Eric J Ostrander's

ClearCase / ClearQuest / Git/Stash "how to" pages.


ClearQuest Errors






Failed to check out a ClearQuest_TeamTest license key: FLEXlm Error -2, Invalid license file syntax OR no licenses in file
This error occurred immediately after initial install. This install was going to be a stand-alone server. Even though this was to be a server, the "client side" of the install needs to know where to find a license.
In CQ 2003.06.00+, in the Rational License Key a Administrator, go to Settings-> Client/Server Configuration. Click the box that says "Add Server". In right-hand pane, fill in the license Server Name in the Values column.
In CQ 2001+, in the Rational License Key Administrator, go to Settings->Service Configuration and check the box "Use License Server. In CQ 2000A-, in the Rational License Key Administrator go to the Settings tab and select Client at the bottom of the page. For both, select "Use single server" and type in the name of the license server. Restart the license server: Start -> Settings -> Control Panel -> FLEXlm License Manager -> Control tab, Stop & Start. If the Stop reports that "Server stop failed", it was probably never started in the first place.

Back to the INDEX.




Failed to check out a key. FLEXlm Error -15, Cannot connect to license server
For some reason, the FLEXlm license manager failed to start after a reboot. Simply go to Start -> Settings -> Control Panel -> FLEXlm License Manager -> Control and Start the License Server.

Back to the INDEX.




ERROR! Execution of a hook failed during the action Action.
Can't locate auto/CQEntity/GetFieldval.in @INC

The hook is attempting call a function called "GetFieldvalue" which doesn't exist. Check the hook code and change it to GetFieldValue, with a capital V.

Back to the INDEX.




Line Number <integer> : Unrecoverable Error: All values need to be quoted. e.g. "BLA BLA BLA"
This error occurred when importing date into CQ using the Rational ClearQuest Import Tool. It's telling you that the quotes surrounding the import field data in the import file are out of order. Ensure that all field values are surrounded by double quotes. Also ensure that there are no field values that have quotes inside them, such as might be found in multi-line text fields. Simply correct the problem and rerun the import.

Back to the INDEX.




Unable to create SQL Anywhere database service.
This error occurred on Windows while attempting to create an SQL database service via the Rational Administrator. At the point where you type in the Account, make sure you have selected Other Account and that the login name is in the syntax of domain\sql-login. If this SQL server is not on a domain and the sql-login is a local account, simply replace the domain name with the local server name.

Back to the INDEX.




Could not start the Flexlm license manager service on \\hostname.
Error 0002: The system cannot find the file specified.

This error occurred immediately after installing Flexlm on Windows and attempting to start the service. The permanent license keys were already successfully imported. In the FLEXlm control panel, under the Setup tab, ensure that the License File entry only has the path to rational_server_perm.dat. If rational_server_temp.dat is listed, remove it.

Back to the INDEX.




FLElm error: -15,10. System Error: 10061 "WinSock: Connection refused"
machine: Cannot connect to license server -15,10:10061 (Winsock: Connection refused)
The first error appeared during the attempted import. It was accompanied by the statement "Bad License" in the Rational License Key Administrator and by "Invalid license key" in the FLEXlm control panel.
The second occured after a successful import of new TLA licenses when checking the Status in the FLexLM Control Panel.
The error was circumvented by manually adding the license information instead of simply invoking the .upd file. Start the Rational License Key Administrator. Enter the information in License -> Enter a License. Restart the FlexLM daemon when prompted.

Back to the INDEX.




!%$ Failed to insert package queries and report formats into user database
AdMetadataMgr: The Project AdEntityDef cannot be gotten because it does not exist
Location ClearQuest Core:admetadatamgr.cpp: 1069

Updated: 08/05/00
Version: CQ 2000
This error occurred when attempting to upgrade a CQ database. However, it only appeared when one manually did an upgrade or attempted to create a new database and not during "Test Work", which does an upgrade as well.
This is a non-fatal error simply warning you that some data is missing in your schema. The datbase gets upgraded anyway. Inspect your schema for any oddities or create a new database that is associated with a previous version of the schema.

Back to the INDEX.




Can't get server: Cannot find license file -1,359:2 (No such file or directory)
Updated: 08/05/00
Version: CQ 2000
This error indicates that the license file on the license server is actually missing. A reenter of the license information via the Rational License Key Administrator solved the issue.

Back to the INDEX.




An error occurred associating the request CQ-record-id
AdEntity: The "cc_change_set" AdField cannot be gotten because it does not exist.
Location: ClearQuest Core:adentity.cpp:575

Updated: 08/05/00
Version: CQ 2.0 P2, NT 4.0 SP6a, CC P9,10
These errors occurred when attempting a checkout in a VOB with a non-UCM integration to CQ. The CQ schema must have the ClearCase package added to it. Also be sure to add any ClearCase package updates, such as "Upgrade 1.0". If you've already added the CC package to the schema, is it checked in? Have the associated databases been upgraded with the latest schema changes?x

Back to the INDEX.




ERROR! AdActionDef: The AdActionUserDef cannot be unbuilt becasue it does not exist. Location: ClearQuest Core:adactiondef.cpp:835
Version: CQ 2000 patch 2
This error occurred when moving an action's Access Control from one group to another. More specifically, if one creates a group such as "test2" and then a group called "test", if in the schema's Actions matrix you set Access Control for an action to User Groups and then choose "test", CQ will set it to "test2" anyway. That is, "test2" appeared first in the groups list and CQ chose the first group that contained the string "test". So, when you attempt to move the check from the "test2" box to the "test" box (to set it the way you intended), it will generate the error message. This appears to be a bug and has been reported to Rational.

Back to the INDEX.




Failed to checkout a key for ClearQuest: FLEXlm Error -18, License server does not support this feature
Version: CQ 2000 patch 2
This error occured after rearranging licenses and installing new TLA floating licenses. It was simply pointing to the incorrect license file. That is, even the import of the new licenses went ok and showed up in green in the Rational License Key Administrator, the FLEXlm Control Panel was still pointing to the _temp.dat license file. In the FLEXlm Control Panel, go to the Licenses tab and ensure the correct one is listed. Use the Show License file button to verify.

Back to the INDEX.




Invalid login name "logon" specified or user "logon" is not subscribed to any database
Unable to logon to the MS_ACCESS database "\\path\mdb".

