Closed Bug 1355207 Opened 7 years ago Closed 5 years ago

Intermittent TEST-UNEXPECTED-TIMEOUT | netwerk/test/httpserver/test/test_host.js | Test timed out

Categories

(Core :: Networking, defect, P3)

Product:
Core
Component:
Networking
Type:
defect

Tracking

()

Status:
RESOLVED FIXED
Milestone:
mozilla66
Tracking Flags:
Tracking Status
firefox65 --- fixed
firefox66 --- fixed

People

(Reporter: intermittent-bug-filer, Assigned: valentin)

Details

(Keywords: intermittent-failure, Whiteboard: [necko-triaged][retriggered][stockwell fixed])

Attachments

(2 files)

Bug 1355207 - Improve test_host.js and add diagnostic info to figure out where it times out
47 bytes, text/x-phabricator-request
Details | Review
Bug 1355207 - Wait for http server shutdown and disable test_host.js on OSX r=dragana
47 bytes, text/x-phabricator-request
Details | Review
Whiteboard: [necko-backlog]
Status: NEW → RESOLVED
Closed: 7 years ago
Keywords: bulk-close-intermittents
Resolution: --- → INCOMPLETE
Priority: -- → P3
Whiteboard: [necko-backlog] → [necko-triaged]
This test seems to be intermittent because we use hardcoded ports for the HTTP server - mainly 4444. Not sure if the test times out because the port is taken (doesn't seem so from the logs), but given that it's OSX only, it could have the same cause as bug 1481963 comment 2.
Looking into it. It may be that https://hg.mozilla.org/integration/autoland/rev/29385deef9ea018dd480de4f4c9ceaa4b1308c5d is to blame. But the test is badly written anyway - I'm inclined to disable it on OSX, as I found out in other bugs, sometimes connections to localhost on an unopened port are not rejected, they just time-out :|
Assignee: nobody → valentin.gosu
Pushed by valentin.gosu@gmail.com:
https://hg.mozilla.org/integration/autoland/rev/28248b27989e
Improve test_host.js and add diagnostic info to figure out where it times out r=kershaw
Keywords: leave-open
This fails a lot for TB as well, I'd love to see it fixed.
The test intermittently fails on OSX. For some reason start(PORT) doesn't always fail when PORT is in use, so we timeout because the handler is never called. I suspect it's because of how SO_REUSEADDR and SO_REUSEPORT are handled on sockets sever-side on OSX. Since httpd is used for unit tests, there is little risk in disabling the test on OSX.
The alternative is to check for a timeout, and in that case switch to another port - which is cumbersome, and not the purpose of the test.
Pushed by valentin.gosu@gmail.com:
https://hg.mozilla.org/integration/autoland/rev/3ecdb3109dc2
Wait for http server shutdown and disable test_host.js on OSX r=dragana
Status: REOPENED → RESOLVED
Closed: 7 years ago → 5 years ago
Resolution: --- → FIXED
status-firefox65: --- → affected
status-firefox66: --- → fixed
Keywords: bulk-close-intermittents, leave-open
Target Milestone: --- → mozilla66
https://hg.mozilla.org/releases/mozilla-beta/rev/1e8a7d7842cb
https://hg.mozilla.org/releases/mozilla-beta/rev/df270c166d44
status-firefox65: affected → fixed
Flags: in-testsuite+
Whiteboard: [necko-triaged] [retriggered][stockwell needswork:owner] → [necko-triaged][retriggered][stockwell fixed]
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: