Created by mfriedrich on 2011-02-04 14:47:38 +00:00
Assignee: mfriedrich
Status: Resolved (closed on 2011-02-07 13:55:52 +00:00)
Target Version: 1.3
Last Update: 2011-02-07 13:55:52 +00:00 (in Redmine)
hint from ton on twitter, that the command_line processed isn't populated.
now from my analysis - the command_line is being put on the initiate call which is now ignored in idomod, but not kept until the check is processed (which is being put as NULL).
by introducing a new entry in the object struct, this will be saved for the broker update then.
important note: since this is a change in objects.h it should be put at the end of the struct. neb modules not being aware of the change will fail on typecast if not.
since i introduced the change for idoutils, it's now a must for 1.3 either way.
maybe we'll put up the ipv6 issue too then in #674
2011-02-07 13:30:29 +00:00 by mfriedrich 010fc8034e8ad6de5da31217ace48ea97927781e
core: save processed command for neb callback #1185
this introduces a new member in host|service struct
in objects.h but as a matter of fact, it'll be
added at the end, non-violating typecasts and faulty
address mapping. Kudos to Opsview for the patches,
we'll take the more processing at the end of the struct
(which is interessting regarding performance vs
compatibility).
tested with IDOUtils and MKLivestatus (1.1.9i8)
fixes #1185
This issue has been migrated from Redmine: https://dev.icinga.com/issues/1185
Created by mfriedrich on 2011-02-04 14:47:38 +00:00
Assignee: mfriedrich Status: Resolved (closed on 2011-02-07 13:55:52 +00:00) Target Version: 1.3 Last Update: 2011-02-07 13:55:52 +00:00 (in Redmine)
hint from ton on twitter, that the command_line processed isn't populated.
now from my analysis - the command_line is being put on the initiate call which is now ignored in idomod, but not kept until the check is processed (which is being put as NULL). by introducing a new entry in the object struct, this will be saved for the broker update then.
important note: since this is a change in objects.h it should be put at the end of the struct. neb modules not being aware of the change will fail on typecast if not.
since i introduced the change for idoutils, it's now a must for 1.3 either way. maybe we'll put up the ipv6 issue too then in #674
Attachments
Changesets
2011-02-07 13:30:29 +00:00 by mfriedrich 010fc8034e8ad6de5da31217ace48ea97927781e