Hi Chris, Maurice

Thanks for the replies.

Was hoping someone would have already done something as I am under a pretty
tight deadline.

Dont think the denyhosts route will be a goer as hosts.deny will only be
writeable as root and dont want to have to run tomcat as root, change
permissions, or step outside the process.
Might look at that for my personal firewall though, good idea that.

Will look at putting something together further down the track to handle
this. Does anyone know if any of the other JWS's handle this issue??

Simon

On 7/29/06, Maurice Yarrow <[EMAIL PROTECTED]> wrote:

Simon, Chris

If you write your own mechanism, you might want to take a look
at the configuration script for "DenyHosts", which is a highly
configurable tool for port blocking (via mods to /etc/hosts.deny)
of sshd upon too-many failed attempts in a given time interval.
This is similar to what you are planning.  Helpful to look at what
they support (look at their "denyhosts.cfg", initially,
denyhosts.cfg-dist) in the way of resetting of failed-count upon
successful ssh login within permitted interval, purging
of denied hosts after configurable interval, etc.

And by the way, I have had their denyhosts stuff running for
nearly a week now, and it has handled sshd port 22 attacks
quite well, which have dwindled significantly as a result.
This has led me to conjecture that the attacks are from a
community of attackers who work kind of like the [EMAIL PROTECTED]
by applying all their cpu resources to a common set of
targets.  Now that their attack tools have had connection-
refused after 5 attempts, their tool has struck my address
off their list as being non-fruitful.  Just a conjecture, anyway.

Maurice Yarrow


Christopher Schultz wrote:

>Simon,
>
>
>
>>Has anyone done anything with tomcat authorisation to configure in a
>>maximum number of retries before an address/account is blocked.
>>
>>
>
>I'm pretty sure that Tomcat's authentication system does not support
>this feature. You could probably write your own authenticator to track
>that kind of thing.
>
>I am going to be adding the same type of feature to an authenticator I
>wrote to be used with securityfilter
>(http://securityfilter.sourceforge.net/). My plan is to use something
>like a synchronized time-sensitive cache of login failures (probably
>something from the commons-collections package such as LRUMap) to store
>login failures (keyed on username). I'll probably do the same thing with
>remote IP address as well (3 failures from the same IP will block future
>logins). The only trick is expiring entries ;)
>
>Let me know if you have any better ideas. I'd love to hear about them.
>
>-chris
>
>
>
>



---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


Reply via email to