Canadian telecom giant Research in Motion declared it’s nearly-week long global BlackBerry outage to be over in a press conference at 10 a.m. ET on Thursday, but said it was still struggling to identify exactly what caused the catastrophic failure in the first place.
“All services are back up globally,” confirmed RIM founder and co-CEO Mike Lazaridis in the teleconference.
Yet when asked by reporters what exactly produced the “core switch failure” that lead to the outage that began Monday in Europe, Africa and the Middle East, Lazaridis said that the company was still working to identify the root cause.
“We don’t know why the switch failed in the way that it did and why it didn’t switch over to it’s redundant pair,” Lazaridis said, referring to earlier statements from the company noting that it had experienced the failure of it’s backup switch as well. “We do know it is an error most likely caused by hardware.”
Lazaridis walked reporters through the exact timeline of the events, noting that on Monday, the main “high capacity core switch designed to protect [RIM server] infrastructure failed, causing outages” in Europe. When the backup switch failed, the outages rippled outward to other continents. Even when RIM was able to restore BlackBerry service in Europe, it faced an enormous backlog of data that prevented it from restoring services more quickly to all customers.
The embattled founder, who is facing a torrent of angry customers calling for compensation as the new iPhone 4S hits stores Friday, said his company had spent the past week working tirelessly to remedy the company’s largest outage ever and was only now turning to the matters of identifying the specific hardware issue and of working with wireless service companies around the world to address customer complaints.
“We plan to come back to customers very soon on that [issue of compensation],” Lazaridis said.
Still, he wasn’t ready to shift the blame to any of BlackBerry’s business partners, even when prodded by reporters. He explained that the company was working with vendors to “correct the cause” of the switch failure.
“There are multiple vendors responsible for the core switch,” Lazaridis explained, “Until we have the root cause identified, it would be premature to identify the vendors.”
Lazaridis added that he couldn’t comment on how long the root cause analysis would take, but admitted the company was baffled.
“Systems like this don’t fail this way, they’re designed not to fail this way,” he added.
