]> git.apps.os.sepia.ceph.com Git - teuthology.git/log
teuthology.git
6 years agoFix timeout rh-multisite 1249/head
Shilpa Jagannath [Mon, 28 Jan 2019 04:59:17 +0000 (10:29 +0530)]
Fix timeout

6 years agoDocstring updated
Shilpa Jagannath [Fri, 18 Jan 2019 08:21:18 +0000 (13:51 +0530)]
Docstring updated

6 years agoRemoved cluster naming
Shilpa Jagannath [Fri, 18 Jan 2019 08:12:57 +0000 (13:42 +0530)]
Removed cluster naming

6 years agoadd a missing comma :(
Vasu Kulkarni [Wed, 19 Dec 2018 16:58:54 +0000 (08:58 -0800)]
add a missing comma :(

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agodisable firewall run in ceph-ansible playbook
Vasu Kulkarni [Tue, 18 Dec 2018 19:02:20 +0000 (11:02 -0800)]
disable firewall run in ceph-ansible playbook

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoAdd client keys on client machines in case we are using ceph-ansible for cluster...
Shylesh Kumar [Thu, 13 Dec 2018 23:14:08 +0000 (15:14 -0800)]
Add client keys on client machines in case we are using ceph-ansible for cluster setup

Signed-off-by: Shylesh Kumar <shmohan@redhat.com>
6 years agodont use the stdout when running playbook, this may cause the prints to buffer.
Vasu Kulkarni [Mon, 17 Dec 2018 22:54:24 +0000 (14:54 -0800)]
dont use the stdout when running playbook, this may cause the prints to buffer.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoRemoved unused variable 'remote'
Shilpa Jagannath [Wed, 31 Oct 2018 10:53:28 +0000 (16:23 +0530)]
Removed unused variable 'remote'

6 years agoUpdated docstring to describe haproxy roles
Shilpa Jagannath [Tue, 30 Oct 2018 07:09:58 +0000 (12:39 +0530)]
Updated docstring to describe haproxy roles

6 years agoAdds haproxy support and other cleanup fixes
Shilpa Jagannath [Tue, 30 Oct 2018 07:02:04 +0000 (12:32 +0530)]
Adds haproxy support and other cleanup fixes

6 years agocheck if osd id is an integer, if so use it and discard any empty or non integer...
Vasu Kulkarni [Tue, 9 Oct 2018 20:27:27 +0000 (13:27 -0700)]
check if osd id is an integer, if so use it and discard any empty or non integer ids.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agofix couple more flake8 issues
Vasu Kulkarni [Mon, 8 Oct 2018 18:38:49 +0000 (11:38 -0700)]
fix couple more flake8 issues

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoAdd osd_disk_info to ctx so that any task can access it like ctx.osd_disk_info
shylesh kumar [Sat, 6 Oct 2018 01:47:21 +0000 (21:47 -0400)]
Add osd_disk_info to ctx so that any task can access it like ctx.osd_disk_info

Signed-off-by: shylesh kumar <shylesh.mohan@gmail.com>
6 years agoadd ansible 2.6 repo
Vasu Kulkarni [Thu, 4 Oct 2018 16:42:05 +0000 (09:42 -0700)]
add ansible 2.6 repo

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agofix rh_install unit test
Vasu Kulkarni [Wed, 26 Sep 2018 17:39:56 +0000 (10:39 -0700)]
fix rh_install unit test

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoexec failure fix
Shylesh Kumar [Thu, 14 Jun 2018 16:25:37 +0000 (21:55 +0530)]
exec failure fix

Signed-off-by: Shylesh Kumar <shylesh.mohan@gmail.com>
6 years agoFix log remote dir with dual clusters
Shilpa Jagannath [Fri, 31 Aug 2018 07:22:35 +0000 (12:52 +0530)]
Fix log remote dir with dual clusters

6 years agomissing nfss key added in generate hosts file section
rakesh [Fri, 22 Jun 2018 17:42:23 +0000 (23:12 +0530)]
missing nfss key  added in generate hosts file section

6 years agoFix compatibility with single cluster run
Shilpa Jagannath [Tue, 12 Jun 2018 08:31:48 +0000 (14:01 +0530)]
Fix compatibility with single cluster run

6 years agoAll fixes included for running dual cluster with ceph-ansible.
Shilpa Jagannath [Tue, 12 Jun 2018 05:17:28 +0000 (10:47 +0530)]
All fixes included for running dual cluster with ceph-ansible.

