[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [hobbit] ACK with Delay
On Mon, Feb 14, 2005 at 09:39:52AM -0500, brent.mccrackin (at) bell.ca wrote:
>
> The ACKCODE in BB is made up of a 5-digit alert code and a 2-digit pager
> code to identify who the alert was sent to. The 5-digit alert code is
> the same for all recipients of that alert. The pager code would change
> depending on the sequence that the matching rule was encountered in the
> bb-warnrules configuration. Replacing the provided pager code with a 99
> in the ACK input would acknowledge the alert for all recipients (except
> the escalation ones marked with ^).
>
> Is this feature retained in Hobbit?
No. the acknowledgement cookie in Hobbit only identifies a
host+service. The same ack-cookie is sent to everyone who is alerted.
The main reason Hobbit does not generate unique cookies for each
recipient is that to do so, the core daemon (hobbitd) and the alert
module (hobbitd_alert) would have to be a lot more aware of each other
than they are now.
> If not, how do we identify who acknowledged an alert?
Make sure you include that information in the ack message you provide
when ack'ing the alert.
I admit that this isn't something I've given much thought, because it
isn't a real problem for me. But I've just added a bit of code to the
bb-ack CGI, so if you have the acknowledgement web form in the secure
CGI directory (which is the default setup), then the username used to
login to the secure area will be included in the ack-message, and
shown on the status page.
If that is not good enough, then we'll have to discuss how to do it
for the next release.
Regards,
Henrik