[Bugs] [Bug 1457936] possible memory leak in glusterfsd with multiplexing

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1457936

Mohit Agrawal <moagrawa at redhat.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |moagrawa at redhat.com



--- Comment #6 from Mohit Agrawal <moagrawa at redhat.com> ---
Hi,

I have checked brick side code path regards to memory leak in case of stop the
volume or
remove the brick while brick mux is enable,I think below are the some code path
where code has leak

1) In notify of changelog.c need to cleanup rpc server/threads after call 
changelog_cleanup_rpc

2) In server xlator in server_setvolume we are creating inode table on xlator
as below
   client->bound_xl->itable =
                                inode_table_new (conf->inode_lru_limit,
                                                 client->bound_xl);
   But we are not cleaning this inode table from xlator at the time of stop the
volume or remove the brick, i think we need to clean this.

3) As of now in server xlator notify we are just updating the next pointer node
in graph to move out node from graph at the time of getting event
GF_EVENT_CLEANUP, i think this also needs to update in more cleaner way.

>>>>>>>>>>>>>
   if (victim_found)
           (*trav_p) = (*trav_p)->next;
   glusterfs_mgmt_pmap_signout (glusterfsd_ctx, victim->name);
   glusterfs_autoscale_threads (THIS->ctx, -1);

>>>>>>>>>>>>>>>>>>>
Regards
Mohit Agrawal

-- 
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=g6WRNtZaPN&a=cc_unsubscribe