These errors occur when attempting to connect to login to CQ when there is no network connection, or a Microsoft Loopback Adapter LAN connection is not employed on Win2k. On NT 4 it wasn't nesessary to be connected to a network or have a loopback apapter in place.
The first error occured using Flexlm 6.1. It prompts for a logon, but then errors out stating that the logon was invalid. Inspection of the Details box reveals the second "real" error about not being able to connect to the database properly.
In either case, simply connecting the computer to a LAN solves the problem. If unable to connect to a LAN with the Win2k CQ server, a loopback adapter will have to be created:
Start -> Settings -> Control Panel -> Add/Remove Hardware Next>
Add/Troubleshoot a device; Next>
Add a new device; Next>
No, I want to select the hardware from a list; Next>
Network adapters; Next>
Manufacturers: Microsoft; Network Adapter: Microsoft Loopback Adapter; Next>
Next>
Finish
The TCP/IP Propertied of the Loopback Adapter do not have to be configured in any way.
Now, when offline, "dialup" the Microsoft Loopback Adapter network connection before accessing CQ.

Back to the INDEX.




Failed to checkout a key for ClearQuest: FLEXlm Error -42, Invalid parameter
This error occurred when attempting to start CQ for the first time after installing it on a new server using FLEXlm 7.0f and CQ2001 on Win2k.
If you are sure the imported licenses are valid, open the Rational License Key Administrator and go to Settings -> Servie Configuration. Ensure the License Key Source in the middle of the page is pointing to the correct server name.

Back to the INDEX.




Error in determining upgrade information. Possibly invalid database.
Version: CQ 2001
This error occurred in the Rational ClearQuest Maintenance Tool when attempting to connect to an existing, previously working, Oracle schema respository. The administrative login to the Oracle instance was invalid, not the database.

Back to the INDEX.




A user information entry in the database user-database is newer than the corresponding entry in the master database.
Version: CQ 2001
This error occurs when attempting to update user information via the User Administration tool in the Designer. This error can also occur when attempting to delete a user database from within the Designer. If the designer is open when a user changes their own information, such as their password, from the client or web interface, they will get an error message that states the database was not upgraded and their password was not changed. However, it really was changed in the user database they are logged into, but not in the master database. Now, when you attempt to push new user information to the user databases from the Designer, it gives the above error message. There is no choice but to log out of the Designer and log back in so that both master and user databases will be in sync.

Back to the INDEX.




Unable to connect to database server; unable to start specified database.
This error occurred when attempting to connect a CQ2001A client to a SQL Anywhere server schema repository when the two machines were in different subnets. Normally, all that is needed to connect to a schema repository is to enter the correct paramters in the CQ Maintenance Tool. Those parameters are normally kept in a cqprofile.ini file on the server. However, as of CQ2001A, if SQL Anywhere is chosen as the vendor, there is no way in the Maintenance Tool to specify any special connect options.
To connect to a SQL Anywhere schema repository across subnets, the following system environment variable must be defined on the client:
  CQ_SA_OPTIONS = "-x tcpip{dobroadcast=no;host=server-ip-address}"
Back to the INDEX.



ERROR! Unable to login to the database using the following connection information: ... Could not find file '(unknown)'.
This error occurred while attempting to delete a database from the CQ2001A Desginer. The reason for error is that the physical database was deleted from the system first. You'll need to following the procedure for logically deleting the db if the physical db is gone.

Back to the INDEX.




Cannot display the page.
This error occurred when attempting to log into CQ in the web interface using CQ 2001A and SQL Anywhere on Windows. This is a registry permission problem.
1) Open regedt32.
2) Go to the key HKEY_LOCAL_MACHINE\Software\Sybase\SQL Anywhere\Sybase SQL Anywhere 5.0.
3) Give the Anonymous user Full Control of this key. Select Allow Inheritable Permissions.
4) Go the HKEY_USER\.Default\Software\Rational Software\ClearQuest and perform step (3).
5) Reboot the machine.

Back to the INDEX.




[State =01000][Error=1326][Microsoft][ODBC SQL Server Driver][SQL Server]"Logon failure: unknown user name or bad password.
This error occurred when attempting to login into CQ via the web. The login error occurred because the IUSR_machine account was not defined on the server where the SQL database resided. To rectify this, the IUSR_machine account must be a domain account or defined locally on the SQL database server machine.
Details:
Error 1326 means "Logon failure: unknown user name or bad password." This error is raised because the IIS Service does not have any rights to use the named pipe.
In order for a client to gain access to a Windows named pipe, the client needs to be validated by the Windows server. This is normally accomplished through a Workgroups-style validation method where identical usernames and passwords are created on the client and the server, or the domain method where both the client and server are domain members.
The Internet Information Server (IIS) setup process creates a Windows ID called IUSR_machinename (where machinename is the name of the Windows server), and adds that user to the Guest local group. The IIS service runs under this Windows account name.
If the connection to IIS is anonymous, IIS uses the Windows account IUSR_machinename and so any network activity performed by IIS is done under this account ID. If the SQL server resides on the same server as IIS, then named pipes connections work and validation is successful. If the SQL server resides on a separate server from IIS, then the IUSR_machinename account needs to be validated on the Windows server that SQL resides on.

Back to the INDEX.




Unable to obtain a ClearQuest web license.
Unable to obtain a CQ web license when CQ Web and CQ Client are running on the same machine. One possible cause could be the permissions required for the FlexLM License Manager key in the registry.
By default, the FlexLM License Manager key in HKEY_LOCAL_MACHINE\SOFTWARE has default permissions for the "Everyone" group (Windows) or the "Users" group (Windows 2000). The account that is used as the Anonymous User for the website accesses this key using those rights.
If those rights are modifed or the group restircted from that key, the web will not be able to access the site, while the licenses will still be served to other workstations from the OS so the ClearQuest client functions normally.
To correct this, add either the approriate Anonymous user or the "Everone/Users" group with the appropriate special permissions.
1) Using Regedt32 navigate to the key HKEY_LOCAL_MACHINE\Software\FlexLM License Manager.
2) Choose Security and Permissions. Add the Anonymous user or the IUSR or the group as appropriate. Be sure to add the selection to cascade changes to the subkeys.
The default permissions are:
Windows: Query Value Set Value Create SubKey Enumerate Subkey Notify Delete Read Control
For Windows 2000: Query Value Enumerate Subkeys Notify Read Control

Back to the INDEX.




User database "dbname" upgrade failed. The dbname database has not been upgraded to the current revision of ClearQuest.
The error occurred using CQ 2002 when attempting to push new user information out to a user database.
At one point, the underlying schema of CQ was upgraded. The database in question had not been upgraded when CQ was. Run the following command to upgrade the user database. This example is for an Access database. Contact Rational Tech Support if working with a different db vendor. The upgrade will create a new, separate database that must be given a name different than the original.
  # installutil upgradeuserdbsystemversion -dbset dbsetname admin-login admin-passwd logicl-dbname MS_ACCESS "" "path-to-upgraded-db" admin admin admin admin
Back to the INDEX.



