Fix RequestError class undefined method issue

[space-r7]

If Metasploit receives an error response from Mettle, Metasploit attempts to return which command failed. The search for the command id currently throws a NoMethodError because start_with? is called on a Symbol instead of a String. This change just makes sure the variable used is a string before start_with? is called on it.

Verification

• Start msfconsole
• Generate a Meterpreter payload
• Get a Meterpreter session
• Use a command that you expect to fail: rm blah.notrealext
• Verify that the error tells you which extension / command was used

Scenarios

Before fix

msf6 > use multi/handler
[*] Using configured payload generic/shell_reverse_tcp
msf6 exploit(multi/handler) > set payload linux/x64/meterpreter/reverse_tcp
payload => linux/x64/meterpreter/reverse_tcp
msf6 exploit(multi/handler) > set lhost 192.168.37.1
lhost => 192.168.37.1
msf6 exploit(multi/handler) > run

[*] Started reverse TCP handler on 192.168.37.1:4444 
[*] Sending stage (3008420 bytes) to 192.168.37.170
[*] Meterpreter session 1 opened (192.168.37.1:4444 -> 192.168.37.170:35530) at 2021-03-12 17:56:58 -0600

meterpreter > rm blah.c
[-] Error running command rm: NoMethodError undefined method `start_with?' for :TLV_TYPE_LOCAL_DATETIME:Symbol

After fix

msf6 exploit(multi/handler) > run

[*] Started reverse TCP handler on 192.168.37.1:4444 
[*] Sending stage (3008420 bytes) to 192.168.37.170
[*] Meterpreter session 2 opened (192.168.37.1:4444 -> 192.168.37.170:35532) at 2021-03-12 17:59:07 -0600

meterpreter > rm blah.c
[-] stdapi_fs_delete_file: Operation failed: 1

[timwr]

For some reason I can't reproduce the original error (without the fix applied):

msf6 exploit(multi/handler) > ruby -v
[*] exec: ruby -v

ruby 2.7.2p137 (2020-10-01 revision 5445e04352) [x86_64-linux]
msf6 exploit(multi/handler) > version
Framework: 6.0.35-dev-a428c5721d
Console  : 6.0.35-dev-a428c5721d
msf6 exploit(multi/handler) >
[*] Sending stage (3008420 bytes) to 192.168.13.37
[*] Meterpreter session 2 opened (192.168.13.37:4444 -> 192.168.13.37:50732) at 2021-03-13 11:03:01 +0000

msf6 exploit(multi/handler) > sessions 2
[*] Starting interaction with 2...

meterpreter > rm blah.notrealfile
[-] stdapi_fs_delete_file: Operation failed: 1

[bcoles]

For some reason I can't reproduce the original error (without the fix applied):

I presumed, perhaps in error, that spacey's examples were around the wrong way. ie, before fix should read after fix.

I am wrong.

[space-r7]

For some reason I can't reproduce the original error (without the fix applied):

msf6 exploit(multi/handler) > ruby -v
[*] exec: ruby -v

ruby 2.7.2p137 (2020-10-01 revision 5445e04352) [x86_64-linux]
msf6 exploit(multi/handler) > version
Framework: 6.0.35-dev-a428c5721d
Console  : 6.0.35-dev-a428c5721d
msf6 exploit(multi/handler) >
[*] Sending stage (3008420 bytes) to 192.168.13.37
[*] Meterpreter session 2 opened (192.168.13.37:4444 -> 192.168.13.37:50732) at 2021-03-13 11:03:01 +0000

msf6 exploit(multi/handler) > sessions 2
[*] Starting interaction with 2...

meterpreter > rm blah.notrealfile
[-] stdapi_fs_delete_file: Operation failed: 1

Thanks for trying that out! Yea, I was using 2.6 at the time, and that was the issue. It looks like the start_with?() method was added to the Symbol class in v2.7:

https://ruby-doc.org/core-2.6/Symbol.html
https://ruby-doc.org/core-2.7.0/Symbol.html

[adfoster-r7]

The fix seems good to me, we currently aim to have framework working with Ruby 2.5, 2.6, 2.7 - and soon in the future 3.0

metasploit-framework/.github/workflows/verify.yml

Lines 49 to 52 in 09704e7

	ruby:
	- 2.5
	- 2.6
	- 2.7

Although Ruby 2.5 will be EOL soon enough:

Ruby 2.5
status: security maintenance
release date: 2017-12-25
EOL date: 2021-03-31

---

Just verified the behavior of start_with in IRB with Ruby 2.7:

2.7.2 :001 > 'a'.start_with? 'a'
 => true
2.7.2 :002 > :a.start_with? 'a'
 => true 
2.7.2 :003 > RUBY_DESCRIPTION
 => "ruby 2.7.2p137 (2020-10-01 revision 5445e04352) [x86_64-darwin19]" 

With Ruby 2.6:

2.6.5 :001 > 'a'.start_with? 'a'
 => true 
2.6.5 :002 > :a.start_with? 'a'
Traceback (most recent call last):
	21: from /Users/user/.rvm/gems/ruby-2.6.5/bin/ruby_executable_hooks:24:in `<main>'
	20: from /Users/user/.rvm/gems/ruby-2.6.5/bin/ruby_executable_hooks:24:in `eval'
	19: from /Users/user/.rvm/gems/ruby-2.6.5/bin/irb:23:in `<main>'
	18: from /Users/user/.rvm/gems/ruby-2.6.5/bin/irb:23:in `load'
	17: from /Users/user/.rvm/gems/ruby-2.6.5/gems/irb-1.3.2/exe/irb:11:in `<top (required)>'
(irb):2:in `<main>': undefined method `start_with?' for :a:Symbol (NoMethodError)
2.6.5 :003 > RUBY_DESCRIPTION
 => "ruby 2.6.5p114 (2019-10-01 revision 67812) [x86_64-darwin18]" 

[adfoster-r7]

Release Notes

Fixed a previous feature which added the readability of Meterpreter error messages via replacing the command ID with the command name to now work with older versions of Ruby.