[Nfd-dev] Fwd: How ccnx sets sock file

[Wentao Shang]

I remember I had the same question during the summer. It is indeed a hard
problem. Using customized config file seems the best (only?) solution. But
you have to be careful because the ndn-cxx library reads default
configuration from /usr/local/etc/ndn/client.conf, which is also shared
across the system. In your applications you have to avoid using the default
unix socket and explicitly specify the correct location.

Wentao

On Tue Dec 09 2014 at 1:01:55 PM Lan Wang (lanwang) <lanwang at memphis.edu>
wrote:

>  No, it's not about ccnx.  We're trying to run multiple instances of nfd.
>
> Lan
>
>
>
>  On Dec 9, 2014, at 2:59 PM, Junxiao Shi <shijunxiao at email.arizona.edu>
>  wrote:
>
>  This is a question about CCNx. I'm forwarding this question to ccnx
> mailing list.
>
> ---------- Forwarded message ----------
> From: Lan Wang (lanwang) <lanwang at memphis.edu>
> Date: Tue, Dec 9, 2014 at 1:56 PM
> Subject: [Nfd-dev] Fwd: How ccnx sets sock file
> To: "nfd-dev at lists.cs.ucla.edu" <nfd-dev at lists.cs.ucla.edu>
>
>
> Any suggestions on how to solve this problem?
>
> Lan
>
>
> Begin forwarded message:
>
>  *From: *"Ashlesh Gawande (agawande)" <agawande at memphis.edu>
>  *Subject: **Re: How ccnx sets sock file*
>  *Date: *December 9, 2014 1:42:12 PM CST
>  *To: *"Lan Wang (lanwang)" <lanwang at memphis.edu>
>  *Cc: *"Vince Lehman (vslehman)" <vslehman at memphis.edu>
>
>  So I am using the mininet (miniccnx) environment where virtual machines
> share file systems. Miniccnx can run multiple instances of CCNx on these
> virtual machines. The sock file it uses can be specified as following:
>
>   export CCN_LOCAL_SOCKNAME=/tmp/file.sock
>
>  This is what miniccnx does on each node before starting CCNx when it
> instantiates the nodes.
>
>  For NFD:
>
>  I set up two nodes connected to each other.
>  Then if I want to start different instances of NFD and applications on
> them. To do this I need to have different sock files, but the sock file
> is hard coded in ndn-cxx.
>  As a temporary work around heres what I did:
>
>  1) Compiled ndn-cxx with a different sock file
> (/var/run/host1-nfd.conf) and then compiled NFD and moved the binaries to a
> folder on desktop.
>
>  2) Copied the /usr/local/etc/ndn/nfd.conf file to /usr/local/etc/ndn/host1-nfd.conf
> and changed the sock file name to /var/run/host1-nfd.conf.
>
>  3) Started nfd from that folder using the following:
>  ./nfd --config /usr/local/etc/ndn/host1-nfd.conf &
>
>  ./nfd-status ​ and ./nfdc works properly.
>
>  4) I repeat the procedure for the other node.
>
>  I was wondering if there is an easy way?  Vince has directed me to how
> Obaid was running it:
>  http://redmine.named-data.net/issues/1790​
>
>  Ashlesh
>
>
>
> _______________________________________________
> Nfd-dev mailing list
> Nfd-dev at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev
>
>
>   _______________________________________________
> Nfd-dev mailing list
> Nfd-dev at lists.cs.ucla.edu
> http://www.lists.cs.ucla.edu/mailman/listinfo/nfd-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/nfd-dev/attachments/20141209/3fce7ab8/attachment.html>