!%$ ===== Caught a (PaMsgException *) exception =====
Updated: 12/15/08
!%$ The database "MASTR" belonging to master database "dbset" is an invalid name. Enter the correct name of a ClearQuest user database.
This error occurred on a Windows system while attempting install a new package into the schema repository using CQ 2002. The command being executed was: [packageutil installpackage -dbset 2002.05.00-slqserver admin "" DefectTrackingSetup 1.0]. The packageutil subcommand called "installpackage" is, according to the documentation, synonymous with the subcommand "installintoschemarepo". However, the use of installpackage generated the error. The subcommand was switched to installintoschemarepo and the package was installed correctly.
This error can also occur if the dbset name has non-word characters in it and has not been double-quoted. For example, if your dbset name is my-dbset, put it in double quotes: "my-dbset".

Back to the INDEX.




***error: Cannot save document: This is not a valid file name.
This error occurred when attempting to a run a report from within the CQ Client 2003.06.00 using SoDA for Word.
Open a SoDA template from the Windows Explorer (C:\Program Files\Rational\SoDAWord\template\ClearQuest) and go to SoDA->Options. A bug in the install of 2003.06.00 didn't create the directory listed in "Tight integration report directory". After simply creating that directory, SoDA reports could be run from within CQ.

Back to the INDEX.




ERROR! The database name for fieldname can't be renamed.
This error occurred in the Designer when attempting to rename a field. Normally when you rename a field, the "Field Name" gets changed, which is what the user sees, and the "DB Column Name" does not. However, in this case, the "Field Name" gets changed, you get the error, but the user still sees the old name.
I don't know if there's better solution, but I got around the error by simply checking the schema in and then checking it back out again.

Back to the INDEX.




The record "" could not be found. It may have been deleted.
Updated: 02/04/10
Version: CQ 7.0.1
This error can occur when a multiline text field has far, far too much data. The connection between that situation and the error message is unknown.
The solution is to access the back-end database and clear out the offending field.
To avoid the error in the future, place a validation hook on multiline text fields limiting the data to, say, a few thousand characters with an error message stating that anything larger than that should be included as an attachment.

Back to the INDEX.




New schema repository doesn't show up in the web interface pulldown list.
Updated: 08/16/06
When you create or link to a new schema repository, when logging into CQ it will prompt you for which schema repo to use. However, note that the list of schema repositories is personal to the user that imported the .ini file or created the link. This means that only the user who originally installed and set up the web interface can make modifications to the .ini for the schema repos that show up when logging into the web. It is unknown if there is a way to have a "system" list of schema repos that all users of that server can see.

Back to the INDEX.




The database that you are trying to open or convert is currently in use, or you do not have permission to open it exclusively.
You attempted to open a database that is already opened exclusively by user 'username' on machine 'machinename'.

Updated: 08/31/06
These errors occurred when attempting to upgrade an MSAccess database. Ensure all users are out of the database. If this is the schema repository, all schemas must be checked in.
If you still get the error, there's a possibility that the lock file is corrupt. In the directory where the database is, there is a corresponding .ldb file. Delete it. If you get the error "Can't delete database-file-name: It is being used by another person or program.", check to see if there are any hung "clearquest.exe" or "cqdesign.exe" processes. If there are, kill them. If not, you'll most likely have to reboot the computer and then delete the .ldb file.

Back to the INDEX.




These fields have invalid values:
Updated: 03/28/07
This error occurs when saving a record. If a field is listed, simply go to that field and fill in a value. If the error message doesn't list any fields, it probably means that it is a hidden field. It is the assigned "Title" of a field and not its name that appears in error messages. If there is no field listed, the schema probably has a bug in it, as there is no way for a user to fill in the value. Unfortunately, because the error doesn't say which is the offending field, you'll have to look at all "hidden" fields for that record type related to the action being performed.

Back to the INDEX.




The database dbname claims to be using schema schemaname revision # version # but the latest version of that schema revision is #.
Updated: 06/08/07
This error can occur when logging into the schema repository or user database, or when attempting to push changes out to the user database from the Designer.
The reason for the version mismatch is unknown. If the version in the user database is higher than the version claimed by the schema repository, the schema repository version can be modified to match. In the schema repository, modify the value in the MASTER_SCHEMAREVS table to match the user database. In the error message, "revision" refers to the schema level, whereas "version" refers to how many times you have pushed the schema out to the schema repository during the current checkout of the schema.

Back to the INDEX.




cqstreambuf::underflow: Conversion error.
Updated: 12/15/08
CQ version: 7.0.1
This error occurred while attempting to import a schema on the CLI using cqload.
A schema dump file has special end-of-line formatting. If that file is opened in an editor, such as Word, and then saved, the end-of-line characters can get changed. Notepad is safe for editing the file and changing the globally replacing the schema.

Back to the INDEX.




The name UCM_CQActBeforeChact is already in use.
Updated: 12/15/08
CQ version: 7.0.1
Received this error when attempting to upgrade package versions.
For unknown reasons, the schema was actually at UnifiedChangeManagement package version 5.0, while the database thought is was at 4.0. So, the upgrade was attempting to apply the 5.0 changes and couldn't.
Note that the only way to determine the actual schema package level was to examine the schema and look at the changes applied by the various package revisions. Eventually you'll be able to see that a given change wasn't applied yet, therefore the previous version is the actual version of the schema.
The fix is to update the database with statements similar to the following. Note that you must quit the SQL editor to see the changes.
	> update packagerev_usage

	set packagerev_dbid =
	(select master_dbid from packagerevs_1 where package_dbid =
	(select master_dbid from packages_1 where name = 'PACKAGE') and rev_string = 'NEWREV')

	where package_dbid =
	(select master_dbid from packages_1 where name = 'PACKAGE') and packagerev_dbid =
	(select master_dbid from packagerevs_1 where package_dbid =
	(select master_dbid from packages_1 where name = 'PACKAGE') and rev_string = 'OLDREV') and
	schema_name = 'SCHEMANAME';

Back to the INDEX.




Valid table name expected.
Updated: 02/05/09
CQ version: 7.0.1
Received this error when attempting to import a schema.
The full error received was:
	Error detected at line 36 of file schema.txt.
	The line was: ADD property ( "Unix Script Language", "", "", "PERL")
	Valid table name expected.
The export was done using "cqload exportintegration" from version 1 to another version. The import was attempted using "cqload importschema". However, even though the export started with version 1, you cannot mix exportintegration with importschema, as there will be missing data; hence the error message. In this particular case, I wanted to exclude several versions from the end of the import. So, the only way to accomplish that is to perform a full import of the schema, and then delete the unwanted versions before pushing to a user db.

