I got some more info on this.
After some searching 1 of my co-workers told me he found a post on some other forum about the status text being capped off at 255 chars. When i was reading through that post i noticed they where talking about the ndo db. From what i have gathered it seems the db is the limiting factor in this whole 255 capping. My question of course is, can this be true and if so how can we change it so it holds a max of 4096 chars (yes i know thats alot).
With that in mind i also noticed something else. On the muticheck for the events the data that is being sent to the main central is partially between ( ). From the looks of it this whole bit gets stripped somewhere between the arriving on the central till showing on the website. I double checked he dell openmanage checks and there are a couple of lines that include ( ) as well so something else is wrong there.
check_multi support
Re: check_multi support
Can you post what the client is trying to send so I can try and recreate it on this end?
Re: check_multi support
Can you provide me with a way to let the send_nrdp command send the full info to a log file or some kind. Perhaps some other way how i can capture this to provide the data. Do note that the sender are core versions on a debian server.
Re: check_multi support
From what I can tell, Nagios is receiving the full data, it's just not being displayed in the interface because of it's length. If you want the raw data that is being received, you can run a grep on the /usr/local/nagios/var/status.dat file.
Re: check_multi support
We took a jump in the deep and we changed the outputs in the sql db from varchar 255 to varchar 8192. This seems to have fixed the data flow.
The only thing that is missing now is the delimiter for the new lines. For some reason it removes all that data from the status output. By the looks of it, it doesn't even transmit the data itself. By the looks of it, it somehow stops at the line delimiter. Its not a big issue but just strange.
The only thing that is missing now is the delimiter for the new lines. For some reason it removes all that data from the status output. By the looks of it, it doesn't even transmit the data itself. By the looks of it, it somehow stops at the line delimiter. Its not a big issue but just strange.
Re: check_multi support
I'm guessing it will be there in the Core interface, but not XI. Since XI's interface fetches it's data from an XML API, we process any characters that might break the XML generation. It's been requested as a feature to allow html in the plugin output of the XI interface, but we haven't implemented it yet.