Age | Commit message (Collapse) | Author | Files | Lines |
|
At cleanup phase, the sample Inspector fails to re-enable the state of
the compute services. Such happens because the novaclient instance it's
using is bound to the 'doctor' project, which has just been deleted at
user.delete(). Thus, it is necessary to postpone the deletion of the
project after stopping the inspector.
Change-Id: I8cd5885cdfd3671a6c28342bdee65eac64dafaa8
Signed-off-by: Carlos Goncalves <carlos.goncalves@neclab.eu>
|
|
Change-Id: Icd110bd532e86a86b82824e849337649c284d3e6
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
A plugin for collectd is added to use collectd on compute as a
monitor type. Monitor files are updated accordingly.
The inspector now listens on all interfaces instead of only
localhost to enable it to communicate with compute node.
JIRA: DOCTOR-86
JIRA: DOCTOR-101
Change-Id: Idc834d428152e4687020eff7d8db36a652b1bf86
Signed-off-by: Umar Farooq <umar.farooq@neclab.eu>
|
|
|
|
|
|
After we modify the congress.conf file, congress server can't be started.
Change-Id: Icc3bd044a5fb7622620c42d6d25802739de875d0
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-100
Change-Id: I684071d35aac99ad1f5b65ae74e0a98ac726af35
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
Change-Id: Id68c9b1206deb58ff94fc942fbef94bffbc1e68b
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-100
Change-Id: I1e2c53f7a716156e35386d42c0df32a228a78fab
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-113
Change-Id: I60f17953e9b1cdf31ea50f313b33f8ede0831bc2
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
|
|
JIRA: DOCTOR-98
Change-Id: I22fa893a1c1973fc63f994b630dd0c6a2c30e641
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-99
Change-Id: I1831d329ffe80435532678fcb23d54f310422ce8
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
|
|
|
|
1.PYTHON_ENABLE=True does not work in bash
The python jobs does not run even we enabled them. The reason is that currently
JJB parse the parameter 'PYTHON_ENABLE' as 'True' which does not work in bash.
Fix it to compare with [Tt]rue.
2.Install tox if it is not installed
3.support only python3.4 in tox.ini
4.install python3-dev
Change-Id: I1a8ebad2fd06fe50ca4f739adb8d6d76caa2963b
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-98
Change-Id: I7110b680d0a0c64b1b2666abcea183e73373dca7
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
Host status returned the wrong output as the "OS-EXT-SRV-ATTR:host"
pattern appears twice in the "openstack server show" output.
The argument connection_pool is deprecated and is not needed.
Change-Id: Ib107e0bad8084bdf6f77689f112bde04bdb97543
Signed-off-by: Umar Farooq <umar.farooq@neclab.eu>
|
|
JIRA: DOCTOR-93
Change-Id: I316608c9627ff38e55154257b798244de3f221c5
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-93
Change-Id: I7645c1cc0c76ff86b9fbcc01c19038469fc9b9b4
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
|
|
|
|
Non-admin user is not allowed to view host_status flag.
To get the host status, policy.json file needs to be
created as it is not available in Ocata by default.
This is created for local and fuel installer types
in this commit.
JIRA: DOCTOR-110
Change-Id: Ie626217a515f821c47b1cb6ed0e3d504d8e01b3a
Signed-off-by: Umar Farooq <umar.farooq@neclab.eu>
|
|
Change-Id: Ib8e696acfe21787904fce6838341f3a256aa0267
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
Change-Id: I68036c295689fae55125854074e3bc4d920fec06
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-93
Change-Id: I85bfb21b7c07c40dfd1bfa6049130ad82fa3996a
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
|
|
The run.sh extracts the name of interface using awk
with 7th value in "ip a" command but in case of DHCP
enabled interface this value is "dynamic". So the fix
takes the last value of the output which is the
correct interface name.
EDIT: Removed the redundant line to get 5th element.
JIRA: DOCTOR-109
Change-Id: I0775d90ef219387e5b68abd4fd98d38d2104f128
Signed-off-by: Umar Farooq <umar.farooq@neclab.eu>
|
|
`os` and `sys` is no longer used in sample monitor, inspector and consumer
Change-Id: I6d2a6ccf1cd68851a80e6e6cf0c50bf06c8d1cbb
Signed-off-by: Yujun Zhang <zhang.yujunz@zte.com.cn>
|
|
|
|
In monitor.py, the self.session value is not defined
in DoctorMonitorSample class. This commit fixes it.
