Recently, we found that several scripts that we typically explicitly
avoided taking into our feed (LSCs that did their work without looking
at package information) slipped through were included with our scripts.

Then, we did a full LSC audit, with credentials, of a system that had
a VERY large file system on it.  The box became essentially unusable,
because the scripts were running "find" operations to locate binaries,
the scripts would time out, but the "find" operations would NOT timeout
and keep running.  They piled up, one on top of another grinding the
box down.  This was with only a handful scripts that were in this
category. I suspect if we had the full suite, the box would probably
have either crashed or become completely unresponsive.

Not being an SSH expert, not quite sure exactly what the solution would
look like, but there ought to be a way to terminate remote commands
a script has executed if the script executing it shuts down.

Thomas
_______________________________________________
Openvas-plugins mailing list
[email protected]
http://lists.wald.intevation.org/cgi-bin/mailman/listinfo/openvas-plugins

Reply via email to