Noah's Randomness

UPDATE/Fix Information: I found out that this was all related to the previous solution of disabling the postgresql instance on the machine. I had (mistakenly) thought that by doing /etc/init.d/postgresql-8.3 stop that BackTrack’s posgresql instance would stop. Unfortunately I was wrong, and a “ps aux | grep -i sql” showed me this. After I did a “kill <PID>” I removed all NeXpose files/folders and reinstalled it. However it still wouldnt run correctly/gave me the same error. After some discussion in #rapid7 on irc.freenode.net, I was informed that NeXpose requires 1GB of RAM. So I edited my VMWare config to provide 1GB of RAM to my BackTrack guest.

However, then I had another issue — the NeXpose server kept crashing after starting. So, again I turned to the guys in #rapid7. With their help, I ran “cd /opt/rapid7/nexpose/nsc && rm conf/nsc.xml && ./nsc.sh” and voila! Success!

So, to reiterate, to fix “Could not start the nxpgsql daemon,” I:

• Determine the running PostgreSQL’s PID by doing “ps aux | grep -i sql”
• “kill <PID>” the offending PID from above
• Reinstall NeXpose
• If you are utilizing VMWare or some virtualization software, increase the VM’s RAM to 1GB (or greater)
• Delete nsc.xml from /opt/rapid7/nexpose/nsc/conf/
• Start the NeXpose daemon

Again, the cause of the “Could not start the nxpgsql daemon” is due to the fact that the nxpgsql server is trying to start, but there is already an existing PostgreSQL server running. The nxpgsql server tries to use the same default port (5432) as the existing PostgresSQL server. Knowing this, I set out to fix this problem. Here’s how I did it:

• Edit /opt/rapid7/nexpose/nsc/conf/nsc.xml
• Change line #4 to say: <Database dbms=”postgresql” db=”//127.0.0.1:5433/nexpose”>
  <Database dbms=”postgresql” db=”//127.0.0.1:5433/nexpose”>
• Save nsc.xml
• Edit /opt/rapid7/nexpose/nsc/nxpgsql/nxpdata/postgresql.conf
• Uncomment line 60 and change it to say: “port = 5433 # (change requires restart)”
• Save postgresql.conf

---

Original Post After seeing this post showing neXpose running on BackTrack 4, I wanted to try installing it on my BT4 VM. So, I followed the instructions in this video.

I went to run the neXpose client/server after install (by doing ./nsc.sh), but was met with the following error message/log:

NSC 12/21/09 6:43 PM: Initializing postgresql database manager for //127.0.0.1:5432/nexpose… postgresql 12/21/09 6:43 PM: Starting up postgresql DB system postgresql 12/21/09 6:43 PM: Nexpose PostgreSQL service status: 0 postgresql 12/21/09 6:43 PM: Nexpose PostgreSQL service status: 1 NSC 12/21/09 6:43 PM: Database initialization failed : java.sql.SQLException: Error starting PostgreSQL: com.rapid7.os.OSException: Could not start the nxpgsql daemon at com.rapid7.nexpose.datastore.A.K.E(Unknown Source) at com.rapid7.nexpose.datastore.n.H(Unknown Source) at com.rapid7.nexpose.datastore.n.E(Unknown Source) at com.rapid7.nexpose.datastore.n.E(Unknown Source) at com.rapid7.nexpose.nsc.NSC.U(Unknown Source) at com.rapid7.nexpose.nsc.NSC.ô(Unknown Source) at com.rapid7.nexpose.nsc.NSC.õ(Unknown Source) at com.rapid7.nexpose.nsc.NSC.run(Unknown Source) at com.rapid7.nexpose.nsc.NSC.main(Unknown Source) Caused by: com.rapid7.os.OSException: Could not start the nxpgsql daemon at com.rapid7.nexpose.datastore.A.K.A(Unknown Source) … 9 more SQLState=null, errorCode=0

So with that in mind I started Googling around to see what I could find. Long story short, I found the following:

• neXpose Known Issues, which didn’t help at all/have any useful information
• Metasploit Mailing List Post regarding a similar issue which told me to disable any existing postgresql instance and reinstall neXpose
• neXpose Common Errors, which went so far as to suggest complete removal of postgresql from your system… which didn’t seem too logical to me seeing as Metasploit uses it…

So with all that said, I’m still stuck. I’m hoping that @rapid7 responds to my Tweet (@rapid7 HELP! Can’t get #neXpose to work on BT4. So far, no go: “Could not start the nxpgsql daemon.” Full error paste: http://bit.ly/6pJ1Ve), but we’ll see.

If anyone has encountered this before and/or has any ideas on how to fix it, please leave me a comment below. Thanks!