This is an update to give a quick update of the Spark calling issues experienced in recent days.
The upstream carrier transiting calls from Spark to our network has tracked the issue down to a memory resource issue on one of their PSTN gateway nodes. Previous attempts to resolve the issue only restarted the server process. However, it was determined that this was not freeing up the memory resources on the underlying operating system node.
At 6pm this evening they performed a full reboot of the server node (after failing the traffic over to a secondary host) and this appears now to have resolved the memory resource problems.
We are tentatively hopeful that this will now bring stability back to the affected platform and there will be no further ongoing related issues.
We thank our customers and partners for their patience relating to this issue.