Re: [Qemu-block] [PATCH] iotests: add 248: test resume mirror after auto pause on ENOSPC

[Kevin Wolf]

Am 20.03.2019 um 10:59 hat Vladimir Sementsov-Ogievskiy geschrieben:
> Test that mirror job actually resume on resume command after being
> automatically paused on ENOSPC error.
> 
> It's a follow-up test for 8d9648cbf3e
>     "blockjob: fix user pause in block_job_error_action"
> 
> Signed-off-by: Vladimir Sementsov-Ogievskiy <address@hidden>
> ---
>  tests/qemu-iotests/248     | 77 ++++++++++++++++++++++++++++++++++++++
>  tests/qemu-iotests/248.out | 12 ++++++
>  tests/qemu-iotests/group   |  1 +
>  3 files changed, 90 insertions(+)
>  create mode 100755 tests/qemu-iotests/248
>  create mode 100644 tests/qemu-iotests/248.out
> 
> diff --git a/tests/qemu-iotests/248 b/tests/qemu-iotests/248
> new file mode 100755
> index 0000000000..ab88c96532
> --- /dev/null
> +++ b/tests/qemu-iotests/248
> @@ -0,0 +1,77 @@
> +#!/usr/bin/env python
> +#
> +# Test resume mirror after auto pause on ENOSPC
> +#
> +# Copyright (c) 2019 Virtuozzo International GmbH. All rights reserved.
> +#
> +# This program is free software; you can redistribute it and/or modify
> +# it under the terms of the GNU General Public License as published by
> +# the Free Software Foundation; either version 2 of the License, or
> +# (at your option) any later version.
> +#
> +# This program is distributed in the hope that it will be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program.  If not, see <http://www.gnu.org/licenses/>.
> +#
> +
> +import iotests
> +from iotests import qemu_img_create, qemu_io, file_path, filter_qmp_testfiles
> +
> +iotests.verify_image_format(supported_fmts=['qcow2'])
> +
> +source, target = file_path('source', 'target')
> +size = 5 * 1024 * 1024
> +limit = 2 * 1024 * 1024

limit is unused...

> +qemu_img_create('-f', iotests.imgfmt, source, str(size))
> +qemu_img_create('-f', iotests.imgfmt, target, str(size))
> +qemu_io('-c', 'write 0 {}'.format(size), source)
> +
> +# raw format don't like empty files
> +qemu_io('-c', 'write 0 {}'.format(size), target)
> +
> +vm = iotests.VM().add_drive(source)
> +vm.launch()
> +
> +vm.qmp_log('blockdev-add', filters=[filter_qmp_testfiles], **{
> +    'driver': iotests.imgfmt,
> +    'node-name': 'target',
> +    'file': {
> +        'driver': 'raw',
> +        'size': 2 * 1024 * 1024,

...because you used the literal value again here.

> +        'file': {
> +            'driver': 'file',
> +            'filename': target
> +        }
> +    }
> +})
> +
> +vm.qmp_log('blockdev-mirror', device='drive0', sync='full', target='target',
> +           on_target_error='enospc')
> +
> +vm.event_wait('JOB_STATUS_CHANGE', timeout=3.0,
> +              match={'data': {'status': 'paused'}})
> +
> +vm.qmp_log('query-block-jobs')
> +
> +vm.qmp_log('x-blockdev-reopen', filters=[filter_qmp_testfiles], **{
> +    'driver': iotests.imgfmt,
> +    'node-name': 'target',
> +    'file': {
> +        'driver': 'raw',
> +        # size parameter dropped
> +        'file': {
> +            'driver': 'file',
> +            'filename': target
> +        }
> +    }
> +})

Instead of duplicating the configuration here, maybe it would be nicer
to store it in a variable and just del options['file']['size'] before
passing them to x-blockdev-reopen.

By the way, I like that you used x-blockdev-reopen. It's much better
than my suggestion to use HMP qemu-io.

> +
> +vm.qmp_log('block-job-resume', device='drive0')
> +vm.qmp_log('query-block-jobs')  # check that job is resumed

This is the expected output for this query-block-jobs:

> +{"return": [{"auto-dismiss": true, "auto-finalize": true, "busy": true, 
> "device": "drive0", "io-status": "ok", "len": 5242880, "offset": 1048576, 
> "paused": false, "ready": false, "speed": 0, "status": "running", "type": 
> "mirror"}]}

So it expects a specific offset of 1048576. Is this reliable without
throttling the job?

Kevin