6 years agoFix unregister during final cleanup, due to stage cdn task that is not needed now.
Vasu Kulkarni [Mon, 11 Jun 2018 00:10:08 +0000 (17:10 -0700)]
Fix unregister during final cleanup, due to stage cdn task that is not needed now.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoneed to debug why check_status is being ignored, for now catch
Vasu Kulkarni [Sun, 10 Jun 2018 22:50:55 +0000 (15:50 -0700)]
need to debug why check_status is being ignored, for now catch
the command failed error, log and ignore so that tests can pass.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agounsubscribe at the start of the test as well, if nodes are not freshly reimaged
Vasu Kulkarni [Sun, 10 Jun 2018 21:41:42 +0000 (14:41 -0700)]
unsubscribe at the start of the test as well, if nodes are not freshly reimaged
then they may have subscription from old configuration

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agotypo fix and ignore status if unregister fails
Vasu Kulkarni [Sun, 10 Jun 2018 15:35:09 +0000 (08:35 -0700)]
typo fix and ignore status if unregister fails

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoEnable standard rhel repos as well after stage cdn configuration
Vasu Kulkarni [Sun, 10 Jun 2018 14:27:54 +0000 (07:27 -0700)]
Enable standard rhel repos as well after stage cdn configuration

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agosubscribe rhel octo nodes using stage cdn from configuration
Vasu Kulkarni [Fri, 8 Jun 2018 23:08:48 +0000 (16:08 -0700)]
subscribe rhel octo nodes using stage cdn from configuration

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoAdd misc cleanup options before we unlock the system
Vasu Kulkarni [Thu, 31 May 2018 17:44:40 +0000 (10:44 -0700)]
Add misc cleanup options before we unlock the system

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoosd addition fix in roles and cleanup shipped files
Vasu Kulkarni [Sat, 26 May 2018 02:11:33 +0000 (19:11 -0700)]
osd addition fix in roles and cleanup shipped files

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoship_utilities import fix
Vasu Kulkarni [Fri, 25 May 2018 22:51:24 +0000 (15:51 -0700)]
ship_utilities import fix

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoship utilities file during ceph-ansible task
Vasu Kulkarni [Fri, 25 May 2018 22:19:18 +0000 (18:19 -0400)]
ship utilities file during ceph-ansible task

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agodetect osd id's and use them in new roles as ansible can change the order
Vasu Kulkarni [Fri, 25 May 2018 05:04:08 +0000 (01:04 -0400)]
detect osd id's and use them in new roles as ansible can change the order
of osd's that are bootstrapped

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoFix roles after cluster is setup for mon/mgr/mds and also
Vasu Kulkarni [Thu, 24 May 2018 22:39:32 +0000 (18:39 -0400)]
Fix roles after cluster is setup for mon/mgr/mds and also
setup daemons for use by ceph_manager

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoFog provides fresh setup so there is no need to do any precleanup
Vasu Kulkarni [Thu, 3 May 2018 22:30:37 +0000 (15:30 -0700)]
Fog provides fresh setup so there is no need to do any precleanup

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agofast fix for nvme test
Vasu Kulkarni [Fri, 16 Mar 2018 01:38:03 +0000 (18:38 -0700)]
fast fix for nvme test

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoDisable running cm-ansible by default, FOG images have already
Vasu Kulkarni [Thu, 15 Mar 2018 00:06:34 +0000 (17:06 -0700)]
Disable running cm-ansible by default, FOG images have already
run cm-ansible code and it would be redundant to run them again.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agodont poweroff nodes, leave it as is for adhoc debug if needed.
Vasu Kulkarni [Tue, 27 Feb 2018 20:50:23 +0000 (12:50 -0800)]
dont poweroff nodes, leave it as is for adhoc debug if needed.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoOcto lab has no VPS and its default FOG, so we dont
Vasu Kulkarni [Thu, 15 Feb 2018 21:19:50 +0000 (13:19 -0800)]
Octo lab has no VPS and its default FOG, so we dont
need any data[os_type] to be sent at all for queue requests

