Test failure with current head

Test failure with current head

[Nick Dokos]

I just updated to release_8.3beta-1286-g20795fd.

``make test'' shows a failure on test 226:

--8<---------------cut here---------------start------------->8---
   passed  225/592  test-org-babel/src-block-regexp
Updating dynamic block `clocktable' at line 2...
Updating dynamic block `clocktable' at line 2...done
Test test-org-clock/clocktable-until-now backtrace:
  (if (unwind-protect (setq value-2215 (apply fn-2213 args-2214)) (set
  (let (form-description-2217) (if (unwind-protect (setq value-2215 (a
  (let ((value-2215 (quote ert-form-evaluation-aborted-2216))) (let (f
  (let ((fn-2213 (function equal)) (args-2214 (list "| Headline       
  (lambda nil (let ((fn-2213 (function equal)) (args-2214 (list "| Hea
  ert--run-test-internal([cl-struct-ert--test-execution-info [cl-struc
  ert-run-test([cl-struct-ert-test test-org-clock/clocktable-until-now
  ert-run-or-rerun-test([cl-struct-ert--stats "\\(org\\|ob\\)" [[cl-st
  ert-run-tests("\\(org\\|ob\\)" #[385 "\306\307\"\203G\211\211G\310
  ert-run-tests-batch("\\(org\\|ob\\)")
  ert-run-tests-batch-and-exit("\\(org\\|ob\\)")
  (let ((org-id-track-globally t) (org-test-selector (if org-test-sele
  org-test-run-batch-tests("\\(org\\|ob\\)")
  eval((org-test-run-batch-tests org-test-select-re))
  command-line-1(("--eval" "(setq vc-handled-backends nil org-startup-
  command-line()
  normal-top-level()
Test test-org-clock/clocktable-until-now condition:
    (ert-test-failed
     ((should
       (equal "| Headline                     | Time    |       |\n|------------------------------+---------+-------|\n| *Total time*                 | *25:00* |       |\n|------------------------------+---------+-------|\n| Relative times in clocktable | 25:00   |       |\n| Foo                          |         | 10:00 |\n| Bar                          |         | 15:00 |\n"
	      (org-test-with-temp-text "* Relative times in clocktable\n** Foo\n** Bar\n" ...)))
      :form
      (equal "| Headline                     | Time    |       |\n|------------------------------+---------+-------|\n| *Total time*                 | *25:00* |       |\n|------------------------------+---------+-------|\n| Relative times in clocktable | 25:00   |       |\n| Foo                          |         | 10:00 |\n| Bar                          |         | 15:00 |\n"
	     #("| Headline                     | Time    |       |\n|------------------------------+---------+-------|\n| *Total time*                 | *24:33* |       |\n|------------------------------+---------+-------|\n| Relative times in clocktable | 24:33   |       |\n| Foo                          |         |  9:33 |\n| Bar                          |         | 15:00 |\n" 2 10
	       (face org-table)
	       33 37
	       (face org-table)
	       104 116
	       (face org-table)
	       135 142
	       (face org-table)
	       206 234
	       (face org-table :org-clock-minutes 1473)
	       237 242
	       (face org-table)
	       257 260
	       (face org-table :org-clock-minutes 573)
	       299 303
	       (face org-table)
	       308 311
	       (face org-table :org-clock-minutes 900)
	       349 354
	       (face org-table)))
      :value nil :explanation
      (array-elt 137
		 (different-atoms
		  (53 "#x35" "?5")
		  (52 "#x34" "?4")))))
   FAILED  226/592  test-org-clock/clocktable-until-now
--8<---------------cut here---------------end--------------->8---


The backtrace was edited to get rid of a NUL that gnus complained
about.

Nick

Re: Test failure with current head

[Rasmus]

Nick Dokos <ndokos@gmail.com> writes:

> I just updated to release_8.3beta-1286-g20795fd.
>
> ``make test'' shows a failure on test 226:
>    FAILED  226/592  test-org-clock/clocktable-until-now
>
>
> The backtrace was edited to get rid of a NUL that gnus complained

I know.  I don't really understand what untilnow does and the org
clocktable tests are really obscure to me.  Fell free to do a bisect
master or/and submit a fix.

