Record test steps into logging.info to make steps looks more clearly;
And fix some coding style issue;
Signed-off-by: NXu Tian <xutian@redhat.com>
Acked-by: NSuqin Huang <shuang@redhat.com>

As this guest performance will get worse when the disk is filled up,
and may influence the normal session. Use serial session instead.
Signed-off-by: NYiqiao Pu <ypu@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Add the cfg for check virtio serial sys file info.
Signed-off-by: NYiqiao Pu <ypu@redhat.com>

Signed-off-by: NYiqiao Pu <ypu@redhat.com>

The old version makes a mistake and makes option rom
never be added.
Signed-off-by: NYiqiao Pu <ypu@redhat.com>

The cleanup action causes a error if guest still alive when removing
images on LVM, thus this patch destroy guest before cleanup method
called.

Other update:
 * Introduct error.context into this script.
 * Add try-except block when calling cleanup method in case unexpected
   exception is raised.
Signed-off-by: NQingtang Zhou <qzhou@redhat.com>
Acked-by: NFeng Yang <fyang@redhat.com>

Update the module patch and module name.
Signed-off-by: NYiqiao Pu <ypu@redhat.com>

This patch add error.context for vlan.py script.
All update the steps in doc string.
Move sub function onto top of function code.

We may still need a patch to sync script with test case.
Signed-off-by: NFeng Yang <fyang@redhat.com>
Acked-by: NSuqin Huang <shuang@redhat.com>

Update the logging message for nmi_bosd_catch, so that we may analyze
the root reason for the case failed.
Signed-off-by: NQingtang Zhou <qzhou@redhat.com>
Acked-by: NFeng Yang <fyang@redhat.com>

Print shutdown steps
Signed-off-by: NSuqin Huang <shuang@redhat.com>
Acked-by: NFeng Yang <fyang@redhat.com>

update test steps into doc string and add test steps into logging.info
Signed-off-by: NXu Tian <xutian@redhat.com>
Acked-by: NShuping Cui <scui@redhat.com>
Acked-by: NSuqin Huang <shuang@redhat.com>

Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

Record test steps into logging.info to make test steps more clearly
Signed-off-by: NXu Tian <xutian@redhat.com>
Acked-by: NSuqin Huang <shuang@redhat.com>

Signed-off-by: NYiqiao Pu <ypu@redhat.com>

Log test steps to logging.info.
Rename '_' to i.
Signed-off-by: NFeng Yang <fyang@redhat.com>
Acked-by: NShuping Cui <scui@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

This patch makes the read/write cmd fully configurable.
Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

And add quote mark for config parameters.
Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

This test script just leaves the disk(s) in mounted
status after read operation, this patch adds 2 more
steps after read: umount the disk and check dmesg
output.

CC: Xu Tian <xutian@redhat.com>
Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

move params.get() out of wait_job_done function;
And drop useless params;
Signed-off-by: NXu Tian <xutian@redhat.com>

Signed-off-by: NXu Tian <xutian@redhat.com>

There is no need to set internal_timeout for "cat /proc/cpuinfo".
It will waste 10s for each case which calls it.
Signed-off-by: NYu Mingfei <yumingfei@cn.fujitsu.com>

Signed-off-by: NYu Mingfei <yumingfei@cn.fujitsu.com>

Signed-off-by: NJason Wang <jasowang@redhat.com>
Signed-off-by: NAmos Kong <akong@redhat.com>

As asset.download_file will only handle urls(unknown url type will be reported)
which makes the nfs mounted to local can not be used for image copy. So use
utils.get_file when we use a nfs mounted in local.