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoskip specific os_type/os_version request when FOG reimage in place
Vasu Kulkarni [Thu, 15 Feb 2018 21:08:05 +0000 (13:08 -0800)]
skip specific os_type/os_version request when FOG reimage in place

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoEnable FOG reimage by default for all machine types
Vasu Kulkarni [Wed, 14 Feb 2018 17:33:01 +0000 (09:33 -0800)]
Enable FOG reimage by default for all machine types

the downside: you can't control from teuthology.yaml what
machines can be enabled for FOG

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agorun git_ignore_ssl task after ansible setup
Vasu Kulkarni [Sat, 6 Jan 2018 19:51:10 +0000 (11:51 -0800)]
run git_ignore_ssl task after ansible setup

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoAdd internal task to ignore ssl errors while
Vasu Kulkarni [Sat, 6 Jan 2018 02:04:25 +0000 (18:04 -0800)]
Add internal task to ignore ssl errors while
cloning from untrusted gitlab

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoAdd option to schedule jobs using simple subset
Vasu Kulkarni [Sat, 27 Jan 2018 02:32:15 +0000 (18:32 -0800)]
Add option to schedule jobs using simple subset

Total number of jobs are divided into equal subsets based on
user options and can be scheduled using 1/10, 2/10 etc

If a suite produces 100 jobs, 1/10 would schedule first 10 jobs,
2/10 will schedule next set of 10 jobs and so on.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agouse update_latest_rh_kernel for redhat configs
Vasu Kulkarni [Fri, 26 Jan 2018 01:11:01 +0000 (17:11 -0800)]
use update_latest_rh_kernel for redhat configs

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoAdd option to retry for exec task.
Vasu Kulkarni [Fri, 5 Jan 2018 01:51:04 +0000 (17:51 -0800)]
Add option to retry for exec task.

retry and sleep_for_retry options can be used with exec task without
needing to use lengthy shell based while loops for similar
task. Also adds timeout option for commands that can be used
to return within specified time.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agowhen cm-ansible breaks!, backup to skip cm-ansible and run jobs
Vasu Kulkarni [Fri, 26 Jan 2018 00:31:56 +0000 (16:31 -0800)]
when cm-ansible breaks!, backup to skip cm-ansible and run jobs

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agomisc run-rh-task fixes during rhcs 3.0
Vasu Kulkarni [Fri, 26 Jan 2018 00:23:06 +0000 (16:23 -0800)]
misc run-rh-task fixes during rhcs 3.0

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agodebian repo is setup during internal task and no longer needed.
Vasu Kulkarni [Thu, 25 Jan 2018 23:03:21 +0000 (15:03 -0800)]
debian repo is setup during internal task and no longer needed.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoSome more cleanup before repo setup and skip subscription manager
Vasu Kulkarni [Thu, 25 Jan 2018 22:49:40 +0000 (14:49 -0800)]
Some more cleanup before repo setup and skip subscription manager
for beta ISO tests

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>
6 years agoMerge pull request #1244 from tchaikov/wip-install-downgrade-packages
Kefu Chai [Thu, 13 Dec 2018 03:01:00 +0000 (11:01 +0800)]
Merge pull request #1244 from tchaikov/wip-install-downgrade-packages

task/install: add "downgrade_packages" option for "install" task

Reviewed-by: Neha Ojha <nojha@redhat.com>
6 years agotask/install: add "downgrade_packages" option for "install" task 1244/head
Kefu Chai [Tue, 11 Dec 2018 12:19:08 +0000 (20:19 +0800)]
task/install: add "downgrade_packages" option for "install" task

in rados/thrash-old-clients, hammer or jewel packages is installed. but
yum does not allow downgrade a package by default, if a newer version is
already installed. in this case, librbd1 and librados2 are installed as
dependencies of qemu-kvm. their version is 1:10.2.5-4.el7 at the time of
writing in CentOS/RHEL 7.5. so if we want to install librbd1 or
librados2 from jewel, yum will simply consider the requirement is
already fulfilled and hence do nothing. if we want to install
ceph-radosgw from jewel, yum will fail, as ceph-radosgw depends on
librados2 and other Ceph packages of the same version from jewel. but
librbd1 and librados2 have been already installed. the error message
looks like:

