how to use the TA-uas_parser

[cdm33]

i've installed it but how do you use it ?

[dshpritz]

Basically, once it is downloaded, you need to run the script to update the data file. After that you can use the lookup. It does expect that the field you are looking up is "http_user_agent", but if your field extractions are different than that, you can use the rename command:

Let's say your field that you want to lookup is named "user-agent":

index="weblogs" | rename user-agent AS http_user_agent | lookup uas_lookup http_user_agent

Your events should now have the additional fields.

Thanks,

Dave

[dshpritz]

Hey splunkears,

Yeah, I'm afraid it can be sluggish, especially if the search returns a lot of values. You could also try out the TA-browscap plugin to see if that is any faster. Another possibility is to create a summary search that runs the lookup in the background to populate the summary index. This way you would have the user-agent string parsing done ahead of time.

Thanks,

Dave

[splunkears]

Though this tip was useful, I found the app TA-uas_parser DEAD slow. while using with
a query like the following:

index="weblogs" |
rename user-agent AS http_user_agent |
lookup uas_lookup http_user_agent | stats dc(myUserApp) by os_family

any tip on making this run faster?

[cdm33]

there is info on the page - but no instructions on how to actually use it !!

[Ayn]

There's info available on the app page: http://splunk-base.splunk.com/apps/48017/ta-uas_parser (also look at the review which has an example on how to use the lookup...)

[Ayn]

It's a lookup, so you call it like you would with other lookups. As the example in the review shows, it expects the input field to be called "http_user_agent".

[cdm33]

there's info on the page but no instructions on how to use it !!!