Software Status Bulletin

NDS Migration     ChemStoreToECM

Preface
-------------------------
This Software Status Bulletin (SSB) documents all known problems in the software
product designated above. The SSB is derived from Known Problem Reports (KPR)
which result from user problems that have been classified as documentation
problems or software defects. When a KPR is written, an identifying number is
assigned to it, and the KPR is added to the next edition of the SSB.

User inputs that have been classified as Enhancement Requests are not documented
in the SSB. User problems that have been submitted, but that have not been
classified by the time the SSB is generated are not included in the SSB.

How to use the SSB
-------------------------
When you experience a problem with a product, first check this SSB to see if the
problem has been reported already, and if there is a temporary workaround
available for the problem, or if the problem has already been fixed by a new
revision. If the problem is not listed in this SSB then you may wish to report
it to the Response Center or to your field support representative.

To determine if your problem is documented in this SSB, first look in the
Keyword Index section of the SSB. Under each keyword is a listing of one-line
descriptions of related KPRs. If any of these sound like yours, locate the KPR
# in the Known Problem Reports section of the SSB, and read the full KPR. The
KPRs in the Known Problem Reports section are sorted by KPR #.

There are two sections in the SSB:
Keyword Index: This index is categorized by keyword. For each KPR there is a
brief description and a KPR #. A KPR may be associated with more than one
keyword.

Known Problem Reports: This section contains KPRs, with all the available
information relevant to the problem. KPRs in this section are sorted by KPR #.

Keyword Glossary

Keyword Index

Keyword: Data Extractor

KPR#:23 Wrong result version counting causing extraction failure
KPR#:25 Blank injection runs of ChemStore legacy data cannot be seqeunce reprocessed

Keyword: 

One-line Description:

Windows Vista: COM call factory error when switching to details view of data viewer

Problem:

On Windows Vista operating systems the user may face a COM call factory error when switching to details view of data viewer rev. A.1.2.5.0.
Windows XP operating systems are not affected.

Temporary Solution:

Use the "Details" tab to switch to the details view.

Fix Information:

This defect is no longer observed in rev. 1.3.0 of the Data Viewer software.

Keyword: 

One-line Description:

Compound amounts may appear multiple times in the compound table of the overview panel

Problem:

The overview panel of the analytical data viewer gives direct access to the results table of each run listed in hte overview table.
For multi-signal methods, each quantified compound result appears mulitple times, equivalent to the number of signals.

Temporary Solution:

This defect is only related to the display and the displayed results are consistent and correct. Alternatively, the "Identified Compounds" nodes of the detail view shows each compound onyl one time per signal.

Fix Information:

This defect is fixed in revision 1.2.0.6

Keyword: 

One-line Description:

The overview panel does not populate various columns

Problem:

The overview panel of data viewer rev. A.1.2.5.0  does not populate the columns:

Injection last modified by
Injection last modified date
Instrument

Temporary Solution:

The missing information is visible in the details view.

Fix Information:

This defect is fixed in revision 1.2.0.6

Keyword: 

One-line Description:

Revision A.1.2.5.0 of the analytical data viewer does not support custom views

Problem:

Revision A.1.2.5.0 of the analytical data viewer does not allow setting up specific custom views in the overview panel.
Adding additional columns from the column chooser and re-arranging the columns and grouping is possible.

Temporary Solution:

n/a

Fix Information:

This defect is fixed in revision 1.2.0.6

Keyword: 

One-line Description:

Series of error messages when zooming chromatograms with no baseline information

Problem:

Whenever zooming a chromatogram of a ChemStore legacy dataset, the Analytical Data Viewer rev. 1.2.5.0 shows a series of error messages. The number of error messages depends of the number of peaks shown in the chromatogram.
The default display setting of the chromatogram viewer includes the display of the baseline and peak start and end times. ChemStore legacy data do not contain this information and the baseline can only be viewed by showing the CHR files stored in the data container.

Temporary Solution:

Do not use the zoom function of the chromatogram view, but open the CHR files stored under the Signal and results > files tab of the detail view.

Fix Information:

This defect does not affect rev. 1.3.0.

Keyword: 

One-line Description:

Refreshing the Data Viewer windows closes the currently loaded file

Problem:

Using the Refresh function of the Analytical Data Viewer rev. A.1.2.5.0. closes the currently loaded file.

Temporary Solution:

The data viewer does not allow modification of data and changes cannot get lost when closing an ACAML file. On error ocurrance, the ACAML file needs to be reloaded.  
The refresh function is not necessary, as changes cannot be perfromed. In order to reset the displayed columns, the 'Restore Grid Layout' menu should be used.

Fix Information:

This defect if fixed in rev. 1.3.0

Keyword: 

One-line Description:

Data viewer cannot open ACAML files generated with uncalibrated methods

Problem:

ACAML files with no calibrated compounds fail to open with the Analytical data viewer rev. A.1.2.5.0.

Temporary Solution:

The sequence container should be reloaded to ChemStation for reviewing the results.

Fix Information:

This defect is fixed in revision 1.2.0.6

Keyword: 

One-line Description:

ChemStore Admin tools cannot connect to DB with changed CSINTERNAL password

Problem:

After changing the ORACLE password of user CSINTERNAL the ChemStore Admin Tools cannot connect to database any longer.

Temporary Solution:

The PW change needs to be done by the Admin Client, as described in the manual.

Fix Information:

There is no fix planned

Keyword: 

One-line Description:

Series of error messages when displaying non-available baseline  of ChemStore legacy chromatograms

Problem:

Whenever turning display the integration baseline of a ChemStore legacy dataset on, the Analytical Data Viewer rev. 1.2.0.6 shows a series of error messages. The number of error messages depends of the number of peaks shown in the chromatogram.
The default display setting of the chromatogram viewer does not include the display of the integration baseline. ChemStore legacy data do not contain this information and the baseline can only be viewed by showing the CHR files stored in the data container.

Temporary Solution:

n/a

Fix Information:

Do not use the “show baseline” function of the chromatogram view, but open the CHR files stored under the Signal and results > files tab of the detail view. 

This defect is planned to be fixed in a future revision of the analytical data viewer. However, ChemStore legacy chromatograms will still need to be viewed under the Signal and results > files tab of the detail view

Keyword: 

One-line Description:

Missing token 'EndElement' error in folder archiver logfile

Problem:

In many cases the folder archiver process reports an error message  at the end of the folder archiver log file:

11/3/2008 7:43:49 PM: 
Error occurred: System.Xml.XmlException: This is an unexpected token. 
The expected token is 'EndElement'. 

The error message is caused by a missing end tag in the CSXTFileList.xml file. The end tag is missing in cases where the verification injection number is different than number of extracted injections.

There a various reasons for failing extractions of individual injections, and these should be subject of a separate deviation form.

Temporary Solution:

The missing token 'EndElement' error can be safely ignored and a printout of this KPR should be attached as a deviation report to the data migration compliance protocol.

Fix Information:

Fixed in rev. 4.3 of the extraction. The file needs to be closed interactivley by using a new menu item.

Keyword: 

One-line Description:

Blank runs acquired with calibrated methods cause verification failures

Problem:

Database extractions performed with extraction tool revision 4.2 may show the following validation error:


Number of Injections does not match (expected: 1, found: 0).
Number of Runs does not match (expected: 1, found: 0).
Number of Chromatograms does not match (expected: 1, found: 0).
Number of Peaks does not match (expected: 3, found: 0).

It affects all runs that do have in fact no peaks, but the analysis method was a calibrated one. 

All the runs were special kind of blank runs.

The ChemStation offers 2 choices:

- an instrument blank run: no vial is specified, the sample type is set to 'blank' and the instrument just runs the standard analysis cycle with no injection

- a method blank run:  A known clean sample is treated in the same way (sample preparation, etc.) as any other sample and at the end the vial is put on the tray and gets analysed like any other sample. The ChemStation assigns a sample type "sample", because a vial number was specified.
If the whole sample preparation process is clean, the chromatogram should not contain any peak.

The 2nd way of operation is very common for e.g. drug screening, or environmental samples.
The analysis are done with calibrated methods, but in most of the samples no peaks are found.

Temporary Solution:

Attach a printout of this KPR to a deviation report listing the affected runs and sequences.

Fix Information:

There is currently no fix planned.

