BIND 10 #760: Conversion of cmdctl to use the new logging interface
BIND 10 Development
do-not-reply at isc.org
Thu Jun 30 08:40:17 UTC 2011
#760: Conversion of cmdctl to use the new logging interface
-------------------------------------+-------------------------------------
Reporter: | Owner: jinmei
stephen | Status: reviewing
Type: | Milestone:
enhancement | Sprint-20110712
Priority: major | Resolution:
Component: | Sensitive: 0
logging | Sub-Project: DNS
Keywords: | Estimated Difficulty: 2.0
Defect Severity: N/A | Total Hours: 0
Feature Depending on Ticket: |
Add Hours to Ticket: 0 |
Internal?: 0 |
-------------------------------------+-------------------------------------
Changes (by jelte):
* owner: jelte => jinmei
Comment:
Replying to [comment:6 jinmei]:
> - I noticed minor inconsistency about whether to capitalize the first
> word or not, see, e.g.:
> {{{
> % CMDCTL_NO_SUCH_USER username not found in user database: %1
> ...
> % CMDCTL_SEND_COMMAND Sending command %1 to module %2
> }}}
>
Oh, yeah that shouldn't be capitalized. Brain overrides that setting once
in a while. Fixed.
> - Like the xfrout case, there may be mapping inconsistency between the
> .mes and .py files. I've not found one myself, but haven't
> performed a comprehensive check either.
>
i did, and i just redid it, so i'm pretty comfortable with it (but we
still need a real automated test for it)
> - CMDCTL_COMMAND_SENT: looks like it should indicate an event where
> the message 'has been' sent, not 'is being' sent. At least the
> description is exactly the same as that for CMDCTL_SEND_COMMAND,
> which seems odd.
>
ack, should be 'has been sent'.
> - is it okay to simply remove this? (or is CMDCTL_COMMAND_SENT
> supposed to be the replacement?)
> {{{
> - if self._verbose:
> - self.log_info("Finish send command '%s' to module '%s'" %
(command_name, module_name))
> -
> }}}
>
yes, that is correct. I moved it up a little so message happens at the
moment the send action is complete, so that it does not get delayed (or
prevented) by a possible delay or failure in the recv that follows it.
> - CMDCTL_SSL_ERROR_USER_DENIED: the ID name and the short description
> may have to be revised, because the error may be local (i.e. the
> CmdctlException case). The current naming and description seem to
> indicate errors happening within SSL.
>
made that last part: "or the local key or certificate file could not be
read."
> - CMDCTL_CC_SESSION_ERROR: hardcoding 'msgq' might be fragile, as it's
> quite likely to be renamed pretty soon. But I'd leave it to you
> whether to do anything about that.
made it 'message bus daemon'
--
Ticket URL: <http://bind10.isc.org/ticket/760#comment:8>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list