Derby
  1. Derby
  2. DERBY-4986

Document the new property derby.stream.error.extendedDiagSeverityLevel for thread dump and extend diagnostic information

    Details

    • Type: Task Task
    • Status: Closed
    • Priority: Minor Minor
    • Resolution: Fixed
    • Affects Version/s: 10.7.1.1
    • Fix Version/s: 10.8.1.2
    • Component/s: Documentation
    • Labels:
      None

      Description

      With new behavior on thread dump and extended diagnostic information, it is needed to document the new property. Please refer to DERBY-4856.

      1. threaddump.txt
        3 kB
        Lily Wei
      2. javacore.20110128.085450.3872.0001.txt
        252 kB
        Lily Wei
      3. DERBY-4986.zip
        7 kB
        Kim Haase
      4. DERBY-4986.stat
        0.1 kB
        Kim Haase
      5. DERBY-4986.diff
        6 kB
        Kim Haase
      6. derby.log
        14 kB
        Lily Wei

        Issue Links

          Activity

          Lily Wei created issue -
          Lily Wei made changes -
          Field Original Value New Value
          Link This issue is related to DERBY-4856 [ DERBY-4856 ]
          Kim Haase made changes -
          Assignee Kim Haase [ chaase3 ]
          Hide
          Kim Haase added a comment -

          In order to document this new property, I'll need information similar to what is provided for other properties, such as the other derby.stream.error ones:

          http://db.apache.org/derby/docs/dev/ref/rrefproper33027.html
          http://db.apache.org/derby/docs/dev/ref/rrefproper18151.html
          http://db.apache.org/derby/docs/dev/ref/rrefproper35028.html
          http://db.apache.org/derby/docs/dev/ref/rrefproper26985.html

          There is some information that I can't find in DERBY-4856 (though it may be there – there's a lot of fine print). I don't see a release note or a functional spec – does either exist?

          1) What does the property do, exactly, if set? There was some discussion of whether the thread dump goes into a separate file or into derby.log. What was the final decision?

          2) What are the levels that the user can set, and what does each of them do?

          3) What is the default value?

          4) I would guess that this is a static property like the other ones and that it is system-wide in scope – is that correct?

          Thanks very much.

          Show
          Kim Haase added a comment - In order to document this new property, I'll need information similar to what is provided for other properties, such as the other derby.stream.error ones: http://db.apache.org/derby/docs/dev/ref/rrefproper33027.html http://db.apache.org/derby/docs/dev/ref/rrefproper18151.html http://db.apache.org/derby/docs/dev/ref/rrefproper35028.html http://db.apache.org/derby/docs/dev/ref/rrefproper26985.html There is some information that I can't find in DERBY-4856 (though it may be there – there's a lot of fine print). I don't see a release note or a functional spec – does either exist? 1) What does the property do, exactly, if set? There was some discussion of whether the thread dump goes into a separate file or into derby.log. What was the final decision? 2) What are the levels that the user can set, and what does each of them do? 3) What is the default value? 4) I would guess that this is a static property like the other ones and that it is system-wide in scope – is that correct? Thanks very much.
          Hide
          Lily Wei added a comment -

          Thanks Kim for the feedback:
          The property is a flag that decide whether thread dump information and extend diagnostic information get created. This is the system wide flag. The default value is 40000(ExceptionSeverity.SESSION_SEVERITY) User can set it to
          WARNING_SEVERITY = 10000,
          STATEMENT_SEVERITY = 20000,
          TRANSACTION_SEVERITY = 30000,
          SESSION_SEVERITY = 40000,
          DATABASE_SEVERITY = 45000,
          SYSTEM_SEVERITY = 50000

          If they use the default value, the thread dump information and diagnostic information will only get print to derby.log when error severity is equal or greater than SESSION_SEVERITY. If they set it to number that is greater than 50000, no information will be print to derby.log i.e. derby.stream.error.extendedDiagSeverityLevel=30000

          You are right about this is a system wide flag. I am in the middle of preparing the releaseNote.html. I am not sure the information is clear enough. Kathey is helping me reviewing it. I am sorry for not provide enough information for documentation.

