]>
 
 
git.apps.os.sepia.ceph.com Git - ceph-cm-ansible.git/log 
 
 
 
 
 
 
Zack Cerza  [Wed, 4 Apr 2018 18:01:53 +0000  (12:01 -0600)] 
 
Merge pull request #392 from ceph/wip-dhcp-server 
 
Add dhcp-server role 
 
David Galloway  [Tue, 3 Apr 2018 16:21:29 +0000  (12:21 -0400)] 
 
dhcp-server: Update set-next-server script to work with role 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Thu, 29 Mar 2018 20:01:47 +0000  (16:01 -0400)] 
 
Add dhcp-server role 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Wed, 21 Mar 2018 19:32:07 +0000  (13:32 -0600)] 
 
Merge pull request #389 from ceph/wip-console 
 
Serial output fixes for testnodes 
 
David Galloway  [Fri, 9 Mar 2018 21:51:25 +0000  (16:51 -0500)] 
 
cobbler: Have rc.local output go to console 
 
Usually if something goes wrong during the rc.local run, the machine 
won't be reachable to debug over the network.  Additionally, since we 
reimage every machine before each job now, it's impossible to debug why 
rc.local failed given a particular job.  This outputs rc.local to the 
tty specified in kernel_options so we can see the output in `$hostname_reimage` run logs. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Fri, 9 Mar 2018 19:58:08 +0000  (14:58 -0500)] 
 
cobbler: Write exact /etc/default/grub 
 
This fixes console output during Xenial and later.  Prior to this, the 
Plymouth boot screen would get loaded and "[37mUbuntu 16.04[-1;-1f[33m. 
[37m.  [37m.  [37m." would get repeated to the console until the login 
prompt shows up. 
 
Writing our own file instead of finding and replacing variables makes 
sure the settings are exactly what we want. 
 
This snippet is only used on Debian-based distros.  The default Cobbler 
snippet is used on RPM-based distros. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Tue, 20 Mar 2018 21:39:57 +0000  (15:39 -0600)] 
 
Merge pull request #388 from ceph/wip-rc-chill 
 
cobbler: Change method used to ping Cobbler host in rc.local 
 
David Galloway  [Tue, 20 Mar 2018 15:22:53 +0000  (11:22 -0400)] 
 
cobbler: Change method used to ping Cobbler host in rc.local 
 
I've observed a *very* occasional race condition where dhclient 
completes but the host can't ping Cobbler.  Instead of timing out 
waiting for one ping packet to return, we'll try pinging X number of 
times (based on $attempts number) and then give up. 
 
I'll paste an example of the race condition observed in the PR notes. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Mon, 19 Mar 2018 20:24:55 +0000  (14:24 -0600)] 
 
Merge pull request #387 from ceph/wip-part-doc 
 
testnode: Fix typo and clarify dict for drives_to_partition 
 
David Galloway  [Mon, 19 Mar 2018 18:38:40 +0000  (14:38 -0400)] 
 
testnode: Fix typo and clarify dict for drives_to_partition 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Thu, 15 Mar 2018 16:00:59 +0000  (10:00 -0600)] 
 
Merge pull request #385 from ceph/wip-ovpn-logs 
 
gateway: Rotate every day.. not when size is 100MB 
 
David Galloway  [Mon, 12 Mar 2018 14:56:25 +0000  (10:56 -0400)] 
 
gateway: Rotate every day.. not when size is 100MB 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 9 Mar 2018 17:50:46 +0000  (10:50 -0700)] 
 
Merge pull request #384 from ceph/wip-ovpn-logrotate 
 
gateway: Finally fix openvpn logging 
 
David Galloway  [Fri, 9 Mar 2018 16:46:31 +0000  (11:46 -0500)] 
 
gateway: Properly configure rotating logs 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Fri, 9 Mar 2018 16:25:29 +0000  (11:25 -0500)] 
 
gateway: Add some additional handlers for openvpn and rsyslog 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Tue, 27 Feb 2018 16:18:40 +0000  (09:18 -0700)] 
 
Merge pull request #382 from ceph/wip-relax 
 
cobbler: Allow a little leeway in rc.local when pinging http_server 
 
David Galloway  [Tue, 27 Feb 2018 02:36:55 +0000  (21:36 -0500)] 
 
cobbler: Allow a little leeway in rc.local when pinging http_server 
 
DHCP might not complete within that one second.  We'll relax the timeout 
a bit. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
vasukulkarni  [Mon, 26 Feb 2018 23:42:58 +0000  (15:42 -0800)] 
 
Merge pull request #381 from ceph/wip-rhel-beta 
 
common: Tolerate new subscription-manager behavior 
 
David Galloway  [Mon, 26 Feb 2018 23:29:28 +0000  (18:29 -0500)] 
 
common: Tolerate new subscription-manager behavior 
 
subscription-manager throws an error now if there are no releases to list 
for a subscribed Beta distro. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Mon, 26 Feb 2018 22:57:06 +0000  (15:57 -0700)] 
 
Merge pull request #380 from ceph/wip-satellite 
 
Add support for Red Hat Satellite 
 
David Galloway  [Mon, 26 Feb 2018 20:04:39 +0000  (15:04 -0500)] 
 
tools: Clean up katello facts when prepping RHEL image capture 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Mon, 26 Feb 2018 18:56:58 +0000  (13:56 -0500)] 
 
pcp: Disable role for now 
 
With the addition of RHEL to Sepia, teuthology will be running 
cephlab.yml on unregistered RHEL testnodes.  Since the PCP playbook gets run 
before the testnodes playbook, RHEL systems in Sepia won't be registered 
to our Satellite yet and PCP installation fails. 
 