Error: Package: 1:ceph-common-0.94.10-87.g116a558.el7.x86_64 (Ceph)
           Requires: librados2 = 1:0.94.10-87.g116a558.el7
           Installed: 1:librados2-10.2.5-4.el7.x86_64 (@base)
               librados2 = 1:10.2.5-4.el7
           Available: 1:librados2-0.94.10-87.g116a558.el7.x86_64 (Ceph)
               librados2 = 1:0.94.10-87.g116a558.el7

so we need to downgrade librbd1 and librados2 first.

in this change, "downgrade_package" option is added for "install" task,
so we can specify packages to be downgraded to given version, these
packages won't be installed twice if they are also specified by
"install" task elsewhere to be installed.

Signed-off-by: Kefu Chai <kchai@redhat.com>
6 years agotask/install: define "project" before using it
Kefu Chai [Tue, 11 Dec 2018 09:31:08 +0000 (17:31 +0800)]
task/install: define "project" before using it

no need to define `project` at the beginning of install(), we can
just use `config.get('project', 'ceph')`. better readability this way,
as we will reference `ceph` in the same expression.

Signed-off-by: Kefu Chai <kchai@redhat.com>
6 years agoMerge pull request #1243 from tchaikov/wip-install-pkg-order
Kefu Chai [Tue, 11 Dec 2018 01:30:46 +0000 (09:30 +0800)]
Merge pull request #1243 from tchaikov/wip-install-pkg-order

task/install: keep the order of pkgs to be installed

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
6 years agotask/install: keep the order of pkgs to be installed 1243/head
Kefu Chai [Mon, 10 Dec 2018 14:50:38 +0000 (22:50 +0800)]
task/install: keep the order of pkgs to be installed

before this change, the package list to be installed are ordered by
the hash function used by set().

after this change, the order of packages are not changed. so we can
ensure the order of package installation is the identical to that
of package list in yaml.

Signed-off-by: Kefu Chai <kchai@redhat.com>
6 years agoMerge pull request #1241 from ceph/wip-job-id
vasukulkarni [Mon, 3 Dec 2018 23:32:42 +0000 (15:32 -0800)]
Merge pull request #1241 from ceph/wip-job-id

Remove unused 'machine_type' parameter from kill_job()

6 years agoMerge pull request #1194 from adamwg/no-validate-refs
vasukulkarni [Mon, 3 Dec 2018 23:32:21 +0000 (15:32 -0800)]
Merge pull request #1194 from adamwg/no-validate-refs

Add no-validate-refs option to teuthology-suite

6 years agoMerge pull request #1239 from ceph/wip-addrvec
Sage Weil [Fri, 30 Nov 2018 19:18:01 +0000 (13:18 -0600)]
Merge pull request #1239 from ceph/wip-addrvec

misc: optionally bind to v1+v2 addrs for mons

6 years agoMerge pull request #1237 from ceph/wip-rhel8
Zack Cerza [Fri, 30 Nov 2018 18:42:13 +0000 (11:42 -0700)]
Merge pull request #1237 from ceph/wip-rhel8

opsys: Add RHEL8

6 years agoMerge pull request #1238 from ceph/wip-instrument-import
Zack Cerza [Fri, 30 Nov 2018 18:41:24 +0000 (11:41 -0700)]
Merge pull request #1238 from ceph/wip-instrument-import

run_tasks.py: log the right ImportError message

6 years agoRemove unused 'machine_type' parameter from kill_job() 1241/head
Shilpa Jagannath [Wed, 28 Nov 2018 07:44:13 +0000 (13:14 +0530)]
Remove unused 'machine_type' parameter from kill_job()

Signed-off-by: Shilpa Jagannath <smanjara@redhat.com>
6 years agomisc: optionally bind to v1+v2 addrs for mons 1239/head
Sage Weil [Mon, 26 Nov 2018 17:59:51 +0000 (11:59 -0600)]
misc: optionally bind to v1+v2 addrs for mons

Signed-off-by: Sage Weil <sage@redhat.com>
6 years agorun_tasks.py: allow _import to raise the right ImportError 1238/head
Nathan Cutler [Thu, 22 Nov 2018 11:59:48 +0000 (12:59 +0100)]
run_tasks.py: allow _import to raise the right ImportError

It turns out it's possible for a file qa/tasks/foo.py to exist,
yet importing it still raises an ImportError because it references a
non-existent symbol.

In this case, teuthology was clobbering the real ImportError with its
own bogus text.

Fixes: http://tracker.ceph.com/issues/37370
Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agoAdd validate-sha1 option to teuthology-suite 1194/head
Adam Wolfe Gordon [Mon, 23 Jul 2018 22:28:05 +0000 (16:28 -0600)]
Add validate-sha1 option to teuthology-suite

Allow users to skip validation of SHA1s passed to -S when submitting
jobs via teuthology-suite. This is useful when testing against
non-GitHub repos, since teuthology can't validate refs for a plain git
repository.

Signed-off-by: Adam Wolfe Gordon <awg@digitalocean.com>
6 years agoMerge PR #1234 into master 1240/head
Patrick Donnelly [Wed, 21 Nov 2018 19:40:26 +0000 (11:40 -0800)]
Merge PR #1234 into master

* refs/pull/1234/head:
only pass system errors to hub handler
small refactor of bash code

Reviewed-by: Kefu Chai <kchai@redhat.com>
6 years agoopsys: Add RHEL8 1237/head
David Galloway [Tue, 20 Nov 2018 20:04:53 +0000 (15:04 -0500)]
opsys: Add RHEL8

Signed-off-by: David Galloway <dgallowa@redhat.com>
6 years agoonly pass system errors to hub handler 1234/head
Patrick Donnelly [Fri, 9 Nov 2018 19:31:04 +0000 (11:31 -0800)]
only pass system errors to hub handler

Fixes: http://tracker.ceph.com/issues/36731
Code suggested/contributed by Kefu.

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
6 years agosmall refactor of bash code
Patrick Donnelly [Fri, 9 Nov 2018 19:26:45 +0000 (11:26 -0800)]
small refactor of bash code

Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
6 years agoMerge pull request #1236 from ceph/wip-netaddr
Yuri Weinstein [Wed, 14 Nov 2018 16:19:43 +0000 (08:19 -0800)]
Merge pull request #1236 from ceph/wip-netaddr

ceph_ansible: Install netaddr

6 years agoceph_ansible: Install netaddr 1236/head
David Galloway [Tue, 13 Nov 2018 20:56:44 +0000 (15:56 -0500)]
ceph_ansible: Install netaddr

See https://github.com/ceph/ceph-ansible/pull/3199

Signed-off-by: David Galloway <dgallowa@redhat.com>
6 years agoMerge pull request #1230 from tchaikov/wip-better-remote-run
Kefu Chai [Mon, 12 Nov 2018 00:09:23 +0000 (08:09 +0800)]
Merge pull request #1230 from tchaikov/wip-better-remote-run

@zmc @smithfarm gents, i've updated the change to address the concerns. if no objections, i'd like merge as it is in two days.

6 years agoMerge pull request #1235 from ceph/wip-suppress-cpython
Sage Weil [Sat, 10 Nov 2018 20:11:08 +0000 (14:11 -0600)]
Merge pull request #1235 from ceph/wip-suppress-cpython

valgind: ignore all leaks relating to CPython code

6 years agovalgind: ignore all leaks relating to CPython code 1235/head
Sage Weil [Fri, 9 Nov 2018 14:53:03 +0000 (08:53 -0600)]
valgind: ignore all leaks relating to CPython code

Yes, this is a big hammer, and we are ignoring a lot.  However, it is a
HUGE step forward to what we do now, which is not check for ceph-mgr
leaks at all.

By adding this suppress I found and fixed 3 separate ceph-mgr leaks.  This
will let us prevent others (in non-Py code) from being introduced.

