make test fails for Redis 2.6.14 fails on Mac OS 10.8.4 #1230

Open
seanthegeek opened this Issue Aug 3, 2013 · 12 comments

Projects

None yet

6 participants

@seanthegeek
!!! WARNING The following tests failed:

*** [err]: Fixed AOF: Server should have been started in tests/integration/aof.tcl
Expected '0' to be equal to '1'
*** [err]: AOF+SPOP: Server should have been started in tests/integration/aof.tcl
Expected '0' to be equal to '1'
*** [err]: AOF+EXPIRE: Server should have been started in tests/integration/aof.tcl
Expected '0' to be equal to '1'
Cleanup: may take some time... OK
make[1]: *** [test] Error 1
make: *** [test] Error 2
@darix
darix commented Aug 7, 2013

this also happens when building on opensuse.

@antirez
Owner
antirez commented Aug 7, 2013

That's a false positive, so your build is fine, thanks for reporting, I'll see if I can reproduce this to fix it.

@darix
darix commented Aug 7, 2013

I can reproduce it 100% and can show you how to reproduce it on a linux system. But i would prefer IRC for that. :)

@darix
darix commented Aug 8, 2013

btw: on the build failure was fixed after adding "ps" to the build environment.

@jbergstroem

@darix A steps to reproduce here would be nice so more people can try it out.

@antirez
Owner
antirez commented Aug 13, 2013

so was that just missing "ps" in the target environment? Thanks.

@darix
darix commented Aug 13, 2013

at least on opensuse. not sure if the osx issue is the same.

maybe the osx problem is parsing the ps output?

@seanthegeek

Output of ps after running make test:

http://pastebin.com/uZX9rz2c

The severs are left running after the tests. I assume that is not the expected behavior.

@moreaki
moreaki commented Aug 19, 2013

Funny, I have just cloned a fresh copy of redis on my MacOSX 10.8.4 and I cannot reproduce the errors you are seeing (maybe they have been fixed):

@surimacpro:.+trials/redis$ git clone https://github.com/antirez/redis.git; cd redis
@surimacpro:.+trials/redis$ make 1>/dev/null 2>&1
@surimacpro:.+trials/redis$ redis-server --version
Redis server v=2.6.14 sha=00000000:0 malloc=libc bits=64
@surimacpro:.+/redis$ uname -a
Darwin surimacpro 12.4.0 Darwin Kernel Version 12.4.0: Wed May  1 17:57:12 PDT 2013; root:xnu-2050.24.15~1/RELEASE_X86_64 x86_64
@surimacpro:.+trials/redis$ sudo make test

[... heaps of output ...]

[33/33 done]: unit/obuf-limits (64 seconds)

                   The End

Execution time of different units:
  1 seconds - unit/printver
  0 seconds - unit/quit
  2 seconds - unit/auth
  2 seconds - unit/multi
  4 seconds - unit/protocol
  10 seconds - unit/expire
  15 seconds - integration/replication
  4 seconds - integration/aof
  20 seconds - unit/type/list
  3 seconds - integration/rdb
  3 seconds - integration/convert-zipmap-hash-on-load
  1 seconds - unit/pubsub
  24 seconds - unit/aofrw
  1 seconds - unit/slowlog
  1 seconds - unit/introspection
  2 seconds - unit/limits
  30 seconds - unit/dump
  7 seconds - unit/scripting
  39 seconds - unit/type/list-2
  42 seconds - unit/type/hash
  55 seconds - unit/other
  28 seconds - unit/bitops
  33 seconds - unit/maxmemory
  58 seconds - unit/type/set
  48 seconds - integration/replication-psync
  59 seconds - unit/type/zset
  59 seconds - integration/replication-2
  63 seconds - unit/sort
  68 seconds - unit/basic
  73 seconds - unit/type/list-3
  78 seconds - integration/replication-4
  83 seconds - integration/replication-3
  64 seconds - unit/obuf-limits

\o/ All tests passed without errors!

Cleanup: may take some time... OK

It seems to work here.

@moreaki
moreaki commented Aug 19, 2013

My above test is not valid, since I didn't use a freshly compiled 2.6.14, but the unstable 2.7.104 one. I forgot to checkout the 2.6 branch.

$ git checkout -b origin/2.6 remotes/origin/2.6
$ make 1>/dev/null 2>&1
$ ./src/redis-server --version
Redis server v=2.6.14 sha=3bacf4c7:0 malloc=libc bits=64
$ sudo make test

[32/32 done]: unit/obuf-limits (52 seconds)

                   The End

Execution time of different units:
  0 seconds - unit/printver
  0 seconds - unit/quit
  1 seconds - unit/auth
  2 seconds - unit/multi
  3 seconds - unit/protocol
  3 seconds - integration/aof
  2 seconds - integration/rdb
  8 seconds - integration/replication-4
  11 seconds - unit/expire
  3 seconds - integration/convert-zipmap-hash-on-load
  1 seconds - unit/pubsub
  1 seconds - unit/slowlog
  1 seconds - unit/introspection
  14 seconds - integration/replication
  2 seconds - unit/limits
  7 seconds - unit/scripting
  19 seconds - unit/type/list
  23 seconds - unit/aofrw
  24 seconds - unit/type/set
  10 seconds - unit/dump
  33 seconds - unit/type/list-2
  36 seconds - unit/type/hash
  35 seconds - integration/replication-3
  20 seconds - unit/bitops
  27 seconds - unit/maxmemory
  40 seconds - unit/other
  45 seconds - unit/type/zset
  46 seconds - integration/replication-2
  47 seconds - unit/sort
  49 seconds - unit/basic
  52 seconds - unit/type/list-3
  52 seconds - unit/obuf-limits

\o/ All tests passed without errors!

Cleanup: may take some time... OK

But it still works perfectly well. So, could you provide the steps to reproduce your failing tests on your MacOSX 10.8.4 system, please?

@sandomingo

I've met the same problem when I run with "make test". However, after plussing "sudo", everything is ok now.

@seanthegeek

Why do tests need root?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment