Package: libdr-tarantool-perl
Severity: serious
Version: 0.35-1
Hi,
while investigating #704267 I noticed the following:
On a not that slow, but virtual machine with a load < 1 and running
amd64 Sid, the test t/110-netsplit-readahead.t, subtest "delay maximum"
occassionally fails (and hence the package FTBFS) because it seems to
expect that some test takes at maximum 200ms more than then minimum
amount of time it expects to be needed.
In the following example the test ran just 0.0024 seconds too long and
hence failed:
# Failed test 'delay maximum'
# at t/110-netsplit-readahead.t line 141.
# '0.26538348197937'
# <
# '0.263'
# Looks like you failed 1 test of 24.
t/110-netsplit-readahead.t ..
Sometimes the difference was over 0.1 seconds, too.
Running t/110-netsplit-readahead.t many times in a row, the test failed
for > 25% of the runs for me.
How to reproduce:
p=0; f=0; for i in `seq 1 100`; do prove t/110-netsplit-readahead.t && (( p++
)) || (( f++ )); done; echo p=$p f=$f
(Since t/110-netsplit-readahead.t does not exist in the version of
libdr-tarantool-perl in Wheezy, only the version in Sid is affected and
704267 is a different issue.)
Regards, Axel
--
,''`. | Axel Beckert <a...@debian.org>, http://people.debian.org/~abe/
: :' : | Debian Developer, ftp.ch.debian.org Admin
`. `' | 1024D: F067 EA27 26B9 C3FC 1486 202E C09E 1D89 9593 0EDE
`- | 4096R: 2517 B724 C5F6 CA99 5329 6E61 2FF9 CD59 6126 16B5
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