Signed-off-by: Sage Weil <sage@redhat.com>
6 years agoMerge pull request #1231 from kshtsk/wip-shortname-logging-for-ip
Gregory Farnum [Fri, 2 Nov 2018 20:07:23 +0000 (13:07 -0700)]
Merge pull request #1231 from kshtsk/wip-shortname-logging-for-ip

Fix ipv4 and ipv6 address logging for Remote.run

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
6 years agoMerge pull request #1233 from kshtsk/wip-remote-sh
Gregory Farnum [Fri, 2 Nov 2018 20:03:50 +0000 (13:03 -0700)]
Merge pull request #1233 from kshtsk/wip-remote-sh

orchestra: add remote.sh commands analogous to misc.sh

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
6 years agoorchestra: add remote.sh commands analogous to misc.sh 1233/head
Kyr Shatskyy [Tue, 30 Oct 2018 13:17:05 +0000 (14:17 +0100)]
orchestra: add remote.sh commands analogous to misc.sh

Adds a remote.sh similar to misc.sh, in fact a shortcut for remote.run,
but return output instead of proc
Example:
    my_name = Remote('127.0.0.1').sh('whoami')

Adds a remote.sh_file run a script as file on a remote with or without sudo
Example 1: Run python script
    Remote('127.0.0.1').sh_file("#!/usr/bin/env python3\n"
                                "import sys\n"
                                "print(sys.version_info)")
Example 2: Run script as root
    Remote('user@host.domain').sh_file("whoami", sudo=True,
                                       label="who-am-i-for-the-real")
Example 3: Run script as other user
    Remote('user@host.domain').sh_file("whoami", sudo='nobody', )

Signed-off-by: Kyr Shatskyy <kyrylo.shatskyy@suse.de>
6 years agoMerge pull request #1228 from zenglg/master
vasukulkarni [Mon, 29 Oct 2018 18:41:32 +0000 (11:41 -0700)]
Merge pull request #1228 from zenglg/master

Fix setenforce failed while selinux is disabled

6 years agoFix ipv4 and ipv6 address logging for Remote.run 1231/head
Kyrylo Shatskyy [Sun, 28 Oct 2018 17:56:14 +0000 (18:56 +0100)]
Fix ipv4 and ipv6 address logging for Remote.run

The Remote class does not respect ip addresses
when it comes to define shortnames. As a result,
the hostname is not shown correctly in the log.
For ipv4 it only shows first number of the octet.
For ipv6 it even does not allow to proceed,
and raises exception in orchestra.run.

Signed-off-by: Kyr Shatskyy <kyrylo.shatskyy@suse.de>
6 years agoorchestra.run: log the ssh command without prefix 1230/head
Kefu Chai [Fri, 26 Oct 2018 06:18:34 +0000 (14:18 +0800)]
orchestra.run: log the ssh command without prefix

run() also supports single string, but if we pass a longstring literal
which contains "\n", it renders log difficult to read.
in this change, multi-line command is logged in multiple lines,
and print the "prefix" in the first line, then print the command in
following lines without "prefix".

Signed-off-by: Kyr Shatskyy <kyrylo.shatskyy@suse.com>
6 years agotask/install: run install commands in a concatenated string
Kefu Chai [Thu, 25 Oct 2018 14:01:14 +0000 (22:01 +0800)]
task/install: run install commands in a concatenated string

this improve the readabiliy of long script.

FWIW, the `arg` will be passed as the `command` parameter of
paramiko.SSHClient.exec_command() if it is a `basestring`.

Signed-off-by: Kyr Shatskyy <kyrylo.shatskyy@gmail.com>
Signed-off-by: Kefu Chai <kchai@redhat.com>
6 years agoMerge pull request #1229 from tchaikov/wip-yum-install
Kefu Chai [Thu, 25 Oct 2018 12:45:36 +0000 (20:45 +0800)]
Merge pull request #1229 from tchaikov/wip-yum-install

task/install: run 'yum ... install' the right way

Reviewed-by: Nathan Cutler <ncutler@suse.com>
6 years agotask/install: run 'yum ... install' the right way 1229/head
Kefu Chai [Thu, 25 Oct 2018 11:25:53 +0000 (19:25 +0800)]
task/install: run 'yum ... install' the right way

after cd1cef6, we install rpm packages like

2018-10-25T06:31:33.578 INFO:teuthology.orchestra.run.smithi076:Running:
"sudo yum -y '' install '' '' libcephfs_jni1"
...
2018-10-25T06:31:34.180
INFO:teuthology.orchestra.run.smithi076.stderr:No such command: . Please
use /bin/yum --help
2018-10-25T06:31:34.193 DEBUG:teuthology.orchestra.run:got remote
process result: 1

which fails the qa run on rpm based distros.

in this change, the command line args are passed in itertools.chain(),
because orchestra.run.quote() takes an iterable which contains string or
Raw.

