1482
79%
master: 5%

Ran 10 May 2017 12:32AM UTC

Jobs 3

Files 92

Run time 407min

Badge

Embed ▾

pending completion

Build # 1482

Build Type

push

travis-ci

Committed by

josephharrington

Commit Message

Add timeout to all outgoing http requests

Currently if a slave becomes unresponsive the master service will
freeze up for a long time (or forever). The slave allocation loop
will hang trying to contact a slave and no builds will be able to
start.

This adds a timeout to all requests between the master and slaves
and corresponding error handling. I made the timeout on
slave->master calls longer because we don't currently have great
metrics and have had issues with request load on the master in the
past (which we should improve separately).

Run Details

2978 of 3787 relevant lines covered (78.64%)

0.82 hits per line

Jobs

ID	Job ID	Ran	Coverage
1	1482.1 (TARGETS="lint test-unit-with-coverage")	10 May 2017 12:33AM UTC	76.68	Travis Job 1482.1
2	1482.2 (TARGETS="test-integration-with-coverage")	10 May 2017 12:32AM UTC	5.62	Travis Job 1482.2
3	1482.3 (TARGETS="test-functional test-unit-via-clusterrunner")	10 May 2017 12:35AM UTC		Travis Job 1482.3

Source Files on build 1482

Detailed source file information is not available for this build.

box / ClusterRunner / 1482
79%
master: 5%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 1482

box / ClusterRunner / 1482 79% master: 5%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Jobs

Source Files on build 1482

box / ClusterRunner / 1482
79%
master: 5%

README BADGES
x