My problem with this is that the saved search takes longer than 60 seconds to run, so I only get partial answers if I try to run it as a subsearch. (it times out)
What is key about my question are the words 'saved search results'. I have created a saved search and set up another search to use it as a subsearch. The problem I encounter is that when used as a subsearch, the results are ignored and the saved search is run fresh. The saved search takes longer than 60 seconds to run, so I only get partial answers when it runs 'live' in a subsearch.
What I need is the ability to retrieve results from a saved search and use those as a subsearch so that I don't time out.
This is an abbreviated example of what I'm doing now. This just runs the saved search fresh for the subsearch instead of pulling the saved results.
sourcetype=Data_Input_File [savedsearch timeless_base_search] | ... etc ... | table IPAddress MACAddress
Have you tried it with a pipe as the first character after the leading [ of the subsearch (e.g. [ | savedsearch timeless_base_search ]
). If that still doesn't work for you, consider loadjob. An example might look like | loadjob savedsearch="admin:search:timeless_base_search"
Another thing you could consider is to constrain the runtime of the subsearch, even if you want a different search scope for the outer search. This can be done using the "earliest" and "latest" keywords in your subsearch.
Have you tried it with a pipe as the first character after the leading [ of the subsearch (e.g. [ | savedsearch timeless_base_search ]
). If that still doesn't work for you, consider loadjob. An example might look like | loadjob savedsearch="admin:search:timeless_base_search"
Another thing you could consider is to constrain the runtime of the subsearch, even if you want a different search scope for the outer search. This can be done using the "earliest" and "latest" keywords in your subsearch.
Excellent! The 'loadjob' advice worked like a charm. The pipe didn't solve this problem, but now with the other advice it all works as I'd hoped.
Thanks much!
I will give those first two options a try. It was also suggested to me that I have the internal saved search output to a lookup table and import THAT as the subsearch, effectively. (egads, trying to describe some of this clearly is difficulty)
As for the time constraints, that's not something I can change. That's why it's 'timeless' in this instance. I don't need to run it often, but I do need a full answer.
Thanks much for the comment. I'll let you know how those suggestions work.