Join GitHub today
GitHub is home to over 36 million developers working together to host and review code, manage projects, and build software together.Sign up
[dev.icinga.com #3329] committing acknowledgement cmd negates flags for send_notification and sticky_ack #1139
This issue has been migrated from Redmine: https://dev.icinga.com/issues/3329
Created by chunghung.chen on 2012-10-23 14:55:37 +00:00
After upgrade to 1.8.0, when acknowledge an event in classic ui, "notification bit" is wrong.
2012-10-27 13:34:47 +00:00 by mfriedrich eda48c8
2012-11-09 16:38:41 +00:00 by ricardo 27f47ba
2012-11-09 17:01:29 +00:00 by ricardo ee330c1
Updated by mzac on 2012-10-24 13:57:26 +00:00
I see the same thing:
With send notification checked:
With send notification unchecked:
Updated by mfriedrich on 2012-10-27 12:12:40 +00:00
true with 1.8.1 as well.
with notifications checked
Updated by mfriedrich on 2012-10-27 12:21:55 +00:00
from the cgi log
Updated by mfriedrich on 2012-10-27 12:40:46 +00:00
the same happens for sticky_ack as well, so this sources from #2927
my guess is, that the cmd.cgi sends the check boxes via POST, but actually the value field is omitted, leaving the default value in place.
need to test further on a possible fix.
Updated by mfriedrich on 2012-10-27 13:33:58 +00:00
though, there's an ugly workaround for that - add a hidden html field, which pushes the value of the original name, if that one was not checked.
actually this works for me.
i'm pretty sure Ricardo will change that to somewhat better, so leaving this around for now.
Updated by ricardo on 2012-11-09 17:02:49 +00:00
fixed in current dev/cgis
Found another way on checking if the default TRUE values for send_notification and sticky_ack are set or not. Don't really like the idea of adding a hidden html input and hoping that every browser behaves the same.
Updated by mfriedrich on 2012-11-10 14:41:43 +00:00
(deleting the ack before every test case)
ticked sticky ack, no notifications
ticked notifications, no sticky ack
so working as expected, thanks! (and for the ugly fix - well, there's always a workaround to html design fails...)