On October 28 2016 a failure in network equipment by a upstream provider caused reachability issues for some of our clients. After 35 minutes the issue got resolved at 9:25am PDT. We are reviewing our policies to identify potential automatic remedies to avoid issues like this in the future.
Aside from this, all system worked normal and within specifications.
For the month of September two issues are worth noting:
As mentioned in an earlier post there was a network event that caused brief connectivity issues for some clients during a 20 minute time window. The cause for this issue has been addressed.
Unrelated to this is the second one:
Since its inception INTERDUBS has been a steadily increased use and growth. We plan accordingly and add capacity as it is needed. During September growth was sudden during a couple of days so that the system got closer to the limits than usual.
For some clients this meant slower than usual page response times. We took steps to mitigate this effect and are happy with the results we are seeing after we stepped up our expansion efforts. We also started to work on a process that will allow us to scale much more quickly and efficiently around the needs of our growing customer base.
Systems worked normally and well for 30 days.
During two distinct phases On August 31st some clients encountered unusual performance issues.
All systems worked perfectly during the month of July.
June – all was perfect. Again.
No problems of any kind in May. Again. This is getting boring. Actually it is not.
A working system only looks dull. Behind the scenes there is quiet a bit of interesting work going into this.
It is fun work. Specially when, as in May, it creates a perfect service record.
Everything worked perfectly in April.
Behind the scenes we worked on the usual amount of improvements to serve our clients needs better.
All INTERDUBS data centers performed flawless in March.
On March 21st around 9am PST some clients had troubles connecting to INTERDUBS.
We had redundant DNS vendors. DNS is the system that translates domain names (like INTERDUBS.COM) to IP addresses.
In theory it appeared to be a good idea to have two vendors for this crucial information. Unfortunately one vendor failed in a way that caused system to think that our web address was not reachable.
We have corrected this problem conceptually.
On February 24th of our upstream providers performed maintenance operations. Due to an error on their end service was impacted for some of our clients starting at 9:11pm. At 9:25pm the issue had been mitigated from our end.
Aside from that there were no issues in February.
Just more infra structure and feature upgrades to the system. Like every month.
There were no issues of any kind during January.
INTERDUBS status 