Our Cloudflare vendor was having some odd problems which began yesterday evening.  These issues seemed largely resolved by Friday morning but they appear to have impacted our AVS API for a period of time today.  We are happy to report that the issues specific to the API have been resolved with Cloudflare.

We estimate that systems are within 24-36 hours from being caught up on processing of new detections within normal parameters.  Most networks are already back to normal.  Next week AdSphere and AVS reporting should be back to normal.  We appreciate everyone’s patience this week and wish everyone a relaxing weekend!

We still have a group of about 30 networks that are not reporting airings since around 4AM on 9/29.  We are being told by our ACR provider that airings data during this time is not lost but the system is still playing catch up.  Apparently there are a lot of thumbnail images and airing level videos to process on the impacted servers.  When we brought them back online, some are having to run file disk check operations to check for, and resolve any possible issues of, data corruption.   While this type of recovery mode is typical, because DRMetrix has millions of airing level thumbnail image files on various servers, a very large number of files are having to be checked which is delaying completion of recovery mode on some servers.  When these servers finish recovery mode, they will begin processing the back airings data and things will return to normal.  We expected this process would have been completed on all servers by now but some servers that are impacting these 30 networks are still in recovery mode and/or currently back logged processing data and trying to catch back up to processing live airings data.

We are in conversations with our co-location vendor about exactly what happened at their facility that caused these issues.  Apparently many companies they service have been negatively impacted.  While we are constantly working to improve our systems, and to achieve a high level of uptime service, when something out of our direct control like this happens, it is extremely discouraging.  We thank all of our customers for your continued patience, support, and understanding.  We deeply regret and apologize for this situation and will do our very best to work with our co-location vendor to find solutions to keep this type of situation from occurring again.

Update 9:30PM ET – Progress!  More of the servers came out of recovery mode today and more data is being reported as the system continues to play catch up.  Some networks that had no data for today or yesterday are now caught up through yesterday and are processing today’s airings.  Some are a bit further behind but quickly catching up.   We’re down to about 10 networks that are still in recovery mode which is a big improvement from yesterday.  More details to follow tomorrow.  We appreciate everyone being so understanding and patient.  Thank you!!

Systems are still in catch up mode following yesterday’s unexpected issues.  Not all networks are showing data per norm in network log view, some thumbnail images are offline, etc.  We understand that a lot of back data is still being processed as the servers are in catch up mode.  We anticipate a lot of progress will be made today and that hopefully data in AdSphere and AVS will quickly return to normal.  More details to follow.

Update 12:09AM ET – While many networks have returned to normal, along with airing level videos and thumbnail images, the system is still playing catch up on other networks that may not be showing recent airings.   We hope the system will finish processing back data and recent data on these lagging networks soon.

DRMetrix is experiencing issues this morning at our main co-location data center that are impacting our services and others.  We are working to get AdSphere back online.  Thank you for your patience this morning.  Updates will be made available as soon as we know more.

Update 12:17PM ET – AdSphere services are back online.  Our co-location center experienced a critical power issue within their infrastructure early this morning that took some of our equipment offline.  We are working to bring all of our capture servers back online and are still assessing the situation and what impacts it may have had on our video feeds.  We expect some impact on our airing detections for today.  More information as it becomes available.

Update 7:55PM ET – We regret to inform that today’s issues impacted some of our video streams from 7:30a to 11:30a EST.  This will result in lost detections on some networks during this period of time.  We are waiting on full details from our provider as to what happened that caused the issue and what remedial steps will be taken to prevent another occurrence.  It is hard to express the frustration and disappointment we experience when something like this occurs and we offer our humble apologies.

Some videos are not playing properly in AdSphere this morning.   There was a fix done overnight on another video related issue that is to blame and our development team is currently working to resolve the current issue as quickly as possible.

10:55AM PDT Update – Video playback issue has been resolved.  We apologize for any inconvenience.

Our ETL is offline due to a technical issue that our team is working to resolve.  Airings will be delayed getting into AdSphere and AVS until the matter is resolved and the system catches up.  We apologize for this inconvenience.  More information will follow.

2:45PM PDT Update – The issue has been resolved and the ETL is in catch up mode.  Should be caught up in the next couple of hours.

Airings data is delayed today due to an issue with some new ETL code that was deployed yesterday evening.  There have been isolated occurrences of our ETL crashing and causing data delays so our team has redesigned aspects of the ETL software to address such issues.  Yesterday’s deployment of the new code base ran into an issue very early in the AM.  The issue was documented and a software patch was created to address.  The system is currently playing catch up.  We assume it will be back to normal before end of day.  We apologize for any inconvenience.

8/19/20 Update at 7AM PT – The ETL is still playing catch up after completing importing all of the airings through end of calendar day 8/18.  Taking longer than expected to catch up.  More details to follow.

8/19/20 Update at 7:50PM PT – ETL is almost caught up at this hour.  The new ETL code seems to be working well!

Yesterday’s storms across the East Coast have taken a toll on some of the OTA (Over the Air) stations we monitor for broadcast digital networks such as Antenna TV.  The signals from more than one broadcast station across the region have been disrupted due to the storms and damage to equipment and broadcast antennas in some markets.  It may take a few days for us to get Antenna TV back on the air .  More information as it becomes available.

8/12  Update – Antenna TV is back online.  Grit and This TV are currently offline also due to storm damage.  Expect these will be back online in the next few days.  More details to follow.

8/13  Update – It appears that Grit and This TV are also back online.  As the region recovers from the recent storm damage, we’re hoping there will be no further interruptions in signal.

1:49AM ET – The ETL which updates DRMetrix’s airing database had a hiccup and is running about 8 hours behind at the present time.  Airings are being delayed getting into the AdSphere and AVS databases.  We expect the issue to be resolved shortly and the ETL will be in catch up mode for the remainder of the day.  This should not impact any of the data but will delay it’s reporting in our system today.  We apologize for any inconvenience.

Just after 5PM ET today, one of our vendors, Cloudflare, which protects DRMetrix.com, AdSphere™, and AVS™ from DOS (denial of service) attacks went offline.  This is a very rare event which likely is impacting thousands of websites that use Cloudflare’s service.  At the time of this writing, we can’t reach cloudflare.com but we did call their offices and they are aware of the problem and have their engineers working on it.  We’re sure this will make some news later today.  We hope Cloudflare is able to quickly resolve their issues!

5:36PM Update – DRMetrix sites appear to be back online.  During the outage, we checked other popular websites such as uber.com, authy.com, yelp.com, and many others were also offline.

Cloudflare DNS goes down, taking a large piece of the internet with it

The following is from Cloudflare’s website: