On Tuesday 18 November 2008 07:00:56 Michael Wiegand wrote: > * Tim Brown [17. Nov 2008]: > > Cheers Michael and Thomas, but actually that's not the problem :(. > > > > Obviously your check is an improvement on that ;). I may have the chance > > to run my scripts on a real world box this week, so I'll see if I can > > nail down where my problem lies. > > That would have probably been too easy. :) Do you have any other hints > as to what the problem might be? What goes wrong? I have access to a > Solaris Box as well, so if there is anything you want me to test, just > let me know.
Okay, I got the chance to do a bit of debugging this evening. At the start of my copy of gather-package-list.nasl: uname = ssh_cmd(socket:sock, cmd:"uname -a"); is called. This successfully results in uname being set to "SunOS obfuscated 5.o Generic_oooooo-oo sun4u sparc SUNW,UltraSPARC-IIi-cEngine". Then at the end of the script, if uname is detected as starting with "SunOS ", then some more commands are run: security_note(port:port, data:uname); osversion = ssh_cmd(socket:sock, cmd:"uname -r"); security_note(port:port, data:osversion); set_kb_item(name: "ssh/login/solosversion", value:osversion); hardwaretype = ssh_cmd(socket:sock, cmd:"uname -p"); security_note(port:port, data:hardwaretype); set_kb_item(name: "ssh/login/solhardwaretype", value:hardwaretype); buf = ssh_cmd(socket:sock, cmd:"pkginfo"); security_note(port:port, data:buf); set_kb_item(name: "ssh/login/solpackages", value:buf); buf = ssh_cmd(socket:sock, cmd:"showrev -p"); security_note(port:port, data:buf); set_kb_item(name: "ssh/login/solpatches", value:buf); However, it appears that osversion etc never get populated. Each debug security_note results in a hole being noted which contains the following data: "This script will, if given a userid/password or key to the remote system, login to that system, determine the OS it is running, and for supported systems, extract the list of installed packages/rpms." This is the default value (description) which is used when (I believe) security_note is passed a null data parameter. If I reorder the script so that the Solaris checks are carried out directly after the initial uname then it works. In essence, the later ssh_cmd do not appear to run correctly. I did try your patch to libopenvas MIchael, but that didn't seem to resolve it. Cheers, Tim -- Tim Brown <mailto:[EMAIL PROTECTED]> <http://www.nth-dimension.org.uk/> _______________________________________________ Openvas-devel mailing list Openvas-devel@wald.intevation.org http://lists.wald.intevation.org/mailman/listinfo/openvas-devel
