Earlier today we experienced an intermittent outage for a period of around 40 minutes in our primary client service region - Amazon Web Service's us-west-2 - caused by an internet connectivity issue on Amazon's end.
We started receiving alerts from our automated external monitoring systems 20 minutes before AWS issued their outage event (at around 7:21am Pacific time) as the connectivity started to "flap" (which is when a service or link goes down but then comes up and then goes down again). The flapping presented as an intermittent outage, and our team followed our incident management procedure with our engineers working to diagnose the problem within our infrastructure, and we fairly quickly inferred that the issue was more widespread and likely related to AWS connectivity. At 7:41am Pacific, 20 minutes after we started working on the issue, the AWS team posted an update reflecting they had an Internet Connectivity operational issue (see screenshot below). Around 18 minutes later, we saw network stability return, confirmed by our external monitoring services reporting solid green connections at 7:58am Pacific.
We're waiting for a more detailed report on the issue, and given the prominence of this AWS region, plenty of other sites also experienced outages too. We apologize for the inconvenience this upstream network interruption caused to our clients.
Additionally, with all of the information in the media around the Log4j vulnerability (CVE-2021-44228, also known as Log4Shell), we thought it would be worth taking a moment to confirm that Accelo is not affected by this vulnerability.
When the news about this vulnerability was published on Friday (US time), our engineering team immediately checked the our infrastructure manifests to confirm whether we were affected by this issue, and confirmed within a few minutes that we are not.