Keyword: 

One-line Description:

Verification error "The key sequence 'XYZ' in Keyref fails to refer to some key."

Problem:

Occasionally the extraction logfiles shows error messages like 
"The key sequence 'XYZ' in Keyref fails to refer to some key"

Temporary Solution:

This errors are associated with a reference to an invalid or orphaned object. Extracted results from this object should be considered suspect. 

Typical affected areas are: Data sets acquired with peak grouping calibrations, acquisition only runs, but also damaged and restored archive units.

One possible attempt to recover the data is a manual extraction of the raw channel data from ChemStore using the review client.
Operators may reprocess this data to re-generate results.

Fix Information:

Revision 4.4.55. ChemStore Extraction Tool introduces an updated worklist creator to generated a OpenLAB  CDS compatible sequence file.

Keyword: 

One-line Description:

Extraction verification sporadically fails on Signal ID

Problem:

Occasionally the export log file shows error messages related to the signal type of the chromatogram, like:

Different content: 
CS_CG.CG_SigTSID differs: should be "Unknown", is "Flame Ionization"

It is suspected that data acquired in "acquisition only" mode are the cause of this problem, as the chromatogram type on the report is  if data analysis was not performed in result revision 1.

If data got reprocessed, the information got added for higher result version.

Temporary Solution:

Generate a deviation report listing all affected sequences and attach a printout of this KPR to the deviation report.

Fix Information:

Fixed in revision 4.4.55 of ChemStore Extraction Tool.

Keyword: 

One-line Description:

Export validation fails sporadically on instrument configuration missmatch

Problem:

Sporadically it may happen that the extraction of sequence related information may cause an error in the extraction logfile, like 
 Different content: CS_INJ.IC_ID differs: should be 2248046, is "2260380".

The error message says that the sequence got acquired on 2 different instruments at a time, which seems to be an artefact caused by forced abortion of acquisition runs.

Temporary Solution:

Generate a deviation report listing all affected sequences and attach a printout of this KPR to the deviation report.

Fix Information:

This defect is fixed in revision 4.4.55 of the extraction tool.

Keyword: 

One-line Description:

The "Seq.Name(MeasData)" column should be labelled 'Sequence comment'

Problem:

The "Seq.Name(MeasData)" column of the overview table contains the sequence comment and it is labelled  incorrectly.

Temporary Solution:

n/a

Fix Information:

This defect is fixed in revision 1.3.2. of the Data Viewer.

Keyword: 

One-line Description:

‘Invalid XML format’ error when loading a file

Problem:

Users of ACAML viewer 1.3.0.2 may experience an error like: ‘Invalid XML format. Method not found:’Boolean COMPOUNDSANDPEAKSRow.IsSAMPLE_APPROVAL_STatusNull() ’

The problem is typically caused by performing an update installation without de-installing a previously installed version, first.

Temporary Solution:

This can be verified by checking the ‘Programs and Features’  panel of the Windows operating systems. If 2 instances of the Agilent Analytical Data viewer are listed, de-install both and re-install rev. 1.3.0.2

Fix Information:

There currently no fix planned.

Keyword: 

One-line Description:

Paused sequences are not always extracted to a single sequence container

Problem:

Certain ChemStation sequences that were interrrupted by intrument hardware errors may not be converted into a single sequence container, but are treated as a separate sequence.

Temporary Solution:

The file structure of the extracted data (Year > Month > Instrument) allows an easy re-allocation of the data.
In cases where the isntrument names have been unique in the lab environment, the 2 sequences appear attached to each other in the instument folder.
For reprocessing the entire sequence, the data can be temporarily merged into one folder.

Fix Information:

This defect is a given limitation of translating a relational database to a file system.
There is currently no fix planned.

Keyword: 

One-line Description:

ACAML  Viewer shows method references pointing to mutiple result versions

Problem:

The details view of the Analytical Data viewer shows a file list pointing to all raw and method data stored in a data set.
Some parts of the Data Analysis method (DA.M) of version X may reference to DA method components stored in the version folder X-1, X-2 or earlier.
This happens if the affected part of the method has not been changed between the result versions X-2 and X. 
The ChemStore database did not store multiple copies of the same method version, but referenced to a single object and the ACAML file reflects these references across mutiple file versions.

