[Spce-user] Spce set up question!

Andreas Granig agranig at sipwise.com
Sat Apr 12 14:45:30 EDT 2014


Hi,

No way to run the SPCE v3.x on a t1.micro due to increased RAM
requirements. It should be fine with v2.8 LTS though. It might need some
tweaking decreasing number of apache processes, mysql innodb buffer pool
size and kamailio pkg/shm size, but it's known to work (unofficially).

Starting with the next release, we're officially providing ready-to-run
AMIs, but then again they will require a certain instance size, and
t1.micro will be too small. I'm running some m1.medium for internal
tests, they are perfectly fine.

Andreas

On 04/12/2014 07:21 AM, Alfonso Lopez wrote:
> So I did a tail of the system log, and I get the following issue when trying to restart the service.
> 
> Out of memory: kill process perl or sacrifice child.
> 
> Mind you, this is for testing purposes and we should only have max 5 users.. 
> 
> Sent from my iPhone
> 
>> On Apr 11, 2014, at 9:45 PM, Alfonso Lopez <soccercrzyboy21 at yahoo.com> wrote:
>>
>> Andrew thanks for your help again.
>>
>> Anyone have some tips for setting it up on amazon t1 micro? I tried to restart the ngcp panel and it said failed to restart.
>>
>> Thanks!
>>
>> Greenkirby21
>>
>> Sent from my iPhone
>>
>>> On Apr 8, 2014, at 2:18 PM, Alfonso Lopez <soccercrzyboy21 at yahoo.com> wrote:
>>>
>>> Thank you so much Andrew! Appreciate your help!
>>>
>>> Sent from my iPhone
>>>
>>>> On Apr 8, 2014, at 2:06 PM, Andrew Pogrebennyk <apogrebennyk at sipwise.com> wrote:
>>>>
>>>> Hello,
>>>>
>>>> it takes sometimes several days to reply, moreover "400 Bad Request: SSL
>>>> certificate error" doesn't pop up here very often.
>>>> The admin login URL is https://<ip>:1443/login/admin, I don't know what
>>>> could have caused the 400 error, try restarting nginx and ngcp-panel.
>>>>
>>>> The networking section in config.yml is not used any more. For SIP and
>>>> RTP however, you need to set advertised IP to your elastic IP:
>>>> ngcp-network --set-interface=eth0 --advertised-ip=x.y.z.w
>>>> ngcpcfg apply
>>>>
>>>> For webrtc, you need to enable ICE in domain preference use_rtpproxy
>>>> (select there "with rtpproxy as only" or "as extra ICE candidate".
>>>> That's it, use WS URI: ws://<YOUR_IP>:5060/ws or
>>>> wss://<YOUR_IP>:1443/wss/sip as your entry point.
>>>>
>>>> Note: this works for browser-to-browser calls, but if you have a mixed
>>>> environment you should also enable transcoding. E.g., I would set for
>>>> every new webrtc user we need to set those preferences:
>>>> srtp_transcoding: Prefer SRTP
>>>> rtcp_feedback: Prefer AVPF
>>>>
>>>> And then set defaults for all (non-webrtc) users in the domain:
>>>> srtp_transcoding: Force RTP
>>>> rtcp_feedback: Force AVP
>>>> Also you the same settings as in domain for PSTN termination.
>>>>
>>>> Regards,
>>>> Andrew
>>>>
>>>>> On 08/04/14 20:21, Alfonso Lopez wrote:
>>>>> Sorry if this is spam! I thought I sent this but I don't see any reply. My apologies, I just joined the list so I wasn't sure if it worked the first time or not!
>>>>>
>>>>> I was wondering if there was a tutorial or anything for setting up sipwise CE on amazon ec2? I am working on a school project and need a easy to set up public server, which is why I chose sipwise!
>>>>>
>>>>> I am having a couple of issues. When I follow the handbook, I get to the part where I need to login to the web console. If I just type in the public IP onto my web browser, I get to the "Self-Administration Login" website. However, when I try the default username and password (administrator), I get "Login failed - Please verify username and password".
>>>>>
>>>>> If I try https://<ip>:1443, I get a pop up to accept the certificate. I hit ok, but then I immediately get 400 Bad Request: SSL certificate error.
>>>>>
>>>>> I noticed after the install (which I didn't get any major failures for), the /etc/ngcp-config/config.yml did not contain "networking". I added the following to it based on another mailing list I came across:
>>>>>
>>>>> networking:
>>>>> iaddress: 127.0.0.1
>>>>> aaddress: <Public IP address given by amazon ec2>
>>>>> eaddress: <ip address for eth0 given by "ip a" command
>>>>>
>>>>> Any help would be appreciated!! Thank you so much 
>>>>>
>>>>> Instance Information:
>>>>> Debian Squeeze 7.4 64 Bit
>>>>> Type: t1.micro
>>>>> sipwise mr 3.2.1
>>>>>
>>>>>
>>>>> Finally, if I want to use webrtc, do I need to configure anything in particular?
>>>>>
>>>>> Thanks!
>>>>>
>>>>> Greenkirby21
>>>>> Sent from my iPhone
> 
> _______________________________________________
> Spce-user mailing list
> Spce-user at lists.sipwise.com
> http://lists.sipwise.com/listinfo/spce-user
> 




More information about the Spce-user mailing list