branch master updated (c0d8538 -> 8590c5d)

[git repository hosting]

This is an automated email from the git hooks/post-receive script.

rharwood pushed a change to branch master
in repository gssproxy.

from  c0d8538   Fix silent crash with duplicate config sections
       new  af666af   Emit debug on queue errors
       new  8590c5d   Do not call gpm_grab_sock() twice

The 2 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails.  The revisions
listed as "adds" were already present in the repository and have only
been added to this reference.

Summary of changes:
 proxy/src/client/gpm_common.c | 9 +++------
 proxy/src/gp_workers.c        | 2 ++
 2 files changed, 5 insertions(+), 6 deletions(-)

01/02: Emit debug on queue errors

[git repository hosting]

This is an automated email from the git hooks/post-receive script.

rharwood pushed a commit to branch master
in repository gssproxy.

commit af666affbd4735ba437e3d89d9e22984a556ed16
Author: Simo Sorce 
Date:   Thu Oct 26 11:47:54 2017 -0400

Emit debug on queue errors

Signed-off-by: Simo Sorce 
    Reviewed-by: Robbie Harwood 
    Merges: #218
---
 proxy/src/gp_workers.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/proxy/src/gp_workers.c b/proxy/src/gp_workers.c
index 2a33c21..18f38f6 100644
--- a/proxy/src/gp_workers.c
+++ b/proxy/src/gp_workers.c
@@ -314,6 +314,8 @@ static void gp_handle_reply(verto_ctx *vctx, verto_ev *ev)
         case GP_QUERY_IN:
             /* ?! fallback and kill client conn */
         case GP_QUERY_ERR:
+            GPDEBUGN(3, "[status] Handling query error, terminating CID %d.\n",
+                     gp_conn_get_cid(q->conn));
             gp_conn_free(q->conn);
             gp_query_free(q, true);
             break;

02/02: Do not call gpm_grab_sock() twice

[git repository hosting]

This is an automated email from the git hooks/post-receive script.

rharwood pushed a commit to branch master
in repository gssproxy.

commit 8590c5dbc6fa07d0c366df23b982a4b6b9ffc259
Author: Simo Sorce 
Date:   Thu Oct 26 16:59:18 2017 -0400

Do not call gpm_grab_sock() twice

In the gpm_get_ctx() call, we unnecessarily call gpm_grab_sock() which
    would cause the lock to be held by one thread and never released.  We
    already call gpm_grab_sock() as the first thing after gpm_get_ctx() in
    gpm_make_call(), plus gpm_make_call() properly releases the socket
    once done.

This corrects the deadlock fix in
    461a5fa9f91a2753ebeef6323a64239c35e2f250, which incorrectly released
    the lock we wanted to grab.  This caused the socket to not be locked
    to our thread.  Another thread could come along and change the global
    ctx while we were still using the socket from another thread, causing
    concurrency issues as only one request can be in flight on any given
    socket at the same time.

In special cases where the "thread" uid/gid changes (like in
    rpc.gssd), we end up closing the socket while we are still waiting for
    an answer from the server, causing additional issues and confusion.

[[e-mail]: squashed 2 commits; minor edits accordingly]
    Signed-off-by: Simo Sorce 
    Reviewed-by: Robbie Harwood 
    Merges: #218
---
 proxy/src/client/gpm_common.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/proxy/src/client/gpm_common.c b/proxy/src/client/gpm_common.c
index 75c64d7..f9f9258 100644
--- a/proxy/src/client/gpm_common.c
+++ b/proxy/src/client/gpm_common.c
@@ -163,7 +163,9 @@ static int gpm_grab_sock(struct gpm_ctx *gpmctx)
         ret = gpm_open_socket(gpmctx);
     }

-    pthread_mutex_unlock(&gpmctx->lock);
+    if (ret) {
+        pthread_mutex_unlock(&gpmctx->lock);
+    }
     return ret;
 }

@@ -518,11 +520,6 @@ static struct gpm_ctx *gpm_get_ctx(void)

pthread_once(&gpm_init_once_control, gpm_init_once);

-    ret = gpm_grab_sock(&gpm_global_ctx);
-    if (ret) {
-        return NULL;
-    }
-
     return &gpm_global_ctx;
 }