          Show
          Lily Wei added a comment - Thanks Kim for the feedback: The property is a flag that decide whether thread dump information and extend diagnostic information get created. This is the system wide flag. The default value is 40000(ExceptionSeverity.SESSION_SEVERITY) User can set it to WARNING_SEVERITY = 10000, STATEMENT_SEVERITY = 20000, TRANSACTION_SEVERITY = 30000, SESSION_SEVERITY = 40000, DATABASE_SEVERITY = 45000, SYSTEM_SEVERITY = 50000 If they use the default value, the thread dump information and diagnostic information will only get print to derby.log when error severity is equal or greater than SESSION_SEVERITY. If they set it to number that is greater than 50000, no information will be print to derby.log i.e. derby.stream.error.extendedDiagSeverityLevel=30000 You are right about this is a system wide flag. I am in the middle of preparing the releaseNote.html. I am not sure the information is clear enough. Kathey is helping me reviewing it. I am sorry for not provide enough information for documentation.
          Hide
          Kim Haase added a comment -

          Thanks so much, Lily – this is great and gives me plenty to get started with. I will be interested in seeing the release note when it is done, in case more information should be provided to users.

          Show
          Kim Haase added a comment - Thanks so much, Lily – this is great and gives me plenty to get started with. I will be interested in seeing the release note when it is done, in case more information should be provided to users.
          Hide
          Lily Wei added a comment -

          Thanks Kim. I put up the releaseNote.html for DERBY-4856. If you need a sample of the derby.log, please just let me know. I will create them.

          Show
          Lily Wei added a comment - Thanks Kim. I put up the releaseNote.html for DERBY-4856 . If you need a sample of the derby.log, please just let me know. I will create them.
          Hide
          Kim Haase added a comment -

          Thanks very much, Lily. Reading the release note, I have a couple of questions about the difference in behavior between the Oracle and IBM JVMs.

          "Oracle jvm will print thread dump information with error severity higher than session error and IBM jvm will print out thread dump information and a javacore."

          Does this mean that the Oracle JVM pays attention to the severity levels the user sets, but the IBM JVM does not? Or does "with error severity higher than session error" apply to both JVMs?

          Also,

          "Users can find thread dump information on derby.log and users of ibm jvm will have more diagostic information in the javacore* file."

          Does this mean that the contents of derby.log are the same for both the Oracle and IBM JVMs, but that IBM users will see additional information in the javacore file?

          Maybe it would be helpful if you attached a sample of the derby.log, too.

          Thanks again!

          Show
          Kim Haase added a comment - Thanks very much, Lily. Reading the release note, I have a couple of questions about the difference in behavior between the Oracle and IBM JVMs. "Oracle jvm will print thread dump information with error severity higher than session error and IBM jvm will print out thread dump information and a javacore." Does this mean that the Oracle JVM pays attention to the severity levels the user sets, but the IBM JVM does not? Or does "with error severity higher than session error" apply to both JVMs? Also, "Users can find thread dump information on derby.log and users of ibm jvm will have more diagostic information in the javacore* file." Does this mean that the contents of derby.log are the same for both the Oracle and IBM JVMs, but that IBM users will see additional information in the javacore file? Maybe it would be helpful if you attached a sample of the derby.log, too. Thanks again!
          Hide
          Lily Wei added a comment -

          Thanks Kim.

          Oracle jvm and IBM jvm will both pay attention to the severity levels and print thread dump information to derby.log. IBM users will see addtional information in the javacore file.

          I include three files in the attachment.
          1. derby.log that include thread dump information
          2. threaddump.txt Extract the thread dump information from derby.log.
          3. javacore.20110128.085450.3872.0001.txt The javacore file created by ibm jvm.

          Show
          Lily Wei added a comment - Thanks Kim. Oracle jvm and IBM jvm will both pay attention to the severity levels and print thread dump information to derby.log. IBM users will see addtional information in the javacore file. I include three files in the attachment. 1. derby.log that include thread dump information 2. threaddump.txt Extract the thread dump information from derby.log. 3. javacore.20110128.085450.3872.0001.txt The javacore file created by ibm jvm.
          Lily Wei made changes -
          Attachment javacore.20110128.085450.3872.0001.txt [ 12469689 ]
          Attachment threaddump.txt [ 12469690 ]
          Attachment derby.log [ 12469691 ]
          Hide
          Kim Haase added a comment -

          Thanks for the additional information, Lily.

          Attaching DERBY-4986.diff, DERBY-4986.stat, and DERBY-4986.zip, with the following changes:

          M src/ref/crefproper22250.dita
          A src/ref/rrefproperextdiagsevlevel.dita
          M src/ref/refderby.ditamap

          The patch adds a new topic on the property to the reference manual and the map file, and lists it in the overview topic on Derby properties. It also fixes the alphabetical order of two other derby.stream.error properties.

