1:09 p.m.
When the connection in virsh got disconnected due to keepalive, virsh
was trying to reconnect. Adding the REASON_KEEPALIVE to the list of
disconnect reasons after virsh should not reconnect (patch 1/1) was
not enough because our rpc code was rewriting those and that needs to
be fixed too (patch 2/2).
Martin Kletzander (2):
rpc: Report proper close reason
virsh: Really disconnect on keepalive timeout
src/rpc/virnetclient.c | 7 +++++--
tools/virsh.c | 13 +++++++++----
2 files changed, 14 insertions(+), 6 deletions(-)
--
2.1.3
1:09 p.m.
New subject: [libvirt] [PATCH 1/2] rpc: Report proper close reason
Whenever client socket was marked as closed for some reason, it could've
been changed when really closing the connection. With this patch the
proper reason is kept since the first time it's marked as closed.
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
src/rpc/virnetclient.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c
index 8657b0e..d7455b5 100644
--- a/src/rpc/virnetclient.c
+++ b/src/rpc/virnetclient.c
@@ -634,8 +634,11 @@ virNetClientMarkClose(virNetClientPtr client,
VIR_DEBUG("client=%p, reason=%d", client, reason);
if (client->sock)
virNetSocketRemoveIOCallback(client->sock);
- client->wantClose = true;
- client->closeReason = reason;
+ /* Don't override reason that's already set. */
+ if (!client->wantClose) {
+ client->wantClose = true;
+ client->closeReason = reason;
+ }
}
--
2.1.3
3:46 a.m.
New subject: [libvirt] [PATCH 1/2] rpc: Report proper close reason
On Sun, Nov 30, 2014 at 20:09:08 +0100, Martin Kletzander wrote:
Whenever client socket was marked as closed for some reason, it
could've
been changed when really closing the connection. With this patch the
proper reason is kept since the first time it's marked as closed.
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
src/rpc/virnetclient.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c
index 8657b0e..d7455b5 100644
--- a/src/rpc/virnetclient.c
+++ b/src/rpc/virnetclient.c
@@ -634,8 +634,11 @@ virNetClientMarkClose(virNetClientPtr client,
VIR_DEBUG("client=%p, reason=%d", client, reason);
if (client->sock)
virNetSocketRemoveIOCallback(client->sock);
- client->wantClose = true;
- client->closeReason = reason;
+ /* Don't override reason that's already set. */
+ if (!client->wantClose) {
+ client->wantClose = true;
+ client->closeReason = reason;
+ }
}
ACK
Jirka
6:58 a.m.
New subject: [libvirt] [PATCH 1/2] rpc: Report proper close reason
On Mon, Dec 01, 2014 at 10:46:38AM +0100, Jiri Denemark wrote:
On Sun, Nov 30, 2014 at 20:09:08 +0100, Martin Kletzander wrote:
> Whenever client socket was marked as closed for some reason, it could've
> been changed when really closing the connection. With this patch the
> proper reason is kept since the first time it's marked as closed.
>
> Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
> ---
> src/rpc/virnetclient.c | 7 +++++--
> 1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/src/rpc/virnetclient.c b/src/rpc/virnetclient.c
> index 8657b0e..d7455b5 100644
> --- a/src/rpc/virnetclient.c
> +++ b/src/rpc/virnetclient.c
> @@ -634,8 +634,11 @@ virNetClientMarkClose(virNetClientPtr client,
> VIR_DEBUG("client=%p, reason=%d", client, reason);
> if (client->sock)
> virNetSocketRemoveIOCallback(client->sock);
> - client->wantClose = true;
> - client->closeReason = reason;
> + /* Don't override reason that's already set. */
> + if (!client->wantClose) {
> + client->wantClose = true;
> + client->closeReason = reason;
> + }
> }
>
ACK
Pushed, thanks.
Martin
1:09 p.m.
