-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Mr Starks put out a call for a wishlist, so I will comply and put forth my top 
feature requests...

1) Please, for the love of <insert deity here>, please add a way to push a 
config to a agent.  I'm ok with having to use some form of scripting to restart 
agents when they receive the new config, but there is currently no way that I 
know of to push a config out to an agent.  I made changes to the config, the 
agent needs them.  I don't typically have the time to wait for the agents to 
come get them.

2) Somewhat related to the previous request, there is no way, that I know of, 
to identify whether the config the agent has is the one that is currently being 
used.  For instance, I update the agent.conf file.  I completely forget to 
restart the agent when it gets the config, likely because of #1.  If I use 
agent_control to look at the agent, it shows the md5sum of the new agent.conf, 
even though it's not the one actively being used..  Perhaps a flag that 
agent_control can show as to whether the agent has been restarted since it 
received the new config?

3) Prior to moving to OSSEC, I used Osiris for integrity checking.  Osiris 
would send a single email detailing all of the changes for a single client.  
This was extremely useful as I could see, at a glance, all of the files that 
changed since the last check.  As I understand it, and I could be wrong, 
syscheck runs every x minutes and is a single process.  ie, when syscheck runs, 
it checks every directory and file specified before going back to sleep.  It 
*should* be possible, I think, to generate a single email detailing all of the 
changes.  I think you can write specific rules that trigger on syscheck events, 
so perhaps OSSEC should continue to act as it does now, but add a flag to allow 
the changes to queue and be sent as a single email?  The agent should be able 
to signal to the server that the syscheck process is complete.

4) It would be nice to be able to monitor ports and kernel modules.  I think 
this can be done by calling an external program, but native support for this 
would be nice.  Plus, if this was done internally, I believe it would thwart 
any root kits that try to hide modules and ports from the common tools used to 
look at them.

...

That's all I have for now..  Let's get chatting and see what else comes of it..

- ---------------------------
Jason 'XenoPhage' Frisvold
[email protected]
- ---------------------------
"Any sufficiently advanced magic is indistinguishable from technology."
- - Niven's Inverse of Clarke's Third Law



-----BEGIN PGP SIGNATURE-----
Version: GnuPG/MacGPG2 v2.0.14 (Darwin)

iEYEARECAAYFAkyybM8ACgkQ8CjzPZyTUTRabACffkSw3EvuYmDuZEisI6qecLbh
nQwAoI6dCCykvwbnsNp/Vo/PVqmLF8UN
=8Mct
-----END PGP SIGNATURE-----

Reply via email to