Re: Snort init script `stats` function
Peter Gallagher via Snort-devel <[email protected]>
| Newsgroups | gmane.comp.security.ids.snort.devel |
|---|---|
| Message-ID | <CAHKuSSuypSBJj55dN=iSqW49Zbh2cWsZgXUfgy1tQHV6zrSJAA@mail.gmail.com> |
Thinking about this further, there is a possibility that additional logs could be interspersed between the snort output in the syslog. So it would be prudent to increase the number of lines returned to be >147 in order to mitigate against this (e.g. 200), this should allows us to capture all 148 lines of snort output. Regards, Peter On 23 August 2017 at 12:31, Peter Gallagher <[email protected]> wrote: > The current `stats` function in the init.d script has been very unreliable > for me, due to variations in the `startdate` variable recorded in the > script and the timestamp when the log lines are actually written by snort. > > Accordingly I have modified my init script to use the following: > > tac /var/log/messages | grep -m1 '*** Caught Dump Stats-Signal' -B147 > | tac | grep snort.*: | cut -d: -f4- > > This example of the basic `stats` function does the following: > > 1. Uses `tac` to reverse the order of `$SYSLOG` > 2. Greps for the first occurrence of the snort stats start signature and > includes an additional 147 lines of output > 3. Reverses the order (again using `tac`) to return it to normal order > 4. Filters out any non-snort lines > > I think this is a more reliable approach. The only risk is if the stats > aren't dumped correctly it could return an earlier collection of > statistics, however I feel this is highly unlikely. > > I looked for a public repository to provide a diff or pull request but > could not find any. > > Please let me know if you think this is suitable for inclusion for future > releases. > > Regards, > > Peter > _______________________________________________ Snort-devel mailing list [email protected] https://lists.snort.org/mailman/listinfo/snort-devel Please visit http://blog.snort.org for the latest news about Snort!