Also fix the error.context_aware missing, update the error image store
path and add the error contexts.
Signed-off-by: NYiqiao Pu <ypu@redhat.com>

Improve physical_resources_check in serial name part.
  - Add a random serial name to guest less than 20 characters
  - Add strict check for serial name

changes from v1:
  - format modify
changes from v2:
  - move randome generate string to virt_utils
changes from v3:
  - move this test only for virtio_blk and only for RHEL 6 guest
    as test plan
changes from v4:
  - clean up and make this test will not run in Windows guests
Signed-off-by: NYiqiao Pu <ypu@redhat.com>
Acked-by: NFeng Yang <fyang@redhat.com>
Acked-by: NQingtang Zhou <qzhou@redhat.com>

Add machine_type check function to check the guest's machine type

change from v1:
1. change 'extra_params' to 'machine type'
2. add try except block when compare the result
Signed-off-by: NGolita Yue <gyue@redhat.com>

Changes from old version:
 - Splits lexical analyze from syntax analyze.
    (not fully because speed optimization)
 - Makes parser more strict for easier work with configs.
 - Error message should be more concrete.
 - Adds new syntax features.
 - Adds optimization for generation of lots of variants.
 - Improve generating of huge number of variants.
       (almost 2 x faster than old version)  50s old -> 23s new

Disadvantage:
  Syntax analyze is more strict and takes 1/3 of time more
  for parse text. 220ms -> 320ms

New syntax features:
variants:
   - added metadata variants _name_ [default=aaa] [xxx] [yyy]:
     where:
        _name_           - Name of variants group.
        [default=aaa]    - Alternative for @. It raise error
                           if variant aaa missing in variants.
                         - It is possible add more [default=xxx]
                           but only first of them is expanded if
                           defaults is set.
        [xxx], [yyy]     - Another possible metadata for future.
   - default could be enabled and disabled. Generate only default variant.
        enable:
          cartesian_config.py -d
          Parser(..., ..., defaults=True, ...)
        disable:
          cartesian_config.py
          Parser(..., ..., defaults=False, ...)
   - default variant could be expanded by parser parameters.
        cartesian_config.py --expand _name1_,_name2_
        Parser(..., ..., ..., expand_defualts="_name1_, _name2_")

filters:
  - new examples
   only (guest_os=Fedora).15
   only (guest_os=Fedora)..(fedora_version=15)
   only (15)

   variants group "guest_os" and variant name "Fedora"

  - variants comparison is based on hash value
    instead of pure string comparison.

Changes:
 * Fix pickling problem
 * Fix use of py 2.6 only constructs
 * Fix compatibility problem with old constructs
Signed-off-by: NJiří Župka <jzupka@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Update ethtool params for different guest and driver
Signed-off-by: NAmos Kong <akong@redhat.com>
Signed-off-by: NXiaoqing Wei <xwei@redhat.com>

log case steps into logging.info.
Signed-off-by: NFeng Yang <fyang@redhat.com>
Acked-by: NSuqin Huang <shuang@redhat.com>

There could be problem when this test fails, so add
try...except block in the end of this script to avoid
unexpected exception when test fails.

Changes from V1:
 * Rebase this patch against merged tree
 * Verify session before close it.
Signed-off-by: NQingtang Zhou <qzhou@redhat.com>
Acked-by: NFeng Yang <fyang@redhat.com>

Some nic driver would print logs into serial console, so send
a return character then wait for the prompt.

This patch also re-formats the ethtool_set function.

Changes from v3:
 * Add a timeout for waiting the shell prompt.
Signed-off-by: NQingtang Zhou <qzhou@redhat.com>

use utils_misc.get_winutils_vol() to instead static volume ID, becase
winutils cdrom has different volume ID when use different driver_format
on different windows guest;
Signed-off-by: NXu Tian <xutian@redhat.com>

use new interface utils_misc.get_winutils_vol() to get winutils cdrom
driver letter; Replace format disk related code with run format_disk sub
tests to avoild duplicate code;
Signed-off-by: NXu Tian <xutian@redhat.com>

iometer.icf for iometer_windows test, iometer-performance.icf for
iometer-performance test;
Signed-off-by: NXu Tian <xutian@redhat.com>