db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Knut Anders Hatlen (JIRA)" <j...@apache.org>
Subject [jira] Commented: (DERBY-4715) Write jvm information and path of derby.jar to derby.log
Date Sun, 11 Jul 2010 09:10:49 GMT

    [ https://issues.apache.org/jira/browse/DERBY-4715?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12887159#action_12887159

Knut Anders Hatlen commented on DERBY-4715:

A couple comments in addition to what Bryan said:

- The methods jarClassPath() and buildJvmVersion() have different indentation than the other

- If formatURL() gets an IOException, it returns the string "IOException". Perhaps it should
also return the exception's message text?

- I don't follow all the logic in formatURL(). Could you add some comments that explain the
various transformations? In my environment, just printing what URL.toString() returns gives
clear enough information:

file:/code/derby/trunk0/classes/ or

- For readability, I think it would be better to separate the properties with a newline character
instead of a space.

- The comments in the policy files say "Add for DERBY-4715". I think it would be better if
they said why the permission was needed since we include them as templates in the distribution
and refer to them from the documentation[1], so they are expected to be read by users. Perhaps
something like "getProtectionDomain is an optional permission needed for printing classpath
information to derby.log."

- In server.policy and template.policy, the new permission is added in the section that is
labeled "These permissions are needed for everyday, embedded Derby usage." My understanding
of the patch is that the new permission is optional, and embedded Derby will work fine without
it, it just won't print the classpath to derby.log. If it's so, I'd suggest moving it to a
less prominent section of the policy files.

- Similarly, in AssertFailureTest.policy, the permission is added under the label "These are
the ones that matter", but AssertFailureTest runs fine without it.

[1] http://db.apache.org/derby/docs/dev/adminguide/tadminnetservcustom.html

> Write jvm information and path of derby.jar to derby.log
> --------------------------------------------------------
>                 Key: DERBY-4715
>                 URL: https://issues.apache.org/jira/browse/DERBY-4715
>             Project: Derby
>          Issue Type: Bug
>          Components: Miscellaneous
>    Affects Versions:
>            Reporter: Lily Wei
>            Assignee: Lily Wei
>            Priority: Minor
>             Fix For:
>         Attachments: DERBY-4715-1.diff, DERBY-4715-2.diff, DERBY-4715-3.diff, derby.log
> The bug is part of DERBY-1272. In production environment, derby.jar can be located different
than the derbyclient.jar It can be easier if we have jvm version information and path of derby.jar
are in the derby.log

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message