block, migration: improve debugging of migration bdrv_flush failure

[PATCH 0/5] block, migration: improve debugging of migration bdrv_flush failure

Daniel P. Berrangé posted 5 patches 3 years ago

Download series mbox

Patches applied successfully (tree, apply log)
git fetch https://github.com/patchew-project/qemu tags/patchew/20210415135851.862406-1-berrange@redhat.com

Test checkpatch passed

block/file-posix.c     | 10 +++++-----
block/trace-events     |  1 +
migration/migration.c  |  1 +
migration/trace-events |  1 +
softmmu/cpus.c         |  7 ++++++-
softmmu/trace-events   |  3 +++
6 files changed, 17 insertions(+), 6 deletions(-)

Expand all Fold all

[PATCH 0/5] block, migration: improve debugging of migration bdrv_flush failure

Posted by Daniel P. Berrangé 3 years ago

I spent a while debugging a tricky migration failure today which was
ultimately caused by fdatasync() getting EACCESS. The existing probes
were not sufficient to diagnose this, so I had to resort to GDB. This
improves probes and block error reporting to make future diagnosis
possible without GDB.

Daniel P. Berrangé (5):
  migration: add trace point when vm_stop_force_state fails
  softmmu: add trace point when bdrv_flush_all fails
  block: preserve errno from fdatasync failures
  block: add trace point when fdatasync fails
  block: remove duplicate trace.h include

 block/file-posix.c     | 10 +++++-----
 block/trace-events     |  1 +
 migration/migration.c  |  1 +
 migration/trace-events |  1 +
 softmmu/cpus.c         |  7 ++++++-
 softmmu/trace-events   |  3 +++
 6 files changed, 17 insertions(+), 6 deletions(-)

-- 
2.30.2

Re: [PATCH 0/5] block, migration: improve debugging of migration bdrv_flush failure

Posted by Connor Kuehl 3 years ago

On 4/15/21 8:58 AM, Daniel P. Berrangé wrote:
> I spent a while debugging a tricky migration failure today which was
> ultimately caused by fdatasync() getting EACCESS. The existing probes
> were not sufficient to diagnose this, so I had to resort to GDB. This
> improves probes and block error reporting to make future diagnosis
> possible without GDB.
> 
> Daniel P. Berrangé (5):
>    migration: add trace point when vm_stop_force_state fails
>    softmmu: add trace point when bdrv_flush_all fails
>    block: preserve errno from fdatasync failures
>    block: add trace point when fdatasync fails
>    block: remove duplicate trace.h include
> 
>   block/file-posix.c     | 10 +++++-----
>   block/trace-events     |  1 +
>   migration/migration.c  |  1 +
>   migration/trace-events |  1 +
>   softmmu/cpus.c         |  7 ++++++-
>   softmmu/trace-events   |  3 +++
>   6 files changed, 17 insertions(+), 6 deletions(-)
> 

For the series:

Reviewed-by: Connor Kuehl <ckuehl@redhat.com>