glusterfs.git/tests, branch dht-stale-layout-fixes

Wrong test in self-heal.t

2014-08-08T05:01:56+00:00

self-heal.t checks for gfid before and after an operation, but fails
to do it on the same file since the prefix is not the same: undefined
$B1 before and $B0 after.

The errorcause the test to fail on NetBSD. I do not know how this test
could pass on Linux before.

BUG: 764655
Change-Id: I6df1175e177855afca1736cf8c847c51fe407576
Signed-off-by: Emmanuel Dreyfus 
Reviewed-on: http://review.gluster.org/8431
Reviewed-by: Ravishankar N 
Reviewed-by: Pranith Kumar Karampuri 
Tested-by: Pranith Kumar Karampuri

build: make GLUSTERD_WORKDIR rely on localstatedir

2014-08-07T08:17:29+00:00

- Break-way from '/var/lib/glusterd' hard-coded previously,
  instead rely on 'configure' value from 'localstatedir'
- Provide 's/lib/db' as default working directory for gluster
  management daemon for BSD and Darwin based installations
- loff_t is really off_t on Darwin
- fix-off the warnings generated by clang on FreeBSD/Darwin
- Now 'tests/*' use GLUSTERD_WORKDIR a common variable for all
  platforms.
- Define proper environment for running tests, define correct PATH
  and LD_LIBRARY_PATH when running tests, so that the desired version
  of glusterfs is used, regardless where it is installed.
  (Thanks to manu@netbsd.org for this additional work)

Change-Id: I2339a0d9275de5939ccad3e52b535598064a35e7
BUG: 1111774
Signed-off-by: Harshavardhana 
Signed-off-by: Emmanuel Dreyfus 
Reviewed-on: http://review.gluster.org/8246
Tested-by: Gluster Build System

Regression test portability: truncate, md5

2014-08-06T11:49:14+00:00

Add shell functions to replace truncate and md5, which are Linux specific
Resubmit because of failed regression test with no apparent cause

BUG: 764655
Change-Id: I07200cf886bd52904a5cf63c66f43f0b1cc91540
Signed-off-by: Emmanuel Dreyfus 
Reviewed-on: http://review.gluster.org/8341
Tested-by: Gluster Build System 
Reviewed-by: Harshavardhana 
Tested-by: Harshavardhana 
Reviewed-by: Vijay Bellur

tests: weighted-rebalance.t shouldn't write to '/dev/tty'

2014-08-06T11:39:20+00:00

On our jenkins instance "/dev/tty" doesn't exist, necessary
output fails as below message

~~~
./tests/features/weighted-rebalance.t: \
         line 72: /dev/tty: No such device or address
~~~

Comment out the debugging code

Change-Id: Iba29b80c8ba2dcaab3d6654d7c54332a915bffb8
BUG: 1114680
Signed-off-by: Harshavardhana 
Reviewed-on: http://review.gluster.org/8421
Tested-by: Gluster Build System 
Reviewed-by: Jeff Darcy 
Reviewed-by: Vijay Bellur

Regression test portability: arequal-checksum

2014-08-05T00:41:19+00:00

Building arequal-checksum on non Linux systems requires a few adjustments:
- use __builtin_alloca() on all platforms
- on systems without argp, get it from contrib/standalone-argp
  (this requires adding cflags support to the build_tester function)
- FTW_ACTIONRETVAL option to nftw(3) is GNU libc specific, work around
  if it does not exist.
- md5sum is Linux-specific. Use md5 -n on NetBSD and md5 -q on FreeBSD/Darwin
- Use 'cc' as synonymous for all compilers, it can behave as gcc/clang depending
  on which is default
- cleanup tabs/whitespaces

BUG: 764655
Change-Id: I9090c17da596fbf00fc1fbd7593163ce8cd3b84c
Signed-off-by: Emmanuel Dreyfus 
Signed-off-by: Harshavardhana 
Reviewed-on: http://review.gluster.org/8283
Tested-by: Gluster Build System

cluster/ec: Fix incorrect management of NFS requests

2014-08-02T07:23:24+00:00

Some operations, specially those comming from NFS, do not use a
regular fd and use an anonymous fd (i.e. a previous open call has
not been sent). Any context information created during open or
create will not be present on these fd's, so we simply return NULL
for contexts of those fd.

Also it seems that NFS can send write requests with a very big
buffer (higher that the default value of 128 KB). Some changes
have been made to correctly handle these large buffers.

Change-Id: I281476bd0d2cbaad231822248d6a616fcf5d4003
BUG: 1122417
Signed-off-by: Xavier Hernandez 
Reviewed-on: http://review.gluster.org/8367
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

ec: Test volume mount point in a better way

2014-08-02T06:48:47+00:00

An 'ls -a1' on an empty volume seems to return 3 entries instead
of the expected 2 ('.' and '..') in the build servers. I changed
the test to a simple 'stat', which is enough and more reliable.

Change-Id: I12d0f47394ad378b40fc9b86507cdb3543f99970
Signed-off-by: Xavier Hernandez 
Reviewed-on: http://review.gluster.org/8313
Reviewed-by: Jeff Darcy 
Tested-by: Gluster Build System 
Reviewed-by: Humble Devassy Chirammal 
Reviewed-by: Raghavendra Bhat 
Reviewed-by: Vijay Bellur

tests: Avoid test bug-884597.t to loop infinitely

2014-08-02T06:32:26+00:00

If some unexpected error happens while this test is trying to
create a linkto file, an infinite loop is generated and the test
never ends.

This patch limits the number of retries it can do to find that
file.

Change-Id: Ie556c509116109ead99c78d671b2669b4428faac
Signed-off-by: Xavier Hernandez 
Reviewed-on: http://review.gluster.org/8318
Tested-by: Gluster Build System 
Reviewed-by: Vijay Bellur

Regression test portability: system-specific cleanup

2014-08-02T05:26:01+00:00

Add NetBSD system-specific cleanup code
Define killall as pkill when it is missing.

BUG: 764655
Change-Id: I3a1cfa3029fd60ede334a670fc115c211700063d
Signed-off-by: Emmanuel Dreyfus 
Reviewed-on: http://review.gluster.org/8290
Reviewed-by: Harshavardhana 
Tested-by: Harshavardhana 
Tested-by: Gluster Build System

Symlink mtime changes when rebalancing

2014-07-31T16:07:59+00:00

Added mtime preservation for special files during rebalance.

Change-Id: If04921d4d66853fde8b4d8a3ab748790864f8f42
BUG: 1122443
Signed-off-by: Anders Blomdell 
Reviewed-on: http://review.gluster.org/8383
Tested-by: Gluster Build System 
Reviewed-by: Shyamsundar Ranganathan 
Reviewed-by: Harshavardhana 
Reviewed-by: Anand Avati