Temporary Solution:

The references shown by the ACAML viewer correct, but for the ease of use, the extraction tool generates redundant copies of the releavant parts of each DA method and stores it in each result version specific V000000X subfolder of the .D file.
If the DA method gets loaded from this version specific folder, the integration and calibration parameters are loaded that generated the result version initiallly.
(Comparing results acquired on a 16-bit ChemStation and processed on a 32-bit ChemStation require special considerations. Please consult the G2170-90235 Upgrade preparation guide)

Fix Information:

Revision 1.3.2. of the Data Viewer introduced a Run Information panel similar to the ChemStore Review Client.  The associated method version can be loaded through this run information panel, using a ChemStore like workflow.

Keyword: 

One-line Description:

The ACAML file verification may fail with error: Number of Injections does not match

Problem:

The problem appears in all data sets with multiple sample injections where the same sample got filled into several vials and each vial got analysed once, as opposed to multiple drawings from the same vial. 
This is common practice in GC analysis to prevent quantification errors caused by evaporation of solvents through the punched septum.
The ChemStore database design indirectly forces a one to one relation between sample names and vial numbers, which is not given in the mentioned example if the identical sample name is used for each vial and related sequence line.
Injections and Sample names are stored in two linked database tables, where multiple injections are linked to the same sample name. This separation allows generating sample summary reports across different techniques use for the same sample.
The ACAML design forces a one to one relation where every vial represents a sample and thus one sample can hold only one vial number.

Temporary Solution:

Attach a printout of this KPR entry to the deviation form section of the data migration compliance protocol.

Fix Information:

There is no fix planned currently.

Keyword: 

One-line Description:

Extraction verification fails due to corrupted custom field strings

Problem:

The CSXTverifyLOG.XML may show the following error:

  Cannot load ACAML file: '', hexadecimal value 0x1F, is an invalid character. Line 11597, position 29. 
  

in cases where custom field strings show characters of the extended ASCII code. 
This is likely to be caused by a missmatch of the NLS character set during Oracle export and import or by using illegal characters in ChemStation at the time of data acquisition.

Temporary Solution:

A printout of this KPR should be attached to the deviation section of the migration verification protocol.

Fix Information:

This defect is fixed in revision 4.4.55. These characters are identified upfront and corrected.

Keyword: 

One-line Description:

Incorrect sample information and missing vial numbers in migrated sequence and batch files

Problem:

ChemStore database migrations performed with extraction tool rev. 4.3 use a new mechanism to restore sequence (.S) and batch (.B) files that have not been stored in the ChemStore database, initially. This restore mechanism is especially required for data acquired with ChemStation rev. B.01.03 and all previous releases. The original design of ChemStation Plus did not consider storing batch files, as the sequence and batch information was stored in the database. 

In cases where the sequence contains repeated injections from a single vial (from single or multiple sequence lines)  sample information in the restored .B file is not properly mapped to the samples and wrong sample information and ISTD amounts (if used) get associated with most of the samples. The vial numbers are all listed as "0".

After loading migrated ChemStore data from OpenLAB ECM, the wrong sample information and ISTD amounts get visible in the respective columns of the ChemStation navigation table and the sequence table.

Note:  This defect does not affect reports generated with OpenLAB ECM Intelligent Reporter or the OpenLAB ECM Filter and Search functions.  These functions are based on the sequence.acaml result file including the correct sample information for each injection.
The file information panel of the integration task of ChemStation Data Analysis view does also show the correct sample information stored along with the data file. 

Note: After sequence or batch reprocessing the sample information gets overwritten and the original sample information gets temporarily lost.

Temporary Solution:

The original vial and sample information can be restored by using the original sequence file stored in the HMD directory of the result set:

- Load the result set from ECM
- Navigate to Sequence>Result set migration
In the result set migration panel:

- Select sequence template:  Browse to \chem32\\Data\\HMD and select the sequence file stored in the HMD directory
- Select method path:  Browse to \chem32\\DATA\
- Select source: Browse to \chem32\\DATA\
- Select Destination:  Browse to \chem32\\DATA\ and create a new folder  "_migrated"
- Press start to start the result set migration
- Close the dialog after migration finished successfully 

- Load the freshly created result set
- Select Sequence > Sequence parameters and modify the sequence output/printing options if needed. Close the dialog
- Start sequence reporessing

Fix Information:

This defect is fixed in revision 4.4.55 of the extraction tool. 

