[dev.icinga.com #6237] no config option set for "Sticky Acknowledgement" in cgi.cfg. #1479
Comments
Updated by mfriedrich on 2014-05-27 20:31:01 +00:00
|
Updated by mfriedrich on 2014-06-20 08:57:18 +00:00
|
Updated by mfriedrich on 2014-10-26 19:00:00 +00:00
Since you already had a look into the other options, please evaluate this one too. |
Updated by crfriend on 2014-10-26 21:05:57 +00:00
What we have here is a failure to read the doco. There is already a CGI configuration option called "sticky_ack" that does what the submitter wants, but it's (1) not universal across the various programs (e.g. config.cgi), (2) doesn't follow the newer standard for Boolean configuration values which begin with "set_" followed by the checkbox name, and (3) doesn't have the code to preserve the setting if the user's browser reloads the page. I seek guidance in this regard. I coded the changes to the CGI pieces before running into the elder option in some of the documentation (which'll need changing, as will config.cgi). The changes, using the newer syntax work, but I worry about possible downstream effects if folks use the older configuration variable (and I've tried to make my change sensitive to both). Which path to follow, then: (1) deprecate the old variable, (2) use the new one, make it universal across the CGI suite, and document it, or (3) tell the submitter to read the documentation? |
Updated by crfriend on 2014-10-28 00:53:41 +00:00
This needs testing on architectures other than Solaris. It behaves OK here in my lab, and I have every reason to believe that it'll be portable to Linux, but we need to be sure of the fact before merging it. The update to the English-language documentation also needs a translation into German. I also strongly advocate for deprecating the already-extant "sticky_ack" CGI configuration option which didn't have full visibility in the Icinga suite. |
Updated by crfriend on 2014-10-28 22:58:19 +00:00
Do NOT commit "Revision 1505fd7". I've got a QA problem with it, and am addressing it now. Specifically, the UI can behave strangely depending on whether the box is ticked or not. I'll flag here when I've got it sorted. |
Updated by mfriedrich on 2014-10-28 23:38:51 +00:00 Besides that, documentation updates must be done in the icinga-doc.git and pushed separately. That may happen after the commit was merged into next. Once you're done, please assign to Ricardo for review. |
Updated by crfriend on 2014-10-28 23:50:16 +00:00 dnsmichi wrote:
Will do, on both counts. The code change is complete; I still need to do the docs. Who do I assign to in order to get the German translation done? |
Updated by crfriend on 2014-10-29 00:23:27 +00:00
OK, it looks like this one is finally good to go for further testing. I pushed another commit this evening which corrects the aberrant behavior I was seeing, but it still needs testing on Linux and others (Are there others?) before being released into the wild. I managed to forget to include a "refs" mention in my last commit, so RedMine isn't going to see it. All 4 permutations have been checked on Solaris with predicted (and desired) results. |
Updated by ricardo on 2014-11-02 18:32:24 +00:00
I'll have a look right now |
Updated by ricardo on 2014-11-02 20:40:02 +00:00
Applied in changeset icinga-core:62e388dad9a609885d4c5d0fce596c3a7db388c9. |
Updated by ricardo on 2014-11-02 21:12:40 +00:00
|
Updated by mfriedrich on 2014-12-08 09:15:46 +00:00
|
This issue has been migrated from Redmine: https://dev.icinga.com/issues/6237
Created by kiranos on 2014-05-19 15:38:24 +00:00
Assignee: ricardo
Status: Resolved (closed on 2014-11-02 20:40:02 +00:00)
Target Version: 1.12
Last Update: 2014-12-08 09:15:46 +00:00 (in Redmine)
Hi,
I want to have "Sticky Acknowledgement" set to unchecked as default but the cgi.cfg doesnt have a configline for this, there is:
send_ack_notifications=1
persistent_ack_comments=0
set_expire_ack_by_default=0
But I dont know what to set as 0 to uncheck "Sticky Acknowledgement" as default. Is there such an option already available but not added to the cgi.cfg or can it be implemented?
Thanks.
Changesets
2014-10-28 00:27:28 +00:00 by crfriend 1505fd7
2014-11-02 19:38:52 +00:00 by ricardo 0911014
2014-11-02 20:12:03 +00:00 by ricardo 62e388d
Relations:
The text was updated successfully, but these errors were encountered: