2

I have the dogecoin daemon running successfully on Ubuntu, started with this command:

dogecoind -daemon

It's listening on port 22556, as verified by netstat.

I can also pass commands to it directly via the Ubuntu terminal, like so:

dogecoind getinfo

However, passing RPC requests to it through HTTP does not work. The daemon doesn't refuse the connection (and I can verify that it does refuse the connection if I use a port other than 22556); instead, the connection hangs, with no data being sent to the client making the HTTP request. Directly connecting to

http://a:b@127.0.0.1:22556/

makes the web browser sit and wait for data. Similarly, making an RPC request via a PHP script, and then accessing that script in a browser, makes the browser wait for 30 second before PHP spits out an error stating that the maximum execution time of 30 seconds was exceeded.

I have my ~/.dogecoin/dogecoin.conf file as such:

server=1
rpcuser=a
rpcpassword=b

(I know the username and password I typed there are insecure; those are temporary until I can at least make a connection.)

I've tried adding the line 

rpcallowip=127.0.0.1

to the end of the config file, but that didn't change the behavior at all.

What am I doing wrong -- why isn't my Dogecoin daemon accepting RPC requests (or any from-an-actual-browser HTTP requests, for that matter)?

edit: Also, I can verify that the daemon is caught up with the blockchain.

Josh1billion
  • 131
  • 1
  • 5
  • 1
    rpcallowiip=0.0.0.0 That is an unsafe configuration if you are listening on a public ip address space. The address map 0.0.0.0 is shorthand for allow connections from all hosts. If you are running in a NAT environment you may be ok for the time being as long as you arent' sharing the network with any unknown hosts. (ie a dorm room network) Casey –  Dec 26 '13 at 18:56
  • @unklStewy: Ah, thanks for pointing that out. I'll be sure to change it before going to production. – Josh1billion Dec 31 '13 at 19:50

1 Answers1

1

Figured it out. It had to do with adding rpcallowip lines to the config file. Adding these three lines solved it:

rpcallowip=127.0.0.1
rpcallowip=192.168.1.*
rpcallowip=0.0.0.0

The last line is what eventually allowed it to work. I'm not sure why, but netstat said it was listening on 0.0.0.0:22556, so adding 0.0.0.0 to the rpcallowip list solved it.

Josh1billion
  • 131
  • 1
  • 5
  • 1
    That's because 0.0.0.0 in network speak essentially means "all IP's". So you've basically allowed all IP's to connect to your RPC server. – taco Jan 27 '14 at 02:19
  • Thanks for the information. I've since removed it to make sure that vulnerability doesn't exist.

    As far as the actual fix, I'm not sure what the issue was. Running a new installation on a different server, I haven't experienced the same issue that I originally posted about.

     – Josh1billion Apr 30 '14 at 22:12