch without any errors. What compounds the error badly, when it occurs, is the fact that ESRI have designed ArcMap so that after the error and any other sketch error occurs, the sketch disappears making it very heard to locate the cause of the error and it also makes one have to do the work again, which is most annoying. It may not seem like much but when your doing large amounts of digitising, the last thing you want is to have to repeat your steps because of such things. It's not as if the software is cheap. Anyway I'm not sure what I would gain by taking this up with ESRI technical support but at least it is here in case it is of help to someone else.
... View more
01-08-2013
01:25 AM
|
0
|
0
|
2369
|
|
POST
|
Hi there I too am accessing ArcSDE 9.3.1 and getting the below error message. I am using ArcMap 10.1.1. "One or more layers failed to draw: Table attachments not supported in this release of the Geodatabase. Missing equals sign between attribute and attribute value." Unfortunately it does state which one or more layers failed to draw. Also it only comes up when I switch to layout mode and not data mode. I suspect it may only be occurring in one MXD or any MXD from which I copy out the layout and layers. However I am experiencing other problems which may or may not be related to this so my issue may be unrelated. I actually think the problem may be related to locally held basemaps. My computer went off for repair and has since came back. Temporarily I was working on another computer. So it may be that the cache was on that computer and now ArcMap does not like it because the MXD is on a different computer without the same cache. Just a educated guess. UPDATE: Having further looked into this. It seems the problem is with a file geodatabase held locally on my PC. This is a version 10 file geodatabase, which I am accessing through version 10.1.1. Whenever I try to view the data held in it, which is just 1 layers worth, it causes ArcMap to hang. I copied this geodatabase to another computer running version 10.0.4 and exported a downgraded version of my MXD to open on that PC and the geodatabase could be accessed without any problems. Out of interest, what are table attachments? I couldn't find much on line when I did a search. Kind regards Tim
... View more
12-06-2012
01:54 AM
|
0
|
0
|
1421
|
|
POST
|
Hi vangelo Thank you for your reply. I believe the use of the SDE user goes back some years. Other Oracle users are in use now but there is still some data as the sde user and it is used at times when data is being created. I will pass on the comments. As I found I could edit the data I then went to another computer running ArcMap 10.0.4. On that Ioaded up a 9.3.1 version of my MXD from late September. Then I went to edit the data but it wouldn't let me. Interestingly when I try to open this MXD in 10.1.1, I get the error: "Opening the selected ArcMap document failed: The data frames of the map document could not be read. The data frames of the map document could not be read due to the following error: ox80040111 (ClassFactory cannot supply requested class) The last successfull loaded component was: Unknown ProgID ({FFFF0000-0000-0000-79CC-56770840FF04}) The specified file does not contain a valid ArcMap Document." Yet the very same document opens in version 10.0.4. My ArcMap 10.1 is a fresh install last week with no customisation or add on install expecting for some of the things that come with the downloaded iso image. The computer was reimaged before that with no previous ArcMap installed either. Also of interest but very unlikely to be related to this, is that my current MXD becomes unstable in ArcMap after zooming and panning for 2-4 hours. The more panning and zooming I do per minute, the quicker it becomes unstable. However this is only for versions of ArcMap from 10.0.5 onwards. It works fine for all versions before 10.0.4. What makes that more intriguing is that when I created a new file and just added the layers I was currently displaying and didn't include the previous layout, it appeared that the MXD was stable. However I've not done enough panning and zooming yet to be sure this is really the case. Following on from above. I took my 10.1 version MXD and saved a copy as a version 10. I opened it in ArcMap 10.0.4. The first message I got was about the columing being NCLOB. Then I tried to edit the feature class and it would not let me. It came up with a message: Cannot edit the data in the folder or database selected. Yet this same feature class can be edited in the 10.1 version MXD, from which the version 10 MXD was just created! I hadn't even closed the 10.1 MXD down before attempting this. My initial thoughts on this are that perhaps there was a change with SP 1 for ArcMap 10.1, which is allowing me to edit the data whilst it is in load only mode. I couldn't do that when the problem first occurred and I was using no service packs. I will forward the information onto the ESRI UK support who are helping me with this. Kind regards Tim
... View more
11-30-2012
04:26 AM
|
0
|
0
|
1962
|
|
POST
|
Thanks for your reply. No external mechanism was used to populate the table with data, other than ArcMap itself. I did enable topology and archiving in ArcCatalog but that was it. The problem occurred on the day I upgraded to ArcMap 10.1. The latest copy of data itself exists in Oracle okay because I was able to copy it to a new feature class. At the time I could not edit the data. Just now however I found I could. I am not sure why I can edit it now but I couldn't then. However when doing the below command it is still saying that the LAYER_NAME in question is in load_only mode, despite the fact I have just created a feature in it and then saved it. sdelayer -o describe -l LAYER_NAME, shape -u sde -p sde It would be interesting to know what caused the problem. Was it an attempted reordering of the NCLOB formatted column? Someone from UK technical support has been helping with this. It's taken a while because I was waiting at my end for backups from before the problem occurred, which I may not now require, if the problem is resolved. The support person is now seeing if they can produce the same problem as me with another feature class. I did think that all that might need to happen is the removal of the duplicate keys. The problem was I couldn't find the duplicate keys. I did not enable topology or archiving on the new feature class. in case it ended up with the same problems as the last one. Out of interest, what is wrong with using the user SDE to store feature classes? I am not the administrator here but would pass on any information to them.
