Some flakiness on builds expected

cintiadr · February 14, 2018, 11:27am

Hi,

We are receiving some alerts about DNS failures in most of our servers. That means that a few builds will fail.

If the provider doesn’t fix the problem in the next 7h, I will be changing the DNS to a public one in all servers.

cintiadr · February 14, 2018, 11:43am

@burke / @pascal / @jeremy is something stop working because of this on the next hours, please feel free to edit /etc/resolv.conf to have 8.8.8.8 and 8.8.4.4 .

Just let me know which machines you had to do that, so I will make sure to clean them up.

Things that might fail: ID (we have multiple machines) and atlassian products (database).

raff · February 15, 2018, 2:08pm

https://ci.openmrs.org/browse/OCL-OA-142 is it related?

r0bby · February 17, 2018, 12:31pm

Potentially and highly likely – but that’s me guessing.

cintiadr · February 18, 2018, 8:50am

Sorry I didn’t respond here earlier, busier than usual week.

Yes, @raff, most likely it was related. The problem has been resolved by our lovely provider

dkayiwa · February 18, 2018, 5:30pm

@raff @cintiadr since the problem was resolved by our lovely provider, does it mean that the current failure is caused by something else? https://ci.openmrs.org/browse/OCL-OUT-90

cintiadr · February 20, 2018, 8:49am

Maybe, I have no idea what’s happening there. Should I restart docker host, @raff?

raff · February 20, 2018, 12:23pm

@cintiadr, no need. I need to fix code to make tests pass.