[Gluster-devel] is tests/basic/gfapi/libgfapi-fini-hang.t broken in NetBSD ?

[Niels de Vos]

On Fri, Jul 07, 2017 at 09:37:35AM +0530, Atin Mukherjee wrote:
> https://build.gluster.org/job/netbsd7-regression/4761/consoleFull

mem_pool_fini() has recently been added. It suggests that there is a
problem with the implementation, maybe not NetBSD specific.

From the test console log:

[01:24:19] Running tests in file ./tests/basic/gfapi/libgfapi-fini-hang.t
Starting libgfapi_fini
glfs_set_volfile_server : returned 0
glfs_set_logging : returned 0
assertion "init_count > 0" failed: file "/home/jenkins/root/workspace/netbsd7-regression/libglusterfs/src/mem-pool.c", line 563, function "mem_pools_fini"
./tests/basic/gfapi/../../include.rc: line 344: 29714 Abort trap              (core dumped) ./libgfapi-fini-hang $H0 $V0 $logdir/libgfapi-fini-hang.log


I didn't see patches that call mem_pool_fini() yet, so it's a bit
surprising to see it fail. Hmm, gfapi calls mem_pool_fini() since
earlier this week (I still to catch up on gfapi patch reviews).


Related patches:
- https://review.gluster.org/17653 api: mem-pool sweeper thread needs to be created
- https://review.gluster.org/17662 libglusterfs: add mem_pools_fini
- https://review.gluster.org/17666 gfapi: add mem_pools_init and mem_pools_fini calls


The segfault is caused by GF_ASSERT() on
https://review.gluster.org/#/c/17662/2/libglusterfs/src/mem-pool.c@563 .
At the moment I'm not sure how this can happen, unless glfs_fini() is
called more than once on a glfs_t object.

Niels
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: not available
URL: <http://lists.gluster.org/pipermail/gluster-devel/attachments/20170707/20baf206/attachment.sig>