... View more
11-30-2012
02:05 AM
|
0
|
0
|
1962
|
|
POST
|
It's not ArcSDE which is hiding the types involved -- the API is quite clear on which is which type (just run 'sdetable -o describe') -- it's actually ArcGIS which is doing the generalization. The actual Oracle CLOB maximum is 4Gb, but ArcSDE uses the SE_CLOB_INFO storage structure, which limits the absolute length to 2^31-1. As you can see, arbitrary length LOB types are difficult to use at times; at a minimum, you need to work with a different set of helper functions than the usual string operators (and there's fewer LOB functions). I don't see any correlation between these two very different problems (besides low-level corruption of your Oracle instance) -- they might share the same cause, but one doesn't cause the other. - V Thank you for your reply. That is really interesting and useful information. I don't have access to the sdetable command myself because I am not the administrator but a colleague has access. I guess it would be helpful if ArcGIS didn't generalise the terms. Had I been presented with the underlying Oracle terms I would have seen the word NCLOB and though is this what I actually require? I would have then looked up the format and probably decided no. However as I was presented with the word text, I didn't look anything up as I didn't see that I would need to. I guess giving the actual format names would make a good idea for the ESRI ideas site, so I will post something when I get time. Kind regards Tim
... View more
11-30-2012
01:17 AM
|
0
|
0
|
5394
|
|
POST
|
Hi there I think I have found a possible cause of this problem. When creating the feature class I created three text columns with the figure 1073741822. From memory that was the largest figure I could enter into ArcCatalog. I was never going to reach that figure but I just wanted the largest number so that I wouldn't run out of space when typing. Even now when I go into ArcCatalog it tells me the column is a text column. Actually it is an NCLOB column in Oracle and I believe this is what is causing my problems below, even though my text it no where near that limit. What happens is that when I try to sort one of the columns in ArcMap, which has this high value, I get the following error message: Underlying DBMS error [ORA-00932: inconsistent datatypes: expected - got NCLOB] Unfortunately it doesn't tell me what it expected. There is another thread I put up about this, when I didn't know these two problems were linked: http://forums.arcgis.com/threads/69452-What-does-following-mean-ORA-00932-inconsistent-datatypes-expected-got-NCLOB?p=252904#post252904 When I went to save the data today I then got the above error and below it: Underlying DBMS error [ORA-01452: cannot CREATE UNIQUE INDEX; duplicate keys found (A2868_IX1)]. I have an on going support call with ESRI UK technical support and they are now going to try and recreate the problem at their end, now that I think I have found a way of replicating the issue. If I find out more I will update this in case it is of help to other users experiencing the same problem. Kind regards Tim
... View more
11-29-2012
02:07 AM
|
0
|
0
|
1962
|
|
POST
|
Hi vangelo Thank you for your reply. Sorry for not replying sooner but I appear to have missed reading your reply on here before now. I haven't been editing the tables outside of ArcMap. I only choose that large figure to make sure I didn't ever ran out of space when typing in text. As ArcCatalog only displays the column type as text I didn't consider what impact it would have since I knew I would never reach that amount or even come close. I don't think it was a random number. From memory I think it was the largest one I could enter into ArcCatlog. However colleague was told that in theory any number could be entered for an NCLOB so I may be misremembering. I didn't know what was going on underneath when I created it because ArcSDE hides all of the underlying data types and just lumps them all under the term text. I just assumed it would be okay because it is only a text column I am creating. However it appears not to have been okay because it is stored as NCLOB and that is causing problems as I am now finding out. I've currently got a support call with ESRI UK technical support and now appears that this error is related to that support call, not that I knew it at the time I raised that call. I have another forum thread I posted, which now is likely to be linked to this issue. I am about to update that thread in case it helps other people: http://forums.arcgis.com/threads/68053-Feature-class-stuck-in-LOAD-ONLY-mode-due-to-duplicate-keys-being-found-ORA-01452 Kind regards Tim
... View more
11-29-2012
01:55 AM
|
0
|
0
|
5394
|
|
POST
|
Never did find out and the problem still occurs even with 10.1.1. Tim
... View more
11-29-2012
01:35 AM
|
0
|
0
|
2369
|
|
POST
|
Hi there I have a feature class which I created thorugh ArcSDE 9.3.1, using ArcCatalog 10. This is stored in Oracle 10g. One of the columns is a string column of 1073741822 characters. This is stored in Oracle as a NCLOB column. I choose this number because I didn't know what my longest string in ArcMap would be. When I try to sort on this column I am getting the following error: ORA-00932: inconsistent datatypes: expected - got NCLOB. Can anyone tell me why that occurs? Kind regards Tim