Back to the INDEX.




ios::failbit set or ios_base::failbit set
Updated: 03/07/17
Version: CQ 8.0.1
This error occurred when attempting "cqload importintegration".
While a very cryptic error message, it's simply telling you that it cannot find the import text file. That is, this is a "file not found" error.

Back to the INDEX.




The name Project is already in use.
Updated: 07/17/09
Version: CQ 7.0.1
This error occurred when attempting "cqload importintegration".
The schema version being imported is attempting to create a record type called "Project" that already exists in the destination repository.
The fix in this case was to log into the destination repository, check out the schema, delete the offending record type, and the check the schema back in, which solved the problem. The import then recreated the Project record type as needed.

Back to the INDEX.




The name "Host" is not a valid entitydef.
Updated: 07/17/09
Version: CQ 7.0.1
This error occurred when attempting "cqload importintegration".
The schema version being imported is attempting to modify a record type called "Host", but that record type doesn't exist in the destination schema.
The solution in this case was that it had been created in the schema version just prior that had not yet been imported. So, including the previous version in the importintegration solved the problem.

Back to the INDEX.




String data code page conversion failed.
Updated: 05/26/10
Version: CQ 7.0.1
This error occurred when attempting to run a query in a script. As with many CQ error messages, it doesn't lead to the actual problem.
The reason for the error was the input to BuildFilter. In this case, the input to BuildFilter was a list of record IDs, as in:
	$operator->BuildFilter("id",$CQPerlExt::CQ_COMP_OP_IN,\@records);
The problem was that the list of records was imported from a text file that was an export of a client query. If you export a query's result set into a text file, even if the query only returned the IDs, the file will also contain a column dbids, as in:
dbid	id
33659828 TEST00105397
33662429 TEST00107988
...
Simply cleaning up the input file to be a single column of IDs solved the issue. It's unknown how it correlated to the error message.

Back to the INDEX.




"String": invalid identifier.
Updated: 10/07/10
Version: CQ 7.0.1
This error occurred when attempting to run a query in a script.
The reason for the error was a misspelling in a fieldname used in a BuildFilter. The "invalid identifier" means that the column name doesn't exist.

Back to the INDEX.




unknown exception from CQQueryFilterNode_BuildFilter
Updated: 10/07/10
Version: CQ 7.0.1
This error occurred when attempting to run a query in a script.
The problem was that the record type in BuildQuery was misspelled (record type doesn't exist in the database). It is unknown why it got all the way to the filter before throwing an error, and then not a very informative error.

Back to the INDEX.




Error 1309.Error reading from file: ...
Updated: 04/06/11
Version: CQ 7.0.1
This error occurred when attempting to patch a release area with msiexec.
The script was being run against a release area on a remote share. As such, each path was a VERY long UNC path. At some point the length of the string to a given file inside the release area exceeded the Windows 256 character limit.
The solution was to do a couple things. 1) Instead of using the UNC path directly, as in "\\machine-name\share\long_path\release_area_parent_dir", the path down to the release area was mapped to a local drive letter, as in "Z:\release_area_parent_dir". 2) Renamed the release area itself from "CQ_701_Installation_Release_Area" to "701_release_area". Together those two reduced the length of the path string by about 75 characters, which was enough to apply the patch properly.
Note that when the error occured there was no way to finish the install of the patch correctly. Because the patch was half applied, the release area had to be rebuilt from scratch.

Back to the INDEX.




CQ Designer: Out of Memory
Updated: 04/12/11
Version: CQ 7.0.1
This error occurred when working in a CQ Designer on records that have large numbers of fields and/or excessive nested reference list fields. The out of memory error can also be accompanied by very high CPU usage.
The ClearQuest 7.0.x Designer can sometimes perform very slow and consume close to 100% CPU usage, or show an increase in virtual memory consumption. It can take several minutes before the properties screen displays. When right-clicking and viewing the properties of a form field, the ClearQuest Designer might temporarily freeze and then display an Out of Memory error. This behavior did not occur in the ClearQuest 2003 releases and is reportedly fixed in 7.1+.
To fix this problem, a registry value will need to be enabled.
1) Enter regedit or regedt32.
2) Go to HKEY_CURRENT_USER\Software\Rational Software\ClearQuest.
3) Create a new key called "Diagnostic".
4) Right click on the new Diagnostic key and select "New -> String Value". Name this Behavior.
5) Double click on the Behavior string value and insert for the Value data: SET_NEST_LEVEL_3=TRUE
Setting this registry value to "TRUE" will now set the nesting level to 3, which will resolve the "out of memory" problem. This registry value only needs to be enabled on a machine in which the Designer is exhibiting the behavior.

Also, for the new Designer, increase the Java heap size to an appropriate value. For example, if a checkedout schema takes up 1GB of memory, then the heap size should be set to at least 1500m. http://www-01.ibm.com/support/docview.wss?rs=988&uid=swg21381316

Back to the INDEX.




SQLFetchScroll odbc_fetch_type: RETCODE=-1, State=S1000, Native Error=-954
Not enough storage is available in the application heap to process the statement

Updated: 04/12/11
Version: CQ 7.0.1
This error occurred when running a programmatic query against various records in a user database.
The reason for the error was an extremely large amount of data (several MB) in a multiline text field. In addition to the large amount in the record itself, the AuditTrailLog tracks the "old" and "new" values, which means when that data changed, it essentially doubled the size in the AuditTrailLog's multiline text field.
The solution is to clean out the multiline text field and place the large amount in an attachment. However, if you do that by a Modify on the record, there will be yet another AuditTrailLog entry that will contain the "old" value. So, the record itself and the associated AuditTrailLog records need to be cleaned out using SQL.
To prevent this from occuring in the future, a BASE action validation hook should check the size of all multiline text fields on the current record to make sure they don't exceed, say, 10,000 characters. If the user has placed more than that in a field, produce a validation error message asking them to put the information into an attachment.

Back to the INDEX.




ERROR! Cannot apply the requested changes to the record-type "id
Updated: 05/20/11
Version: CQ 7.0.1
This error occurred when attempting to Apply changes to a record that was being edited by a different user at the same time.
The only recourse is to copy and paste any lengthy work to a separate file, Revert the current action, and the perform the action again.
By default, CQ is in a first-come first-served mode. That is, if two users have the same file open for edit at the same time, the first user to Apply the changes gets their record committed, while the second user must Revert and start over.
As of CQ 7.1, administrators can set up pessimistic locking. If enabled, a user is immediately alerted that another user is currently editing the record and they'll have to wait. See the release notes and admin manual for details on setting it up.

Back to the INDEX.




'FieldName' is null or not an object
Updated: 08/22/11
Version: CQ 7.1.1
This error occurred when testing in the eclipse web interface after a new field was added to a form.
Testing of the same change passed in the client and in the old web interface. This is apparently a bug in the new web interface.
See https://www-304.ibm.com/support/docview.wss?uid=swg21440085
The workaround requires removing a cache file, and is detailed in the technote. Restarting the web services or rebooting the server also resets the cache file. The fix is to patch CQ.

Back to the INDEX.




ERROR! SQLNumResultCols: RETCODE=-1, State=08S01, Native Error=0
[IBM Rational DataDirect][ODBC DB2 Wire Protocol driver]Unexpected Network Error. ErrNum = 10052
SQL statement="select max(schema_rev) from master_schemarevs where schema_id = 15"

Updated: 07/14/11
Version: CQ 7.0.1.12
This error occured after making some schema changes and then attempting to apply those changes to the user db. Many other changes had been applied to the user db during the same schema checkout without issue. The schema could be accessed by starting a different designer session, but it obviously only shows the changes since the last save. Clicking the Save Work button caused the error too.
Unfortunately, the only resolution found was to Undo Check Out. Yes, all unsaved changes were lost. The lesson is to keep pushing the Save Work button even when not ready to push to the user db for unit testing.

Back to the INDEX.




The system is attempting to delete and object that does not exist: Object Type: ActionUser
Updated: 07/20/11
Version: CQ 7.0.1.12
This error occured when attempting to import a schema revision into the next environment using cqload importintegration.
The schema version was being migrated from the Dev environment to the Test environment. Part of the schema change in Dev was to remove a User Groups definition from a record type's action permissions. Unfortunately, sometime in the past, somebody removed that definition manually from the Test environment. An inspection of the import file revealed that the following line was attempting remove a definition that didn't exist in Test.
	DEL actiondef_usergroups  ( "ReleaseRecord", "Validate", "Admin")
The offending line was manually removed from the import .txt file and the cqload importintegration completed without error.

Back to the INDEX.




The name Unix Script Language is already in use.
Updated: 07/18/11
Version: CQ 7.0.1.12
This error occured when attempting to import a schema revision into the next environment using cqload importintegration.
The cause is that the exportintegration was done from schema version 1 onward. The exportingregration can only be from version 2 changes onward. If you include version 1, it will attempt to create information that already exists in the destination schema.

Back to the INDEX.




The cursor is not positioned on a valid current row.
Updated: 07/25/11
Version: CQ 7.0.1.12
This error occured when running a programmatic query.
Inspection of the code revealed the cause to be that the query didn't return anything, yet the script still made a call to GetColumnValue.
The solution is to test if the query returned anything prior to attempting that API call, as in:
	if ( $resultSetObj->MoveNext == $CQPerlExt::CQ_SUCCESS ) {
		$value = $resultSetObj->GetColumnValue(1);
	} else {
		print "The query didn't return anything.\n";
	}
Back to the INDEX.


The user "login" does not exist in the database.
Updated: 07/26/11
Version: CQ 7.0.1.12
This error occured when attempting to copy personal queries from one userdb to another with a Perl script.
The cause is that even though you can log into a db using a case-insensitive login_name, when calling $workSpaceObj->SetUserName($login), the $login must match the case of how the login is defined in CQ.

Back to the INDEX.




Execute perl hook script failed, no detail information available (RC=1)
Updated: 08/02/11
Version: CQ 7.0.1.12
This error occured after creating a new record type and attempting to test it. The error occurred multiple times immediately after attempting to create a new record and before the form appeared.
If a new reference field's choice list was left as "Default", there was no error. If even simple code was added to the choice list hook, the error occurred.
Along with the new record type and its fields, a new global script was defined. However, just to look at the new form, the global script had simple dummy code in it, as in: "return;".
The problem was the global script "return" was not contained inside a "sub" function. It was a bare global return. When that was discovered and properly placed inside a "sub" function, the error went away.

Back to the INDEX.




Type mismatch in expression
Updated: 08/02/11
Version: CQ 7.0.1.12
This error occured when accessing a newly added field on a form. The field was originally defined as a short string, but then changed to a reference.
The error included a SQL statement that had in it "where T1.name = T2.dbid", which is a problem, as the name is a reference to a different record type and dbid is an integer. I'm not sure how it got that SQL statement.
The fix was to delete the new field, and recreate it as a reference. Perhaps the change of type got messed up somewhere.

Back to the INDEX.




Failed condition: AdIsNonEmpty(value)
Updated: 08/09/11
Version: CQ 7.0.1.12
This error occurred when programmatically updating a stateless record.
A value was being added to a list box field whose choice list was set "Limit to List". To avoid overwriting the old value, the old value was placed into a variable and prepended to the new value, as in:
	$entity->SetFieldValue("Fieldname","$orig_value\n$new_line\n");
However, in this case, the was no original value, so the SetFieldValue was effectively entering a blank line above the new line. A blank line isn't allowed by the choice list. The proper way to handle it is.
	if ( "$orig_value" ne "" ) {
		$entity->SetFieldValue("Fieldname","$orig_value\n$new_line\n");
	} else {
		$entity->SetFieldValue("Fieldname","$new_line\n");
	}
Back to the INDEX.


dbwin32: ERROR: Unable to execute path\dbwin32.exe: No such file or directory
Updated: 08/12/11
Version: CQ 7.1.1
This error occurred when attempting to launch dbwin32 on a box with a working installation of CQ. This error doesn't mean that dbwin32.exe itself isn't found, but rather that dbwin32 can't find some other file it needs.
When you run dbwin32 from Start->Run, it executes a pointer in the ...\common directory of your CQ install, which points to dbwin32.exe in the ...\ClearQuest directory of your install. The dbwin32.exe in the ClearQuest directory was missing. It's unknown how it came to be missing, as the rest of the installation works fine. I copied the executable over from a different install and it worked. Inspection of another 7.1.1.0 install had the dbwin32 executable in the correct place, so go figure.

Back to the INDEX.




Caught an exception of an unknown type
Updated: 08/25/11
Version: CQ 7.0.1.12
This error occurred when doing a cqload importintegration.
The import file did not contain a previous version from the source schema, so did not have what it needed to apply the new changes. It's unknown what exactly was missing, but an inspection of versions revealed that one was missing from the import file.
For this reason, it's an extremely good idea to place comments on schema version imports indicating what versions of the source schema are being imported. For example, if you export Development schema versions 98-100 and import them into the Test environment, during importintegration use a comment like "Dev versions 98-100". If those versions create, say, version 80 in the Test environment, when that schema version is migrated to Production, use a comment such as "Test version 80". In the Development environment, be sure write change ticket numbers into the comments upon checkin of the schema. In that way you'll have traceability from a Production schema version back to ticket numbers that were worked on in Development. It's extremely important to get into this habit so that errors like this one can be easily diagnosed.

