This morning, around 9 a.m., it was impossible to connect to many French sites. According to the numerous Down Detector reports, this is due to an outage at OVH. Work in progress on the material would be at the origin. Very bad news for the host who has just gone public.
General panic this morning on the French web. Within a minute, several hundred sites refused to load. At Phonandroid, we have not been spared. Unable to load the pages or even publish our articles. Usually, this kind of failure only means one thing: OVH is having technical problems.
At the start of the day, the host announced a maintenance between 9 am and 10:30 am. Many Internet users then affirmed that this was the source of the problem, although OVH had not yet at that time. nothing confirmed. On Down Detector and Twitter, reports are increasing. A few months after the fire that ravaged the company’s data centers, thousands of users no longer have access to their work tool.
The French web down after an OVH maintenance
“What could be better than starting on Wednesday with a general blackout at OVH?“, sums up a user on Down Detector. “Work has been underway since this morning on Roubaix 4: intervention on network equipment such as bays and switches”, indicates another. “The cut is widespread and no service is responding for the moment. Let us be patient then. “
A few minutes before the start of the problems, the host claimed on Twitter to be “On the bridge to answer your requests”. A coincidence that might not be one, then. The information was finally confirmed by its CEO Octave Klaba: “Following a human error during the reconfiguration of the network on our DC at VH (US-EST), we have a problem on the whole backbone. We are going to isolate the DC VH then fix the conf ”.
Subsequently, Octave Klaba specifies that the maintenance was aimed at strengthening the security of the servers following the proliferation of attacks: “In recent days, the intensity of DDoS attacks has increased a lot. We have decided to increase our DDoS processing capacity by adding new infrastructures in our DC VH (US-EST). A bad configuration of the router caused the failure of the network ”.
The OVH failure was repaired after one hour
Finally, around 10:30 am, all the sites affected by the outage reappeared online. On Twitter, Octave Klaba explains that it was necessary to cut “theThe router with the wrong D2 configuration ″ to get things back to normal. Unfortunately for the company, it didn’t take longer for the #ovhdown hashtag to hit Twitter trending.
Moreover, although the blackout is now a thing of the past, many on the social network continue to react to the event. It must be said that recently, Internet users have seen failures follow one another. The most important is undoubtedly the one that affected all Facebook applications last week. To pass the time, some say that the person responsible for this failure would have been hired by OVH …
Which sites were affected by the blackout?
As with the fire in OVH’s Strasbourg data centers last March, this failure affected a large number of French sites. On Down Detector, several of them have spoken to warn their users of ongoing maintenance. Among the most important, we find in particular Interflora (which moreover had fun by declaring that its site had “plant”), Strasbourg airport, Clublic, the data.gouv.fr site and even Simplified Procedures.