module.compatible_payloads error

[GoogleCodeExporter]

What steps will reproduce the problem?
1. perform Hail Mary attack
2.
3.

What is the expected output? What do you see instead?
I expect for it to run it's course and show exploits, if any. Instead, I get 
over 30 error messages all stating "Something went 
wrong:module.compatible_payloads

Error:java.lang.RuntimeException:can't add a new key into hash during iteration"

What version of Metasploit are you using (type: svn info)? On which
operating system?
Version 4.5  on Backtrack 5 R3

Please provide any additional information below.

This error was not occuring with the version of Metasploit/Armitage supplied 
with BT5R3, but then after doing msfupdate, this error started occuring. 

Can that payload file be removed, and if so, what is the command or path to do 
so? 

Original issue reported on code.google.com by mpr0...@gmail.com on 6 Dec 2012 at 1:11

[GoogleCodeExporter]

I uninstalled the 4.5 version from BT5, and then reinstalled (the apt-get 
reinstall is for 4.4), and it is working well now, however I was able to 
recreate this issue by doing an msfupdate again. There were three conflicts 
when performing the update, which might have caused the issue.

Original comment by mpr0...@gmail.com on 6 Dec 2012 at 3:55

[GoogleCodeExporter]

There is not enough information here for me to help. Here's what will make it 
easier for me to help you:

1. What kind of targets did you run the hail mary against? I will happily spin 
up a few VMs to match your environment and attempt to reproduce the problem. I 
don't know what your environment is--so I'm stuck. 

2. You see this lovely hash iteration error thing. That's fine. I want to help 
you, but I don't know where to start. To help me, help you, do this:

setg LogLevel 5

Reproduce the problem. Find ~/.msf4/logs/framework.log and paste in the 
relevant pieces into this bug report.

3. Which revision of the Metasploit Framework are you using? While msfupdate 
uses svn, you can do this:

cd /opt/metasploit/msf3
source ../scripts/setenv.sh
svn info .

I can share some information about the symptom you're seeing though. The 
Metasploit Framework introduced a thread that builds a cache of modules in the 
background. Sometimes, Ruby detects a thread modifying data while another 
thread loops through it. Ruby then flips out with this exception. I'm not a 
Ruby dev, but this situation is similar to the ConcurrentModificationException 
in Java. These issues suck to track down and make programmers pull their hair 
out.

From time-to-time, this bug reintroduces itself *because* of a commit to the 
Metasploit Framework. This is bad and when it happens--I do everything I can to 
help narrow it down to fix it.

Other times, this bug surfaces because you have a payload with a syntax error 
in it. This can happen because of an incomplete msfupdate. Sometimes a file 
with an error is committed, sometimes an old file is left in your install that 
was meant to be removed by msfupdate. Stale files are a cause of pain and your 
framework.log will help narrow the problem down if this is the case.

I updated to the latest version of the framework and tried a Hail Mary against 
an XP box and two Linux boxes. I was not able to reproduce this problem.

I can't test every revision of the framework against Armitage. If you run 
msfupdate and update to the latest development version of the framework, you're 
kind of on your own. Expect problems. I publish the last version of the 
framework I tested against. For some information on this, go to:

http://www.fastandeasyhacking.com/manual#msfupdate

Finally, I love BackTrack, but for my sanity and yours--install with Rapid7's 
installer. I don't know what the package you used looks like. If the package 
includes files that would be removed by msfupdate and something changes to 
cause those left over files to reference something that's no longer 
there--you'll experience this pain. 

All of this is just a guess though. If your troubleshooting yields the answer, 
let me know so I can keep it in mind when helping others in the future.

Since this is not an Armitage issue, I'm closing the ticket as invalid.

Original comment by rsmu...@gmail.com on 6 Dec 2012 at 8:56

• Changed state: Invalid