Back to the INDEX.




Invalid Credentials: Either the login name or the password is incorrect.
Updated: 08/26/11
Version: CQ 7.0.1.12
This error occurred when attempting to log into a database using a script.
The error message seems to indicate that the specified username or password was incorrect. If those two are known to be correct, ensure the specified user is marked as active, assigned to the desired database, and that the user information has been pushed to the user db.

Back to the INDEX.




There is a reference to an object that does not exist: Object Type: MasterdataObject
Updated: 08/31/11
Version: CQ 7.0.1.12
This error occurred when attempting to log into the ClearQuest User Administration tool. The login succeeded after acknowledging the error, but the list of groups appeared truncated. Right-clicking on a user was ok, but right-clicking on any group crashed the tool without error. An Import of userinfo from a different schema repo resulted in the same error. See technote: https://www-304.ibm.com/support/docview.wss?uid=swg21152566
It's reported that this error can be caused by starting the creation of a new user or group and then cancelling, but that issue has been fixed in 7.1.1 and later.
The following series of SQL commands will clear up the problem. Log into the affected schema repo using pdsql. The master_dbid is from the error message. For safety, perform a backup of the schema repo database before attempting this.
# Run the following sql to check if the user record exists in the master_users table.
# There must be 0 record returned.

	SQL> select * from master_users where master_dbid = master_dbid;
	(0 record).

#Run the following sql to check if any related records exists in the
# master_links table for types 1, 2, 3, and 4.  Here is an example SQL statement:

	SQL> select master_dbid,link_type, from_master_dbid, to_master_dbid from master_links where to_master_dbid = master_dbid and Link_type = 2;

	master_dbid,link_type  from_master_dbid, to_master_dbid
	----------- ---------- ----------------  --------------
	nnnnnnnn    2          nnnnnnnn          
	nnnnnnnn    2          nnnnnnnn          
	(n records).

# If there were no records in the SQL statement, but there are any related records
# in the next, you must delete the related records you found.

	SQL> delete from master_links where to_master_dbid = master_dbid and Link_type = 2;
	(n row affected).
	SQL> delete from master_links where from_master_dbid = master_dbid and Link_type = 4;
	(n row affected).

# Exit the SQL connection.
# Launch the ClearQuest User Administrator Tool to check if the error doesn't occur.
# If it does, there will probably a different master_dbid, so the steps must be run again.
PDSQL
pdsql -v db2 -db prdcqmst -u clqadmin -p fri12dec -s rsprtl01 -co PORT=65015

select * from master_links where to_master_dbid = 16829249;

delete from master_links where to_master_dbid = 16829249;

quit;
Back to the INDEX.


Resource 'cq.repo.cq-record:dbid@dbset/db': not found.
Updated: 09/02/11
Version: CQ 7.1.1
This error occurred clicking a button associated with a reference list field while in the Eclipse web interface.
The button was attempting to execute a Post-action record script that wasn't defined. That is, the Properties of the button was set for "Enable for Web -> Post Action Hook", but no hook was selected in the Hooks list.

Back to the INDEX.




CQDBSETUP could not instantiate dbsetup object
Updated: 01/17/12
Version: 7.0.1.8
This error occurred when attempting to import a .ini file just after install.
One problem could be that one or more registry entries did not get created properly. Run "cq-home/rereg.bat" and then reboot.

Back to the INDEX.




Unable to resolve activity
Updated: 12/17/12
Version: 7.0.1.8
This error occurred when viewing or modifying a record in the full client.
You can log into any db you want by importing the correct profile via the ClearQuest Maintenance Tool. You can then view and transition records as normal. However, if the current machine doesn't have CC installed, or is pointing to a region other than where the change set's VOB is tagged, you'll get the error.

Version: 7.1.2
This error also occurred when attempting to access a CQ record after an associated VOB had been renamed and its permissions changed. Stopping and restring the CQ client solved the problem.

Back to the INDEX.




Plug-in com.ibm.rational.clearquest.ui was unable to load class com.ibm.rational.clearquest.ui.query.CQQueryExtension.
Updated: 05/15/12
Version: 7.0.1.0
This error occurred when starting the eclipse CQ client.
The cause is unknown.
Remove the "C:\Documents and Settings\login\.Rational\ClearQuest" folder. The path is "C:\users\login\... on Windows 7.

Back to the INDEX.




<Unavailable> in UCM change set
Updated: 02/08/12
Version: 7.0.1.8
This error occurred when clicking on the Unified Change Management tab in an acivity record.
You must have a at least one UCM view started on your local box to get file name information from CC. The easiest way to start a view is to start the CC Explorer and simply click on a view. CC will start the view process for you. You can use "cleartool startview viewtag" from the CLI as well.

Back to the INDEX.




Error 1316. A network error occurred while attempting to read from file ...\Installer\1033_ClearQuest.msi
Updated: 02/08/12
Version: 7.0.1.8
This error occurred when attempting to uninstall ClearQuest through the Add and Remove Programs applet.
The error indicates a corrupt file. In this case, it's the file needed to perform the automated uninstall.
CQ will have to be uninstalled manually. Contact Rational Support for instructions on how to manually remove everything from the C:\ drive and registry.

Back to the INDEX.




Unable to read workbench state. Workbench UI layout will be reset.
Updated: 06/01/12
Version: 7.0.1.8
This error occurred starting the Eclipse client, which had been previously running without this error.
The cause is unknown.
Acknowledge the message, close the current client, and start a new client.

Back to the INDEX.




ERROR : CRMMD1717E The package file "C:\Rational\ClearQuest\packages\EmailPlus\2.0\package.ini" is missing. Either the corresponding package is not installed or it is corrupted.
Updated: 07/30/12
Version: 7.1.2.4
This error occurred when attempting to push schema changes to a test database. The error occurred after upgrading from 7.0 to 7.1.
The error only occurs for packages that were manually installed into the schema. The problem is that the registered package location has changed.
The solution is to copy the offending package to the new CQ home: ...\IBM\RationalSDLC\ClearQuest\packages and register it there. The following command is specific to the case that caused the error, but can be easily configured for your error.
	packageutil registerpackage EmailPlus 2.0 "C:\Program Files\IBM\RationalSDLC\ClearQuest\packages\EmailPlus\2.0"
Back to the INDEX.


java.lang.IllegalArgumentException: No enum const class com.ibm.rational.wvcm.stp.cq.CqQuery$Filter$Oprationa.boolean equals(java.lang.Object)
Updated: 12/05/12
Version: 7.1.2
This error occurred when attempting a preview of a BIRT report.
The problem is that the query associated with the data set has a dynamic filter. That is, the query prompts for a value. BIRT reports are suppose to handle dynamic query filters, but it's unknown how that's accomplished.

