It's looking better in the SIP world. I have the 5501 currently routing 50 calls at a rate of 5 calls a second. Just to make sure that there's no network losses etc I also have a softphone connected through the 5501 to the P4 so that I can hear the MOH. I guess technically that makes 51 calls.
I've noted that the temperature of the processor on the 5501 climbs rapidly when its under this load (erm ... duh!!) almost to the point where one cannot touch it. Whilst all this is going on I am still seeing the occasional bitch message about the T1 (which is up but not in use in this test) which is making me suspect that something is wrong in my T1 chain. Time to break out another P4. Mark On 07/16/2010 06:44 AM, Mark Phillips wrote: > Well, it doesn't look good. > > My test rig is as follows .... > > P4 3GHz/2GB/Digium TE110P T1 card running latest 1.4 sources from Digium > site. T1 card configured as PRI_NET and internal timing. span=1,0,0,esf,b8zs > > NET5501-70 (500MHz/512MB) with TE110P running Astlinux 0.7.2. T1 card > configured as PRI_CPE and external timing. span=1,1,0,esf,b8zs > > Both machines are connected to each other via a T1 cross-over cable. > > SIPp is running on my desktop linux machine and is pointed to the 5501 > which in turn relays the calls up the T1 to the P4. All calls terminate > into the MOH app on the P4 and last for 10 seconds before hanging up. > > So far I'm only able to get 8 calls going before I start to get this > error chan_dahdi.c:9099 pri_dchannel: PRI got event: HDLC Abort (6) on > Primary D-channel of span 1 showing up on both sides of the T1. > > Question is "which end of the link is causing this?" > > Are my tests flawed? Should I be routing the calls onwards out of the P4 > and back to a UAS instance on another machine? > > I'm gonna pull the T1 board from the 5501 and put it in a bigger machine > (prolly another P4) and try it there. I'm trying to work out if the 5501 > is running out of horsepower or something (PCI bus not wide enough?) > > Mark > > On 07/15/2010 03:35 PM, Mark Phillips wrote: >> Looks like I'll be busy then :) >> >> On 07/15/2010 09:06 AM, Darrick Hartman wrote: >>> Mark, >>> >>> I'm not aware of any stress tests other than the ones you've provided in >>> the past ;) >>> >>> Lonnie may have done some, but not likely with a T1 card. >>> >>> Darrick >>> >>> On 07/15/2010 07:51 AM, Mark Phillips wrote: >>>> Hi all, >>>> >>>> Has anyone done any stress tests using the 5501 and a T1 card? >>>> >>>> I need some data for a project I'm doing and was hoping to get a head >>>> start. >>>> >>>> If not then I'll do it myself and post the results. >>>> >>>> Mark >>> >> >> ------------------------------------------------------------------------------ >> This SF.net email is sponsored by Sprint >> What will you do first with EVO, the first 4G phone? >> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first >> _______________________________________________ >> Astlinux-users mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/astlinux-users >> >> Donations to support AstLinux are graciously accepted via PayPal to >> [email protected]. >> > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by Sprint > What will you do first with EVO, the first 4G phone? > Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first > _______________________________________________ > Astlinux-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/astlinux-users > > Donations to support AstLinux are graciously accepted via PayPal to > [email protected]. > ------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ Astlinux-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/astlinux-users Donations to support AstLinux are graciously accepted via PayPal to [email protected].
