request expired being raised does not increase the opstats timeout count
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Launchpad itself |
Fix Released
|
Low
|
Colin Watson |
Bug Description
Failure in test lib/canonical/
Traceback (most recent call last):
File "/usr/lib/
testMethod()
File "/usr/lib/
raise self.failureExc
AssertionError: Failed doctest test for xx-opstats.txt
File "lib/canonical/
-------
File "lib/canonical/
Failed example:
report()
Differences (ndiff with -expected +actual):
503s: 1
5XXs: 1
http requests: 1
requests: 1
- timeouts: 1
This occurs if the timeout tracer starts raising RequestExpired rather than TimeoutError, because the *tracer* does the opstats incrementing, rather than publication machinery.
I'm fixing this specific example by catching both exceptions, but other timeouts like the google search service timeout are probably not being recorded accurately.
Related branches
- Ioana Lasc (community): Approve
-
Diff: 372 lines (+106/-33)4 files modifiedlib/lp/app/stories/basics/xx-opstats.txt (+90/-28)
lib/lp/services/webapp/errorlog.py (+4/-1)
lib/lp/services/webapp/publication.py (+9/-4)
lib/lp/services/webapp/servers.py (+3/-0)
- Ioana Lasc (community): Approve
-
Diff: 52 lines (+4/-5)2 files modifiedlib/lp/services/webapp/adapter.py (+0/-5)
lib/lp/services/webapp/publication.py (+4/-0)
Changed in launchpad-foundations: | |
importance: | Undecided → Medium |
status: | New → Triaged |
Changed in launchpad: | |
importance: | Medium → Low |
Changed in launchpad: | |
status: | Triaged → In Progress |
assignee: | nobody → Colin Watson (cjwatson) |
Changed in launchpad: | |
status: | In Progress → Fix Committed |
summary: |
- request expired being raised does not increased the opstats timeout - count + request expired being raised does not increase the opstats timeout count |
Changed in launchpad: | |
status: | Fix Committed → Fix Released |