Signed-off-by: Kefu Chai <kchai@redhat.com>
6 years agoFix setenforce failed while selinux is disabled 1228/head
Linggang Zeng [Thu, 25 Oct 2018 06:49:03 +0000 (14:49 +0800)]
Fix setenforce failed while selinux is disabled

Signed-off-by: Linggang Zeng <linggang.zeng@easystack.cn>
6 years agoMerge pull request #1218 from kshtsk/wip-test-repo
Zack Cerza [Thu, 25 Oct 2018 01:07:32 +0000 (19:07 -0600)]
Merge pull request #1218 from kshtsk/wip-test-repo

Add custom repos before installing rpm packages

6 years agoMerge pull request #1227 from rjfd/wip-jwt-dependency
Zack Cerza [Wed, 24 Oct 2018 22:01:59 +0000 (16:01 -0600)]
Merge pull request #1227 from rjfd/wip-jwt-dependency

requirements.txt: add JWT dependency required by ceph-dashboard tests

6 years agoMerge pull request #1224 from ceph/wip-36522
Sage Weil [Tue, 23 Oct 2018 13:49:01 +0000 (08:49 -0500)]
Merge pull request #1224 from ceph/wip-36522

valgrind: suppress __libc_csu_init leaks

6 years agorequirements.txt: add JWT dependency required by ceph-dashboard tests 1227/head
Ricardo Dias [Tue, 23 Oct 2018 06:50:16 +0000 (07:50 +0100)]
requirements.txt: add JWT dependency required by ceph-dashboard tests

Signed-off-by: Ricardo Dias <rdias@suse.com>
6 years agoMerge pull request #1225 from ceph/wip-36499
Gregory Farnum [Tue, 23 Oct 2018 05:15:03 +0000 (22:15 -0700)]
Merge pull request #1225 from ceph/wip-36499

kernel: Update kernel using `DEBIAN_FRONTEND noninteractive`

Reviewed-by: Greg Farnum <gfarnum@redhat.com>
6 years agovalgrind: suppress __libc_csu_init leaks 1224/head
Sage Weil [Fri, 19 Oct 2018 14:08:26 +0000 (09:08 -0500)]
valgrind: suppress __libc_csu_init leaks

Fixes: http://tracker.ceph.com/issues/36522
6 years agokernel: Update kernel using `DEBIAN_FRONTEND noninteractive` 1225/head
David Galloway [Fri, 19 Oct 2018 12:23:34 +0000 (08:23 -0400)]
kernel: Update kernel using `DEBIAN_FRONTEND noninteractive`

Fixes: https://tracker.ceph.com/issues/36499
Signed-off-by: David Galloway <dgallowa@redhat.com>
6 years agoopenstack: add --test-repo CLI option 1218/head
Kyr Shatskyy [Thu, 26 Oct 2017 16:15:41 +0000 (18:15 +0200)]
openstack: add --test-repo CLI option

Add custom repos before installing rpm packages on test nodes.
Repository can be specified as a NAME:URL pair. Several repos
can be provided by specifying the option multiple times.
For example,
    --test-repo foo:http://example.com/repo/foo \
    --test-repo bar:http://example.com/repo/bar

gives two test package repositories named "foo" and "bar".