Back to the INDEX.




Column binding "field" has referred to a data set column "field" which does not exist.
Updated: 12/20/12
Version: 7.1.2
This error occurred when previewing a BIRT report layout. An element in the layout is referring to field, but the query that is bound to the report is not returning that field.
Either remove the reference to that field from the layout, or add that field to the query's display fields.

Back to the INDEX.




cleartool: Error: There are no credentials registered for database db.
Updated: 07/09/13
Version: 7.1.2
This error occurred when attempting setactivity inside a scheduled job while the job's owner was logged out. This error was accompanied by a request for the user to run the crmregister command.
The crmregister command was known to be correct, as the scheduled job would run without errors while the user was logged in.
To ensure a scheduled job gets run with the correct privs while logged out, perform the following steps.

1) Open the Control Panel -> Administrative Tools -> Component Services.
2) Navigate to Component Services -> Computers -> My Computer.
3) Right-click to get the Properties and go to the "COM Security" tab.
4) Perform the following steps for both the "Access Permissions" and "Launch and Activation Permissions".
   a) Click on Edit Default.
   b) Click Add.
   c) Add the user that needs to run the scheduled job while logged out.
   d) Ensure Local Access and Remote Access are checked.
5) Close Component Services.

Back to the INDEX.




CRVAP0226E Internal error: Status 500; Condition null; Message: CRVSV0072E server error occurred while processing request.
Updated: 04/10/14
Version: 7.1.2
This error occurred when clicking on the "View Change Set" button in the "Unified Change Management" tab of a record while in CQ web.
Since change set information is actually stored in a CC activity, the CC full client must be installed on the CQ web server to able to view change set information from the web interface. That is, the CQ web server needs access to the PVOB where the activity is.

Back to the INDEX.




CRVAP0075E After completion of the 'CqWsDeliverChangeContext' operation, not all resources were delivered as requested.
Updated: 05/27/14
Version: 7.1.2
This error occurred when clicking OK to save a record opened as New from within CC.
IBM states that it is a temporary error and asks you to just try clicking OK again. If that fails, restart CQ and try again.
Howver, in this case, clicking the Details button revealed a field validation hook error message. Resolving the cause of the validation error fixed the problem. So, the real error was inside the Details.
If you don't want users submitting CQ tickets from within CC, that policy can be changed in the UCM project's Policies dialog.

Back to the INDEX.




CRVAP0075E The operation could not be completed because of a licensing error: Cannot read data from license server.
FLEXlm error: -16,287. System Error: 22 "Invalid argument"

Updated: 06/17/16
Version: 2003
This error occurred in an isolated lab when attempting CC/CQ integration actions such as checkout or describe activity. Because most servers were running Solaris 9, could not upgrade to 7.1.2. Because of a serious CQ bug in 7.0, workstations were reverted to 2003. However, some workstations were running Solaris 10.
CQ 2003 is incompatible with Solaris 10. Upgrading to 7.0.1 solved the problem.
This can also be caused by having incompatible cqdb_registry entries. Ensure there are only entries in "/opt/rational/clearquest/CQDB_rgys/cqdb_registry" that you need/recognize. Use "cqreg drop_dbset" to remove unwanted ones.
Also for unknown reasons, if installs and cqreg commands are run multiple times, you can wind up with multiple embedded cqreg entries, like "2003.06.00", "7.0.0", "2003.06.00/2003.06.00", and "2003.06.00/7.0.0". Do a drop_dbset on all of them to clean up. Ensure "/opt/rational/clearquest/CQDB_rgys/cqdb_registry" is empty, then run "cqreg initialize" and "cqreg add_dbset" to create the proper one.
This error was also encountered on a workstation running CQ 2003 and Solaris 9. Upgrading CQ to 7.0.1 solved the problem. Unfortunately, CQ 7.0.1 on Solaris 9 sometimes has problems rendering forms properly, which can be a pain for users.

Back to the INDEX.




Cannot create ActiveX component
Updated: 06/22/16
Version: 7.1.2.14
This error occurred when accessing a web page in a custom website on Windows that accesses CQ.

Solution 1
The problem was that when CQ was installed, apparently the DLLs didn't get registered properly.
1) Open a command prompt (Run as Administrator).
2) Change to the cq-home folder, such as C:\IBMRational\RationalSDLC\ClearQuest.
	rereg.bat -i -s
Solution 2
The problem was that CQ is a 32-bit application and it was running on a 64-bit system.
IIS was running the VB aspx scripts with the default csscript.exe, which then throws the error when it attempts CreateObject("CLEARQUEST.SESSION").
1) In IIS, click on Application Pools.
2) Right-click on the pool name and select Advanced Settings.
3) Change "Enable 32-Bit Applications" to True.

Back to the INDEX.




unknown exception from CQResultSet_Execute in CQPerlExt
Updated: 07/11/16
Version: 7.0
This error occurred when a query within a Perl script on Unix. The error occurred when attempting $resultSet->Execute().
The script had worked previously.

This error was intermittent on various workstations. The problem was narrowed down to the size of the result set. The results were attempting to return ~100K records. If a filter was placed in the same query, the results would be returned consistently without error. It's not clear if the error was related to memory issues, a bug in 7.0, or other cause.

Back to the INDEX.




Unable to refresh since artifact was not found
Updated: 03/01/18
Version: 8.0.1.14
This error occurred when attempting to click on a record in a query result.

Scenario 1:
This occurred in the Eclipse CQ interface. The session has timed out, even though it hasn't told you that. Rerunning the query will prompt you to refresh your logon. Or, you can disconnect and reconnect to the database.

Scenario 2:
An attempt was made to use SQL to clear a value from a REFERENCE field. Simply setting the field's value to NULL using SQL will result in this error. Using SQL to replace the dbid that was previously in that column will fix the issue. I don't know how to properly clear a REFERENCE field using SQL.

Back to the INDEX.




CRMCU0018E Failed condition: groups_fielddef_dbid != ((AdId_t)0)
Updated: 09/12/16
Version: 8.0.1.09
This error occurred when attempting to update a user database from the ClearQuest User Administration tool.

Cause and solution unknown yet.

Back to the INDEX.




Object doesn't support this property or method: 'fieldname'
Updated: 09/19/16
Version: 8.0.1.09
This error occurred when attempting to submit a record after update a field's vbscript validation hook.

You cannot name variables the same as fieldnames in vbscript.

Back to the INDEX.




