Re: [Qemu-devel] [PATCH] Ensure migrate_cancel does not block doing I/O

2012-05-31 Thread Amos Kong
On Fri, Aug 26, 2011 at 9:48 PM, Paolo Bonzini wrote: > On 08/26/2011 01:25 PM, Daniel P. Berrange wrote: >> >> diff --git a/migration.c b/migration.c >> index f5959b4..6448d0b 100644 >> --- a/migration.c >> +++ b/migration.c >> @@ -319,6 +319,11 @@ ssize_t migrate_fd_put_buffer(void *opaque, cons

Re: [Qemu-devel] [PATCH] Ensure migrate_cancel does not block doing I/O

2011-08-26 Thread Paolo Bonzini
On 08/26/2011 01:25 PM, Daniel P. Berrange wrote: diff --git a/migration.c b/migration.c index f5959b4..6448d0b 100644 --- a/migration.c +++ b/migration.c @@ -319,6 +319,11 @@ ssize_t migrate_fd_put_buffer(void *opaque, const void *data, size_t size) FdMigrationState *s = opaque; ssi

Re: [Qemu-devel] [PATCH] Ensure migrate_cancel does not block doing I/O

2011-08-26 Thread Daniel P. Berrange
On Fri, Aug 26, 2011 at 11:59:28AM +0100, Daniel P. Berrange wrote: > From: "Daniel P. Berrange" > > There are two common cases where migrate_cancel is intended to be > used > > 1. When migration is not converging due to an overactive > guest and insufficient network bandwidth > 2. When

[Qemu-devel] [PATCH] Ensure migrate_cancel does not block doing I/O

2011-08-26 Thread Daniel P. Berrange
From: "Daniel P. Berrange" There are two common cases where migrate_cancel is intended to be used 1. When migration is not converging due to an overactive guest and insufficient network bandwidth 2. When migration is stuck due a network outage, waiting for the TCP transmit timeout