Rasmus

-- 
A page of history is worth a volume of logic

Re: Test failure with current head

[Nick Dokos]

Rasmus <rasmus@gmx.us> writes:

> Nick Dokos <ndokos@gmail.com> writes:
>
>> I just updated to release_8.3beta-1286-g20795fd.
>>
>> ``make test'' shows a failure on test 226:
>>    FAILED  226/592  test-org-clock/clocktable-until-now
>>
>>
>> The backtrace was edited to get rid of a NUL that gnus complained
>
> I know.  I don't really understand what untilnow does and the org
> clocktable tests are really obscure to me.  Fell free to do a bisect
> master or/and submit a fix.
>

Seems like a heisenbug - first attempt to bisect fingered
20795fd1c4e48096226f3072de8d93bd99d761bf but only because it is the HEAD
and I had declared it as bad: it turns out that `make test' does not
get any failures now. And nothing changed - weird.

Nick

Re: Test failure with current head

[David Engster]

Nick Dokos writes:
> Rasmus <rasmus@gmx.us> writes:
>
>> Nick Dokos <ndokos@gmail.com> writes:
>>
>>> I just updated to release_8.3beta-1286-g20795fd.
>>>
>>> ``make test'' shows a failure on test 226:
>>>    FAILED  226/592  test-org-clock/clocktable-until-now
>>>
>>>
>>> The backtrace was edited to get rid of a NUL that gnus complained
>>
>> I know.  I don't really understand what untilnow does and the org
>> clocktable tests are really obscure to me.  Fell free to do a bisect
>> master or/and submit a fix.
>>
>
> Seems like a heisenbug - first attempt to bisect fingered
> 20795fd1c4e48096226f3072de8d93bd99d761bf but only because it is the HEAD
> and I had declared it as bad: it turns out that `make test' does not
> get any failures now. And nothing changed - weird.

Yes, it's not always failing, but it's been happening for a few days now
on the buildbot:

  http://randomsample.de/org-buildbot

First time it happened was after this commit

  http://orgmode.org/cgit.cgi/org-mode.git/commit/?id=22c652599c8bfedcd27e78d7ad9544826eae7dd0

but it looks harmless. Could very well be something else that triggers
this.

-David

Re: Test failure with current head

[Rasmus]

Nick Dokos <ndokos@gmail.com> writes:

> Rasmus <rasmus@gmx.us> writes:
>
>> Nick Dokos <ndokos@gmail.com> writes:
>>
>>> I just updated to release_8.3beta-1286-g20795fd.
>>>
>>> ``make test'' shows a failure on test 226:
>>>    FAILED  226/592  test-org-clock/clocktable-until-now
>>>
>>>
>>> The backtrace was edited to get rid of a NUL that gnus complained
>>
>> I know.  I don't really understand what untilnow does and the org
>> clocktable tests are really obscure to me.  Fell free to do a bisect
>> master or/and submit a fix.
>>
>
> Seems like a heisenbug - first attempt to bisect fingered
> 20795fd1c4e48096226f3072de8d93bd99d761bf but only because it is the HEAD
> and I had declared it as bad: it turns out that `make test' does not
> get any failures now. And nothing changed - weird.

I cannot reproduce the heisenbug.

Maybe it depends on the system time.  Changing my clock (via gnome
control center):

- At 12:59 there's an error.
- At 13:00 there is not.

Rasmus

-- 
ツ

Re: Test failure with current head

[Rasmus]

David Engster <deng@randomsample.de> writes:

> Nick Dokos writes:
>> Rasmus <rasmus@gmx.us> writes:
>>
>
>>> Nick Dokos <ndokos@gmail.com> writes:
>>>
>>>> I just updated to release_8.3beta-1286-g20795fd.
>>>>
>>>> ``make test'' shows a failure on test 226:
>>>>    FAILED  226/592  test-org-clock/clocktable-until-now
>>>>
>>>>
>>>> The backtrace was edited to get rid of a NUL that gnus complained
>>>
>>> I know.  I don't really understand what untilnow does and the org
>>> clocktable tests are really obscure to me.  Fell free to do a bisect
>>> master or/and submit a fix.
>>>
>>
>> Seems like a heisenbug - first attempt to bisect fingered
>> 20795fd1c4e48096226f3072de8d93bd99d761bf but only because it is the HEAD
>> and I had declared it as bad: it turns out that `make test' does not
>> get any failures now. And nothing changed - weird.
>
> Yes, it's not always failing, but it's been happening for a few days now
> on the buildbot:
>
>   http://randomsample.de/org-buildbot
>
> First time it happened was after this commit
>
>   http://orgmode.org/cgit.cgi/org-mode.git/commit/?id=22c652599c8bfedcd27e78d7ad9544826eae7dd0
>
> but it looks harmless. Could very well be something else that triggers
> this.

This is in contrib.  AFAIK it is not covered by make test.

-- 
This is the kind of tedious nonsense up with which I will not put

Re: Test failure with current head

[Nick Dokos]

Rasmus <rasmus@gmx.us> writes:

> Nick Dokos <ndokos@gmail.com> writes:
>
>> Rasmus <rasmus@gmx.us> writes:
>>
>>> Nick Dokos <ndokos@gmail.com> writes:
>>>
>>>> I just updated to release_8.3beta-1286-g20795fd.
>>>>
>>>> ``make test'' shows a failure on test 226:
>>>>    FAILED  226/592  test-org-clock/clocktable-until-now
>>>>
>>>>
>>>> The backtrace was edited to get rid of a NUL that gnus complained
>>>
>>> I know.  I don't really understand what untilnow does and the org
>>> clocktable tests are really obscure to me.  Fell free to do a bisect
>>> master or/and submit a fix.
>>>
>>
>> Seems like a heisenbug - first attempt to bisect fingered
>> 20795fd1c4e48096226f3072de8d93bd99d761bf but only because it is the HEAD
>> and I had declared it as bad: it turns out that `make test' does not
>> get any failures now. And nothing changed - weird.
>
> I cannot reproduce the heisenbug.
>
> Maybe it depends on the system time.  Changing my clock (via gnome
> control center):
>
> - At 12:59 there's an error.
> - At 13:00 there is not.
>

I ran ``make test'' in a loop, 3 iterations per minute,
for about an hour - no (unexpected) failures. Whatever it is,
it's hiding.
-- 
Nick

Re: Test failure with current head

[Rasmus]

Nick Dokos <ndokos@gmail.com> writes:

> I ran ``make test'' in a loop, 3 iterations per minute,
> for about an hour - no (unexpected) failures. Whatever it is,
> it's hiding.

But did you change the clock to before 13:00?  Now it's 10:30 I get the
error.

I get c76fef6b9df3ffbb6a1cf4047f391721c4076caa is the bad commit.  This is
also when the test and untilnow was implemented.

Rasmus

-- 
When the facts change, I change my mind. What do you do, sir?

Re: Test failure with current head

[Nick Dokos]

Rasmus <rasmus@gmx.us> writes:

> Nick Dokos <ndokos@gmail.com> writes:
>
>> I ran ``make test'' in a loop, 3 iterations per minute,
>> for about an hour - no (unexpected) failures. Whatever it is,
>> it's hiding.
>
> But did you change the clock to before 13:00?  Now it's 10:30 I get the
> error.
>
> I get c76fef6b9df3ffbb6a1cf4047f391721c4076caa is the bad commit.  This is
> also when the test and untilnow was implemented.
>

No, I misinterpreted your comment. I thought you meant the hour
transition, not the specific hour. I just repeated it now and I got
the failure (and it is 9AM here).

-- 
Nick

Re: Test failure with current head

[Nicolas Goaziou]

Hello,

Nick Dokos <ndokos@gmail.com> writes:

> I just updated to release_8.3beta-1286-g20795fd.
>
> ``make test'' shows a failure on test 226:
>
>    passed  225/592  test-org-babel/src-block-regexp
> Updating dynamic block `clocktable' at line 2...
> Updating dynamic block `clocktable' at line 2...done
> Test test-org-clock/clocktable-until-now backtrace:

Fixed. Thank you.

Regards,

-- 
Nicolas Goaziou