We have several remote sites connecting back to a central site through AT&T= 's switched metro ethernet service, CSME. We have several CSME connections = at the central side, each providing 100 Mbps bandwidth to about 10 sites ea= ch.
Recently, they're having us convert over to a new service, Opt-E-Man. Since= , some of the central site connections were not set up correctly in the beg= inning, the existing central side connections are needing to have the sites= they serve moved elsewhere temporarily so they can be reconfigured. This p= rocess is underway, but we're running into a problem. Two of the sites whic= h were moved temporarily and have been moved back are losing connectivity f= or periods of time (these sites are now Opt-E-Man sites). They come and go = seemly randomly. As far as I can see, nothing is being changed on either si= de.
When the sites become unreachable from the central side (pings are failing)= , I can still connect to other sites served by the same central side connec= tion and get to the "unreachable" site. Once on the "unreachable" site, eve= rything looks fine. The default gateway is set correctly (but can't be ping= ed), the results of 'show ip eigrp neighbors' seems fine. And then suddenly= , without changing anything, that site will return to normal.
AT&T is not considering this to be a problem on their equipment. The config= uration on our equipment is basically exactly the same as it's always been.= Two sites have problems, and others have no problems at all.
Is there some way I can prove it is AT&T's problem or if it is on our end?