JIRA: DOCTOR-111
Change-Id: Ib61caa4146d78d6fe4857299c4730f13a7ac7a27
Signed-off-by: Umar Farooq <umar.farooq@neclab.eu>
|
|
This option will be removed, once we confirmed tox/python testing code
works fine and functest trigger has changed from 'run.sh' to 'tox'.
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
Signed-off-by: Carlos Goncalves <carlos.goncalves@neclab.eu>
Change-Id: Ia4421f6e39a415bd7b6212af612249deb482e7af
|
|
The values of OS_USER_DOMAIN_ID and OS_PROJECT_DOMAIN_ID
are needed as auth parameters else the nova API gives
HTTP 400 error.
JIRA: DOCTOR-108
Change-Id: I58b746e5216c00d2f78b0ccf899b69fe66c15848
Signed-off-by: Umar Farooq <umar.farooq@neclab.eu>
|
|
Change-Id: Ifb650f52fed641a7146358eada571e5c4beaf3d6
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
JIRA: DOCTOR-93
Change-Id: I6551d5e31799b967ce442cb05a664c0d64a3dcf8
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
|
|
JIRA: DOCTOR-93
Change-Id: Ife3255c4fe2d170c5d1b8354aaab24d5b1e3bc64
Signed-off-by: dongwenjuan <dong.wenjuan@zte.com.cn>
|
|
Change role of doctor user to _member_
Do needful changes to API calls
Add admin a role to project
Add Fuel support
JIRA: DOCTOR-80
Change-Id: Ic7f88b9ceb6c4fc90e5e985ecfbc50c907b2367e
Signed-off-by: Tomi Juvonen <tomi.juvonen@nokia.com>
|
|
JIRA: DOCTOR-96
Change-Id: I5b0923ffeb53ff53ad9f780b8b5ebdce834d995a
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
- use relevant identity version estimated from OS_AUTH_URL
- (apex) unset OS_PROJECT_ID
Change-Id: I7e51c7d5510baaa4d14b16600f9efa6fcfc011b9
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
- create "doctor_net" and use it in VM creation in order to avoid
virtual network issues
- make sample inspector use keystone v3
- use "openstack compute service set" for unset force_down
- drop nova_force_down.py as keystone returns nova v2.1
- use '--os-project-name' except for ceilometer cli
- avoid ping check (wait_ping) with empty "COMPUTE_IP"
- add missing '$' in quota update args
- move ping check (wait_ping) into unset_forced_down_hosts() and
perform ping check only to downed compute hosts
- use wait_until() instead of wait_ping()
- put out unset_forced_down_hosts from collect_logs()
JIRA: DOCTOR-95
Co-Authored-By: Carlos Goncalves <carlos.goncalves@neclab.eu>
Change-Id: I3275ff5dd993b82029dac6a58087096baa251022
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
Change-Id: Id8f307624564a6b822595689dbde64fee351b17f
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
Add support to run Doctor test with more than one VM by setting
VM_COUNT to wanted amount of VMs. Normally half of those VMs
gets scheduled to host that is used for Doctor testing.
JIRA: DOCTOR-77
Change-Id: I9fac96c6f53c290fccc35a2fd933b9fb106b96f2
Signed-off-by: Tomi Juvonen <tomi.juvonen@nokia.com>
|
|
|
|
- make sure expected parameters and functions are set
- remove all installer dependent codes from run.sh
except non-used function which will be removed later on
Change-Id: I4628173ef0891d893146976c7553ed6a95329957
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
|
|
Change-Id: I5c486c5f08880f6b3b6b4b393360752fb6077bed
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
JIRA: DOCTOR-58
Change-Id: Ic9326791007a3afa754cc64423b46adc17c59dc0
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
- collect disable_network.log
- use Python oneliner instead of `bc`
- rename T00 from hostdown to linkdown
- add step collect_logs
- move profiler to a dedicated step
- reorder function definitions to elimate warning on unresolved variable
- redirect also stderr to log
JIRA: DOCTOR-94
Change-Id: Ic63e5163945c5b21fd1d1c0bf99be353c11c2ada
Signed-off-by: Yujun Zhang <zhang.yujunz@zte.com.cn>
|
|
- make sure all hosts are unset force_down flag
- load all functions for inspector and installer first in order
to avoid 'command not found' error
- also, print VM if it has error state
Change-Id: If56cba201e1a5f3669a97ffb3d28b3c55c7cdb58
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|