A ChemStation C.01.03 hot fix is available for customers operating with affected data. Other ChemStation revisions are subject of further tests and qualifications.
This hot fix introduces a new ChemStore Result Set Migration feature that allow users to convert sequences runs to single runs or prepare the result for reprocessing with original sequence file stored in the historical meta data directory (hmd).

Keyword: 

One-line Description:

Extraction tool can’t handle missing archive units.

Problem:

In cases where an archive unit is lost or where the storage media got damaged, the database consistency cannot be established and the database report shows plenty of expected orphaned records in the injection table.
his might affect all result versions of injection or maybe only a few of the versions associated with an injection.

Temporary Solution:

n/a

Fix Information:

Revision 4.4.55. ChemStore Extraction Tool offers an interface to add all ID’s of missing archive units and a selection of possible reasons:
- archive unit unavailable
- archive unit damaged
Failures during the extraction are linked to those missing archive units and the reason is added to the log file.

Keyword: Data Extractor

One-line Description:

Wrong result version counting causing extraction failure

Problem:

During a ChemStore database extraction it may happen that result version counting of a specific run is starting with  #2, rather than #1. As a consequence, the validation of the extracted results against the original database content fails and the verifiction log files shows error  like:

Different content: CS_RN.RN_Ts differs: should be 2003-09-02T09:45:01, is 2003-08-30T03:55:05.
CS_RN.RN_SqLine differs: should be 0, is 23.
CS_RN.RN_SqRepNum differs: should be 0, is 1.
CS_RD.RD_App differs: should be "Rev. A.09.01 [1206]", is "n.a.".
CS_RD.RD_Hash differs: should be "u6vm3s408mFHTXaLWEVOPg==", is "Hspi57GTbFmvWr9dp3mcsg==".
....

The whole sequence gets marked as "failed" and it separated from the successfully exported and verfied  sequence.

This problem is caused by the unlike case of a hash collision when a result version 1 gets identified as already present in the result set. As a consequence it gets labeled as version #2. The Analytical data viewer shows this specific injection as version 2,3, 4 where it should be 1,2,3. The Audit Trail of this run starts with version 2, and corresponds to the version numbering of the affected run.

The likelyhood increases in case of "data acquisition only" runs, as many ChemStore database fields (e.g. sample acquisition order number) did not get populated in earlier versions of ChemStore.

Temporary Solution:

Whenever a failed result set needs to be reviewed in an audit situation, the verfication failure needs to checked against he deviation reports delivered as part of the migration execution and validation protocol.
A printout of this KPR needs to be added to the deviation report of the affected failure class.

In this specific case, the symptoms are limited to revision counting of a single injection only and no other data are affected.

Fix Information:

This defect is fixed in revision 4.4.55.

Keyword: Data Extractor

One-line Description:

Blank injection runs  of ChemStore legacy data cannot be seqeunce reprocessed

Problem:

ChemStore data extracted from legacy databases in 2013 or earlier will show the following problem when sequence reprocessing blank injection runs:
All blank injections with raw data file names starting with "NV-"  (no vial) will be skipped and no report will be generated.  These blank runs are also not shown in the overview view of the Analytical Data Viewer. However, they appear in the details view of the Analytical data viewer (Sequence.ACAML>Results>Sequence Results>XXX.S>Sample Results)

The problem is caused by a file name cleanup mechanism that is supposed to remove special characters like %,&,/,'  from instrument and user names used to make up the unique result set folder name.
However, this mechanism does also replace any double dash '--' with a single dash "-"  in file names. The auto-naming prefix of blank injections "NV--" is also affected by this change.

Temporary Solution:

Renaming the NV-....... .D   files back to NV--......D can be used as an emergency procedure.

Fix Information:

This defect is fixed in revision 4.4.55 of the extraction tool. 

A ChemStation C.01.03 hot fix is available for customers operating with affected data. Other ChemStation revisions are subject of further tests and qualifications.
This hot fix introduces a new ChemStore Result Set Migration feature that allow users to convert legacy ChemStore sequences to modern type of ChemStation sequences and automatically restores the original file name.