Error from the RPC server: The attempt to save the record failed; got CRMDB0001E SQLExecDirect: RETCODE=-1, State=S1000, Native Error=-289
SQL statement="insert into history ...

Updated: 01/20/17
Version: 8.0.1.09
This error occurred when attempting to save any record.

The solution suggested by IBM Support is that the DB2 database is out of space.

Back to the INDEX.




CQUADMT1.DBGLOBAL IS AN UNDEFINED NAME
SQL statement="select count(*) from dbglobal where working_master=site_uuid"

Updated: 04/19/17
Version: 8.0.1.13
This error occurred when attempting to logically delete a database from the schema repository.

The physical database was removed before logically removing it from the schema repo. Use pdsql to execute the following SQL command, where "SAMPL" is the logical name for your database.
	update master_dbs set is_deleted = 1 where name = 'SAMPL';
  
Back to the INDEX.


Schema repository does not contain any user database
Updated: 04/28/17
Version: 8.0.1.13
This error occurred when attempting to log in using the web interface.

If you had a previously functioning system, the cause may be that there are now no user databases with the "production" property. For some quirky reason, the old ClearQuest for Windows client and the web interface must see at least one "production" database to make the connection. If you designate any database, even a dummy one as "production", then you'll be able to connect to the other "test" databases without error.
Note that the new Eclipse interface doesn't have that quirky requirement.

Back to the INDEX.




Unknown error: SQLCODE -1773 State: S1000Native: -1773
Updated: 08/23/17
Version: 8.0.1.14
This error occurred when attempting run a select statement using pdsql in a standby database.

The database in question was a standby database, which is a copy using DB2 HADR (high availability disaster recovery). All select statements must be appended with "with UR", which stands for "uncommitted read".
	# This throws the error.
	select * from master_dbs;

	# This is the correct select statement in the standby database.
	select * from master_dbs with UR;
Back to the INDEX.


CRMUD0070E There is no value available for the field fieldname
Updated: 10/04/17
Version: 8.0.1.00
This error occurred when attempting save a record after the Modify action. Actually, the error was occurring for different field names for different actions.

The reason for the error is unknown. However, the action was being performed on a computer running the unpatched 8.0.1.00 version. The error did not occur on a computer running the fully patched version 8.0.1.14.

Back to the INDEX.




CRMDB0001E SQLExecute: RETCODE=-1, State=22001, Native Error=-302 ... Character data, right truncation occurred ... SQL statement="insert into ...
Updated: 03/29/18
Version: 8.0.1.14
This error occurred when attempting to create a record in a DB2 datbase using the API.
The data was being pulled from a company database of people. The user's name had a long string of strange, non-printable characters embedded in it. The temporary solution was to strip the non-printable characters from the name before import. The permanent solution was to contact HR and them fix that user's fullname in their system.

The full error message was:
CRMDB0001E SQLExecute: RETCODE=-1, State=22001, Native Error=-302 [IBM Rational DataDirect][ODBC DB2 Wire Protocol driver][UDB DB2 for Windows, UNIX, and Linux]Character data, right truncation occurred; for example, an update or insert value is a string that is too long for the column, or a datetime value cannot be assigned to a host variable, because it is too sm all. SQL statement="insert into ...

Back to the INDEX.




AdMasterUserDb[10000], total bytes=28320000 @ 0x00000000
Failed to create the Java Virtual Machine

Updated: 07/31/18
Version: 9.0.1.3
This error occurred when attempting to access a schema repository in the Eclipse-based designer.
The memory allocated to Java for the designer is specified in "C:\Program Files\IBM\RationalSDLC\ClearQuest\Designer\cqdesigner.ini". If the value of "-Xmx=512m" is reset to anything higher than 1400, starting the designer results in the error Java Virtual Machine error. If the value is set to 1400, the designer will launch, but will unable to access any schema repo because of the AdMasterUserDb error, which is presumably due to not enough memory allocated.
Previously when working with CQ 8.0.1.x, that value had to set to around 1900 to get the designer to work with our very large schema. They must have changed some things because now it doesn't work if you set that value that high. Resetting it to "-Xmx=756m" solved the problem.

Back to the INDEX.




CRMDB0000E SQLDriverConnect: RETCODE=-1, State=S1000, Native Error=-1060 [IBM Rational DataDirect][ODBC DB2 Wire Protocol driver][DB2]Unknown error: SQLCODE -1060
Updated: 08/08/18
Version: 8.0.1.14
This error occurred when attempting to access a database after restoring it from backup after a failed feature level upgrade attempt.
The cause is unknown, but may be related to https://www-01.ibm.com/support/docview.wss?uid=swg21289255.

Back to the INDEX.




CRMUD0030E No history will be recorded for this modification. Update history.expired_timestamp field FAILED.
Updated: 01/14/19
Version: 9.0.1.4
This error occurred when attempting modify an old record.
This error usually means that the latest history entry for the record probably already has an expired_timestamp entry, which is usually only added when the next history entry is added. If history is removed, the expired_timestamp is not cleared for the last remaining history entry. This limitation is supposedly rectified in CQ feature level 9.

Back to the INDEX.




ActiveX component can't create object: 'CDO.Message'
Updated: 01/28/19
Version: 9.0.1.4
This error occurred when updating a record from a Windows scheduled task. Errors involving 'Scripting.Dictionary' were also seen in the logs.
The error only occurs periodically. That is, the external script run by the task and the internal schema code function as expected most of the time. The task was exported from the Scheduled Tasks utility on an older server and imported on a server running a newer operating system. The true root cause is unknown, but importing a task into a different operating system can cause environmental issues.
The safest thing to do is to manually re-create the task definition on the new server.

Back to the INDEX.




CRMCU0019E Failed condition: GetState() == AD_VALIDATED
Updated: 05/31/19
Version: 9.0.1.4
This error occurred when updating a record. The current record's action validation hook was updating a secondary record.
The error occurred because there was no error trapping around the "entityObject.Validate" method. The validation failed, but the code then moved on to the "entityObject.Commit". You cannot commit a record that has not passed validation.

Back to the INDEX.




Function sequence error
Updated: 06/03/20
Version: 9.0.1.4
This error occurred when running query in the schema.
The exact cause of the error is still unknown, but there is a solution/workaround. The query itself was not timing out. That is, the execution of the query completed returning a result set. However, while looping through the query results, the system was a taking a very long time to complete and somehow timeing out. The "Function sequence error" apparently means the result set is no longer valid, yet being consulted with something like resultset.GetColumnValue(1).
The error no longer occurred after saving the query results in a dictionary so that the query is completely finished, and then loop through the entries in the dictionary.

Back to the INDEX.


ejostrander@cox.net
Return to the home page .

This page last modified: 08/27/2020