Conflicts:
scripts/openstack.py
teuthology/openstack/__init__.py
teuthology/orchestra/opsys.py
teuthology/task/install/rpm.py

6 years agoinstall: rpm: only one option per variable
Nathan Cutler [Wed, 30 Aug 2017 10:18:30 +0000 (12:18 +0200)]
install: rpm: only one option per variable

Although the variables are entitled, e.g., "pkg_mng_opts" they
really can only contain at most one option. Here's what happens
when they contain more than one:

Running: "sudo zypper '-n --no-gpg-checks' install --capability --no-recommends ceph-radosgw"
stderr:Unknown option ' '
stderr:Unknown option '-'
stderr:Unknown option '-'

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agotask: install: rpm: zypper install with --no-recommends
Nathan Cutler [Mon, 5 Jun 2017 17:55:23 +0000 (19:55 +0200)]
task: install: rpm: zypper install with --no-recommends

Since teuthology is used to automate CI testing, it is undesirable to
install anything more than is necessary to run the tests.

Fixes: https://github.com/SUSE/teuthology/issues/84
Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agoOverride failing package signature checks
Nathan Cutler [Wed, 30 Aug 2017 09:55:15 +0000 (11:55 +0200)]
Override failing package signature checks

The RPMs built by teuthology's buildpackages task are not signed and
after a recent zypper update the install task started to fail with

File 'repomd.xml' from repository 'ceph-rpm-under-test' is unsigned, continue? [yes/no] (no): no
Error building the cache: [ceph-rpm-under-test|http://149.202.175.91/ceph-rpm-sle12-x86_64-basic/sha1/3804e807353c9d125753b1cf4f6405f79db83d4e/x86_64] Valid metadata not found at specified URL

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agopackaging: add ceph-rpm-under-test zypper repo with high priority
Nathan Cutler [Thu, 11 May 2017 20:18:55 +0000 (22:18 +0200)]
packaging: add ceph-rpm-under-test zypper repo with high priority

Otherwise the ceph packages from OBS are preferred because RPM evaluates, e.g.,
12.0.2+git.1493341348.9148e53 as a higher version number than
12.0.2-276.gf27d4b00ed.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agopackaging: call the zypper repo ceph-rpm-under-test
Nathan Cutler [Sat, 11 Feb 2017 08:21:16 +0000 (09:21 +0100)]
packaging: call the zypper repo ceph-rpm-under-test

The zypper repo must have a name/alias and "ceph-rpm-under-test" seems better
than just "ceph-rpm"; it's a repo containing the ceph RPMs that are being
tested.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agopackaging: SUSEify GitbuilderProject._remove_rpm_repo()
Nathan Cutler [Sat, 11 Feb 2017 04:46:46 +0000 (05:46 +0100)]
packaging: SUSEify GitbuilderProject._remove_rpm_repo()

When a job completes, all Ceph packages installed are removed and then the repo
is removed as well. Repo removal fails with:

    Running: 'sudo yum -y erase ceph-release'
    sudo: yum: command not found

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agopackaging: add repo instead of installing ceph-release
Nathan Cutler [Thu, 9 Feb 2017 21:54:38 +0000 (22:54 +0100)]
packaging: add repo instead of installing ceph-release

The ceph-release RPM probably doesn't do what we want, anyway.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agoopsys.py: make from_os_release support SUSE
Nathan Cutler [Tue, 28 Feb 2017 11:11:42 +0000 (12:11 +0100)]
opsys.py: make from_os_release support SUSE

Fixes: #47
Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agoEnable SUSE Linux Enterprise as a teuthology target
Kyr Shatskyy [Wed, 10 Oct 2018 04:25:39 +0000 (06:25 +0200)]
Enable SUSE Linux Enterprise as a teuthology target

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agosetup: do not set ceph_qa_suite_git_url in ~/.teuthology.yaml
Nathan Cutler [Fri, 10 Feb 2017 14:38:06 +0000 (15:38 +0100)]
setup: do not set ceph_qa_suite_git_url in ~/.teuthology.yaml

When this value is set, it is necessary to explicitly give --suite-repo and
--suite-branch. We would rather have the defaults for these come from
--ceph-repo and --ceph.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
6 years agoopenstack: switch to --{ceph,suite}-repo
Loic Dachary [Fri, 10 Feb 2017 10:45:37 +0000 (11:45 +0100)]
openstack: switch to --{ceph,suite}-repo

--ceph-git-url and --ceph-qa-suite-git-url are now
arguments of teuthology-suite but were named --ceph-repo and
--suite-repo instead.

Signed-off-by: Loic Dachary <ldachary@redhat.com>