You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We have been having a lot of trouble with random test failures on Travis, many of which are caused by the 10-minute timeout when no output is produced. This seems to have become frequent over the last three months, and may be related to similar problems encountered by other people: travis-ci/travis-ci#8162, travis-ci/travis-ci#8165 and flux-framework/flux-core#1141.
Following the hints, especially in the last of those links, I have put in PR #709 which changes the Travis environment to a VM instead of a container. Since it's a one-liner affecting only the test configuration, I'll merge it as soon I am sure the tests pass.
However, I'll leave this issue here for a few days because there is something slightly worrying about needing to do this. The failures all seem to occur when compiling the actual test modules: usually the last line reported is Compiling PersistentTest.hs. Could it be something to do with memory usage (the VMs have more than the containers)?
The text was updated successfully, but these errors were encountered:
We have been having a lot of trouble with random test failures on Travis, many of which are caused by the 10-minute timeout when no output is produced. This seems to have become frequent over the last three months, and may be related to similar problems encountered by other people: travis-ci/travis-ci#8162, travis-ci/travis-ci#8165 and flux-framework/flux-core#1141.
Following the hints, especially in the last of those links, I have put in PR #709 which changes the Travis environment to a VM instead of a container. Since it's a one-liner affecting only the test configuration, I'll merge it as soon I am sure the tests pass.
However, I'll leave this issue here for a few days because there is something slightly worrying about needing to do this. The failures all seem to occur when compiling the actual test modules: usually the last line reported is
Compiling PersistentTest.hs
. Could it be something to do with memory usage (the VMs have more than the containers)?The text was updated successfully, but these errors were encountered: