BIND 10 #2445: suppress initial log
BIND 10 Development
do-not-reply at isc.org
Mon Dec 10 23:44:13 UTC 2012
#2445: suppress initial log
-------------------------------------+-------------------------------------
Reporter: jinmei | Owner:
Type: defect | jinmei
Priority: medium | Status:
Component: Boss of BIND | reviewing
Keywords: | Milestone:
Sensitive: 0 | Sprint-20121218
Sub-Project: Core | Resolution:
Estimated Difficulty: 4 | CVSS Scoring:
Total Hours: 0 | Defect Severity: N/A
| Feature Depending on Ticket:
| Add Hours to Ticket: 0
| Internal?: 0
-------------------------------------+-------------------------------------
Changes (by jelte):
* owner: jelte => jinmei
Comment:
Hmm, that's actually slightly tricky, and we have a little bit of a trade-
off here.
in createBufferAppender(), I set the logger level to log anything. If this
isn't done, any level past INFO won't be logged, even if the later config
says DEBUG (since it has been filtered out already and did not even get to
the appender).
But this causes everything to be dumped in case it gets to flushStdout().
I just considered another approach; storing the initial log level, and
checking it in flushStdout (normal flush should use the log level of the
spec it just applied). But then for at least bind10-boss, the DEBUG
messages wouldn't show up in flushStdout even if -v was given, because
boss sets the log level of the logger manually, and right now I can't
think of any way to do this 'right';
- these two approaches either show too much or too little
- removing setLogLevel(TRACE) causes debug messages to be dropped in the
case where -v is *not* given, but the config says DEBUG (that is, the
messages until process() on the config is first called)
- storing the *current* loglevel of the logger during append() would also
not work, since we'd still need to set it to TRACE initially then, which
would mean the same result as what it originally did (dump too much).
For now I'm removing the setLogLevel(TRACE), so it will dump DEBUG on -v,
and not dump them in normal mode, but then, in the -v is not given, and
working configuration that sets it to DEBUG, some initial DEBUG messages
get lost....
Maybe some idea will come to me after some rest, but for now I think we'll
have to live with that.
--
Ticket URL: <http://bind10.isc.org/ticket/2445#comment:28>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list