... View more
10-19-2012
06:50 AM
|
1
|
4
|
13291
|
|
POST
|
Hi there Some months ago I created a feature dataset and within this a feature class. This was created through ArcSDE 9.3.1, via ArcCatalog 10. The back end being Oracle 10g. With this feature class I enabled archiving and topology. It is of course versioned. The other week I upgraded my computer to ArcGIS 10.1 and carried on working with the feature class. The following morning I suddenly found I could not edit the data. The data was now stuck in load only mode. I rang the UK technical support and myself and another colleague followed the steps to take it out of load only mode. The commands being: sdelayer -o describe -l LAYER_NAME, shape -u sde -p sde sdelayer -o normal_io -l LAYER_NAME,shape -u sde -p sde However the second command then generated the error: Error: Underlying DBMS error (-51). Error: Setting Layer to NORMAL I/O mode. ORA-01452: cannot CREATE UNIQUE INDEX; duplicate keys found (A2868_IX1) Layer is still in LOAD-ONLY mode! Since then myself and one of the UK technical support staff have been unable to find out what is going on. Using SQL Developer I have gone through various tables stored in Oracle and can find no duplicate keys on the columns I've been checking, although I may be checking the wrong columns of course. There is no A2868 table in existence but there is an F2868 and S2868, these referencing the archiving table Does anyone know which column or table might have the offending duplicate? Would it be an A or D table or simply the LAYER_NAME or LAYER_NAME_H tables? If I check the shape column in the archiving table then I found 541 duplicates. select count(*), shape from LAYER_NAME_H GROUP BY shape HAVING COUNT(*) >1; However I am not even sure if that is the correct column and being the archive table, I might expect duplicates. Whilst it would be possible to recreate the feature class, archives cannot be recreated; copied or anything much done to them apart from making them no longer an archive. So removing the archive isn't an acceptable solution in my opinion but it may be what I have to put up with if an answer can't be found. However I don't want to start again, only to find the problem reoccurs. Kind regards Tim
... View more
10-01-2012
06:39 AM
|
0
|
7
|
4912
|
|
POST
|
Thank you for your reply. I had not thought of using any of the land parcel tools. The data set I am working with is a record of features that are maintained. For example grass; flower beds; roads; pavements and so on. I've generally been spiting and merging polygons via digitising the split line for a split or selecting the require polygons and merging them. Then I would manually change the ID numbers to assign a new one or I would note the unused number and assign it else where to remove id number gaps. I only do this because the process is manual. Kind regards
... View more
09-24-2012
01:02 AM
|
0
|
0
|
1207
|
|
POST
|
Thank you for your reply vangelo. I'm not looking to do anything risky. I was just looking to do it all within the safe boundaries of ArcGIS itself. I assume from your comments that it is not possible. I have now upgraded to 10.1 and I get the impression it isn't possible in that either. Kind regards
... View more
09-21-2012
08:16 AM
|
0
|
0
|
1207
|
|
POST
|
Some positive feedback... Some tools simply can't give you a percent because the process doesn't lend itself to estimates - but this is a nice improvement. Would be nice if more tools provided an estimate even if it's an inaccurate one, just so we know it's working. Followers of this thread may want to know that Esri plans to add 64 bit background processing as an optional additional install for Desktop 10.1 SP 1, which will let you run models and scripts in 64 bit. This could be pretty useful, especially for big geospatial data processes that can crash 32-bit geoprocessing because of memory requirements. Unless I am mistaken in 10 all tools that had a background process, provided an estimate even if it was an inaccurate one. Over the years I have looked at the progress bars in different applications and often thought they were not great at estimating. However now that no bar exists at all, I think I miss it. I guess inaccurate is better than no accuracy. The other day I ran a geoprocess that took almost 3 hours. I didn't have a clue how long it would take or any idea when it might be finished until it had finished. I guess the problem is that no one has been able to design a progress bar which Incorporates the fact different parts of a process take different lengths of time. Offering 64 bit scripts will be good. Personally I don't have a 64bit computer at work but can see the usefulness of it.
... View more
09-21-2012
08:09 AM
|
0
|
0
|
1709
|
| Title | Kudos | Posted |
|---|---|---|
| 1 | 07-28-2017 03:18 AM | |
| 1 | 06-19-2019 04:30 AM | |
| 1 | 06-19-2019 07:29 AM | |
| 1 | 04-30-2015 02:15 AM | |
| 1 | 11-22-2011 02:22 AM |
| Online Status |
Offline
|
| Date Last Visited |
11-04-2024
05:38 PM
|