          I didn't add details on how to set the property, since the information is the same for all properties and is provided in the overview topic (crefproper22250.dita).

          Please let me know what changes are needed.

          Show
          Kim Haase added a comment - Thanks for the additional information, Lily. Attaching DERBY-4986 .diff, DERBY-4986 .stat, and DERBY-4986 .zip, with the following changes: M src/ref/crefproper22250.dita A src/ref/rrefproperextdiagsevlevel.dita M src/ref/refderby.ditamap The patch adds a new topic on the property to the reference manual and the map file, and lists it in the overview topic on Derby properties. It also fixes the alphabetical order of two other derby.stream.error properties. I didn't add details on how to set the property, since the information is the same for all properties and is provided in the overview topic (crefproper22250.dita). Please let me know what changes are needed.
          Kim Haase made changes -
          Attachment DERBY-4986.diff [ 12469703 ]
          Attachment DERBY-4986.stat [ 12469704 ]
          Attachment DERBY-4986.zip [ 12469705 ]
          Kim Haase made changes -
          Issue & fix info [Patch Available]
          Hide
          Lily Wei added a comment -

          +1 for the DERBY-4986 patch

          Show
          Lily Wei added a comment - +1 for the DERBY-4986 patch
          Hide
          Kim Haase added a comment -

          Thanks very much, Lily.

          Committed patch DERBY-4986.diff to docs trunk at revision 1066111.

          Show
          Kim Haase added a comment - Thanks very much, Lily. Committed patch DERBY-4986 .diff to docs trunk at revision 1066111.
          Kim Haase made changes -
          Status Open [ 1 ] Resolved [ 5 ]
          Issue & fix info [Patch Available]
          Resolution Fixed [ 1 ]
          Hide
          Kim Haase added a comment -

          Closing, since change appears in latest alpha docs.

          Show
          Kim Haase added a comment - Closing, since change appears in latest alpha docs.
          Kim Haase made changes -
          Status Resolved [ 5 ] Closed [ 6 ]
          Hide
          Kim Haase added a comment -

          For some reason, the new file src/ref/rrefproperextdiagsevlevel.dita was committed without its svn:eol-style set to native. This needs fixing.

          Show
          Kim Haase added a comment - For some reason, the new file src/ref/rrefproperextdiagsevlevel.dita was committed without its svn:eol-style set to native. This needs fixing.
          Kim Haase made changes -
          Resolution Fixed [ 1 ]
          Status Closed [ 6 ] Reopened [ 4 ]
          Hide
          Kim Haase added a comment -

          Committed new svn:eol-style = native property setting for topic at revision 1076360.

          Show
          Kim Haase added a comment - Committed new svn:eol-style = native property setting for topic at revision 1076360.
          Kim Haase made changes -
          Status Reopened [ 4 ] Resolved [ 5 ]
          Resolution Fixed [ 1 ]
          Rick Hillegas made changes -
          Fix Version/s 10.8.1.1 [ 12316356 ]
          Fix Version/s 10.8.1.0 [ 12315561 ]
          Rick Hillegas made changes -
          Fix Version/s 10.8.1.2 [ 12316362 ]
          Fix Version/s 10.8.1.1 [ 12316356 ]
          Hide
          Kim Haase added a comment -

          Changes appeared in 10.8.1 documentation, so closing.

          Show
          Kim Haase added a comment - Changes appeared in 10.8.1 documentation, so closing.
          Kim Haase made changes -
          Status Resolved [ 5 ] Closed [ 6 ]
          Gavin made changes -
          Workflow jira [ 12543765 ] Default workflow, editable Closed status [ 12797770 ]
          Transition Time In Source Status Execution Times Last Executer Last Execution Date
          Open Open Resolved Resolved
          5d 23h 6m 1 Kim Haase 01/Feb/11 18:18
          Closed Closed Reopened Reopened
          19d 3h 54m 1 Kim Haase 02/Mar/11 19:56
          Reopened Reopened Resolved Resolved
          18m 36s 1 Kim Haase 02/Mar/11 20:15
          Resolved Resolved Closed Closed
          255d 19h 55m 2 Kim Haase 03/Nov/11 18:27

            People

            • Assignee:
              Kim Haase
              Reporter:
              Lily Wei
            • Votes:
              0 Vote for this issue
              Watchers:
              0 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved:

                Development