|
On Fri, Jan 12, 2007 at 09:16:24PM -0800, Dung Nguyen wrote:
> Jaro,
> I've thought about the verbage and do agree with you. However, I wanted to
> keep the change to a minimal because some organization out there may already
> be monitoring the logs for the existing strings. That's why I choosed to
> change the string for the lesser critical of the two messages. I opted to
> retain the exact string for the failure/restore due to status checks since
> these are unplanned events. I was hoping that this patch could be applied
> with minimal changes to organizational practices/procedures.
Hi,
sorry for not getting back to you earlier.
I'm pretty comfortable with the idea of denoting the difference
between administratively triggered additons and removals, and
status-check triggered ones. I'm not particularly fussed by what
exactly the text looks like. Though here is my 2c worth.
On configuration file reload event:
delete : "Purged real server (configuration reload): ..."
restore: "Re-added real server (configuration reload): ..."
On status check:
delete : "Deleted real server: ..."
restore: "Added real server: ..."
And now some mechanics:
If you want a patch to go into ldirectord, then it would help me a lot
if you could provide a sign-off-line, as described at the link
below. Basically its a way of you saying, I wrote this code, and
its ok for it to be included in GPL code. Its primary purpose
is so that if someone else claims they own the code there is
a way of tracking where it came from.
See section 5 of http://linux.yyz.us/patch-format.html
And, patches should really be against ldirectord.in. You can get
the lastest from:
http://hg.linux-ha.org/dev/raw-file/466b5d495b37/ldirectord/ldirectord.in
Thanks
>
> -Dung
>
>
>
> On 1/12/07, jarol1@xxxxxxxxx <J.Libak@xxxxxxxxxx> wrote:
> >
> >Dung Nguyen wrote:
> >> Per request, I've made the following modification to the patch.
> >>
> >> On configuration file reload event:
> >> delete : "Purged real server: ..."
> >> restore: "Re-added real server: ..."
> >>
> >> On status check:
> >> delete : "Deleted real server: ..."
> >> restore: "Added real server: ..."
> >>
> >> So for monitoring service, one would want to monitor these strings:
> >> "Deleted real server" and "Added real server".
> >>
> >I think Re-added would make more sense for status check and Added for
> >reload.
> >
> >Or another way of dealing with this problem would be mentioning "status
> >check" and "reload" in the log. Then we could have only "added" and
> >"deleted". The same fix should be made for the email subject that is
> >sent to the admin by ldirectord.
> >It could look like this:
> >
> >On status check:
> >delete : "Deleted real server due to status check failure"
> >restore: "Added real server due to status check success"
> >
> >On configuration file reload event:
> >delete : "Deleted real server due to reload event"
> >restore: "Added real server due to reload event"
> >
> >Either way, Horms is the maintainer of ldirectord so we have to find a
> >solution that he will accept. In my opinion the 2nd solution is better
> >as it clearly states why a server was added or deleted.
> >
> >Jaro
> >
> >_______________________________________________
> >LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> >Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> >or go to http://www.in-addr.de/mailman/listinfo/lvs-users
> >
> _______________________________________________
> LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://www.in-addr.de/mailman/listinfo/lvs-users
--
Horms
H: http://www.vergenet.net/~horms/
W: http://www.valinux.co.jp/en/
|