New subject: [libvirt] [PATCH 2/2] virsh: Really disconnect on keepalive timeout
Commit 676cb4f4e762b8682a06c6dab1f690fbcd939550 added client keepalive
support for virsh. But because whenever the connection is closed, virsh
tries to reconnect, unless the disconnection happened with the reason
VIR_CONNECT_CLOSE_REASON_CLIENT. Because of that, virsh kept trying to
reconnect even when the connection was closed due to keepalive.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1073506
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
tools/virsh.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/tools/virsh.c b/tools/virsh.c
index bcfa561..98d8d18 100644
--- a/tools/virsh.c
+++ b/tools/virsh.c
@@ -327,9 +327,14 @@ static int disconnected; /* we may have been disconnected */
static void
vshCatchDisconnect(virConnectPtr conn ATTRIBUTE_UNUSED,
int reason,
- void *opaque ATTRIBUTE_UNUSED)
+ void *opaque)
{
- if (reason != VIR_CONNECT_CLOSE_REASON_CLIENT)
+ vshControl *ctl = opaque;
+
+ vshDebug(ctl, VSH_ERR_INFO, "Received disconnect with reason '%d'",
reason);
+
+ if (reason != VIR_CONNECT_CLOSE_REASON_CLIENT &&
+ reason != VIR_CONNECT_CLOSE_REASON_KEEPALIVE)
disconnected++;
}
@@ -407,7 +412,7 @@ vshReconnect(vshControl *ctl)
vshError(ctl, "%s", _("failed to connect to the
hypervisor"));
} else {
if (virConnectRegisterCloseCallback(ctl->conn, vshCatchDisconnect,
- NULL, NULL) < 0)
+ ctl, NULL) < 0)
vshError(ctl, "%s", _("Unable to register disconnect
callback"));
if (connected)
vshError(ctl, "%s", _("Reconnected to the hypervisor"));
@@ -484,7 +489,7 @@ cmdConnect(vshControl *ctl, const vshCmd *cmd)
}
if (virConnectRegisterCloseCallback(ctl->conn, vshCatchDisconnect,
- NULL, NULL) < 0)
+ ctl, NULL) < 0)
vshError(ctl, "%s", _("Unable to register disconnect
callback"));
return true;
--
2.1.3
4:05 a.m.
New subject: [libvirt] [PATCH 2/2] virsh: Really disconnect on keepalive timeout
On Sun, Nov 30, 2014 at 20:09:09 +0100, Martin Kletzander wrote:
Commit 676cb4f4e762b8682a06c6dab1f690fbcd939550 added client
keepalive
support for virsh. But because whenever the connection is closed, virsh
tries to reconnect, unless the disconnection happened with the reason
VIR_CONNECT_CLOSE_REASON_CLIENT. Because of that, virsh kept trying to
reconnect even when the connection was closed due to keepalive.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1073506
Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
---
tools/virsh.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
It's nice this patch makes non-interactive virsh stop, however, it
prevents interactive virsh from automatically reconnecting when a next
command is issued, which is a very useful feature.
Jirka
4:10 a.m.
New subject: [libvirt] [PATCH 2/2] virsh: Really disconnect on keepalive timeout
On Mon, Dec 01, 2014 at 11:05:04AM +0100, Jiri Denemark wrote:
On Sun, Nov 30, 2014 at 20:09:09 +0100, Martin Kletzander wrote:
> Commit 676cb4f4e762b8682a06c6dab1f690fbcd939550 added client keepalive
> support for virsh. But because whenever the connection is closed, virsh
> tries to reconnect, unless the disconnection happened with the reason
> VIR_CONNECT_CLOSE_REASON_CLIENT. Because of that, virsh kept trying to
> reconnect even when the connection was closed due to keepalive.
>
> Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1073506
>
> Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com>
> ---
> tools/virsh.c | 13 +++++++++----
> 1 file changed, 9 insertions(+), 4 deletions(-)
It's nice this patch makes non-interactive virsh stop, however, it
prevents interactive virsh from automatically reconnecting when a next
command is issued, which is a very useful feature.
Thanks for finding that out. v2 coming up.
Martin
3671
days inactive
3679
days old
6 comments
2 participants
participants (2)
-
Jiri Denemark -
Martin Kletzander