We're not currently using PCP so we can disable the role and save some 
time and headache. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Mon, 26 Feb 2018 17:50:30 +0000  (12:50 -0500)] 
 
cobbler: Subscribe to satellite during RHEL kickstart if use_satellite 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Fri, 7 Oct 2016 20:26:22 +0000  (16:26 -0400)] 
 
common: Add support for registering with Red Hat Satellite 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Tue, 20 Feb 2018 21:46:05 +0000  (14:46 -0700)] 
 
Merge pull request #378 from ceph/wip-update-reboot 
 
Fixes for https://tracker.ceph.com/issues/23052 
 
David Galloway  [Tue, 20 Feb 2018 20:55:39 +0000  (15:55 -0500)] 
 
cobbler: Update kernel on RPM distros and subscribe to CDN on RHEL 
during kickstart 
 
Fixes: https://tracker.ceph.com/issues/23052  
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Tue, 20 Feb 2018 19:42:51 +0000  (14:42 -0500)] 
 
cobbler: echo short hostname into /etc/hostname during rc.local 
 
Fixes: https://tracker.ceph.com/issues/23052  
Ceph should be able to handle the FQDN but whatever. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 9 Feb 2018 21:49:56 +0000  (14:49 -0700)] 
 
Merge pull request #377 from ceph/wip-set-iface 
 
More rc.local NIC fixees 
 
David Galloway  [Thu, 8 Feb 2018 23:16:53 +0000  (18:16 -0500)] 
 
cobbler: Only write network config for our routable NIC 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Thu, 8 Feb 2018 23:12:15 +0000  (18:12 -0500)] 
 
cobbler: Remove commented if statement 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Thu, 8 Feb 2018 20:36:17 +0000  (13:36 -0700)] 
 
Merge pull request #376 from ceph/wip-unsubscribe 
 
tools: Unsubscribe RHEL machines during FOG image capture 
 
David Galloway  [Thu, 8 Feb 2018 20:22:17 +0000  (15:22 -0500)] 
 
tools: Unsubscribe RHEL machines during FOG image capture 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Thu, 1 Feb 2018 21:53:33 +0000  (14:53 -0700)] 
 
Merge pull request #375 from ceph/wip-ugh-ddns 
 
nameserver: Let records tasks coexist with DDNS 
 
David Galloway  [Thu, 1 Feb 2018 20:42:36 +0000  (15:42 -0500)] 
 
nameserver: Let records tasks coexist with DDNS 
 
It takes about 3 minutes for ansible to compile all the zone files. 
That was causing nsupdate/DDNS to overwrite any new records we wanted to 
add or change before named could be reloaded. 
 
This PR: 
  - Writes zone files to a temporary location 
  - Dumps pending DDNS changes into zone files 
  - Freezes DDNS zone files from updates 
  - Moves temporary zone files into place all at once 
  - Unfreezes DDNS zone files 
 
This results in about a 3 second window where DDNS updates will be 
refused which isn't great but we can at least update records while OVH 
jobs are running now. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 26 Jan 2018 17:15:20 +0000  (10:15 -0700)] 
 
Merge pull request #374 from ceph/wip-fog-server 
 
fog-server: Create role 
 
David Galloway  [Thu, 25 Jan 2018 20:27:24 +0000  (15:27 -0500)] 
 
Merge pull request #368 from ceph/wip-fw-update 
 
firmware: flashrom syntax changed for Xenial 
 
David Galloway  [Wed, 24 Jan 2018 18:47:15 +0000  (13:47 -0500)] 
 
fog-server: Create role 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Tue, 23 Jan 2018 23:26:04 +0000  (16:26 -0700)] 
 
Merge pull request #373 from ceph/wip-ntp-server 
 
ntp-server: Create NTP server role 
 
David Galloway  [Tue, 23 Jan 2018 16:44:32 +0000  (11:44 -0500)] 
 
ntp-server: Create NTP server role 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 19 Jan 2018 23:16:29 +0000  (16:16 -0700)] 
 
Merge pull request #372 from ceph/wip-eth0-mac 
 
cobbler: Use MAC address specified in ansible inventory instead of eth0 
 
David Galloway  [Fri, 19 Jan 2018 20:31:03 +0000  (15:31 -0500)] 
 
cobbler: Use MAC address specified in ansible inventory instead of eth0 
 
I concede.  Name it whatever you want, RHEL. 
 
This will allow the OS to use the "predictable naming" during anaconda 
and after firstboot preventing NIC names from switching like we're 
seeing in http://tracker.ceph.com/issues/22732 and http://tracker.ceph.com/issues/22643 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 19 Jan 2018 20:30:48 +0000  (13:30 -0700)] 
 
Merge pull request #370 from ceph/wip-nvme-down 
 
testnode: Check for and mark down systems with missing NVMe devices 
 
David Galloway  [Wed, 17 Jan 2018 19:40:15 +0000  (14:40 -0500)] 
 
Merge pull request #371 from ceph/wip-nvme-firmware 
 
tools: Playbook to update Intel NVMe firmware 
 
David Galloway  [Wed, 17 Jan 2018 19:37:44 +0000  (14:37 -0500)] 
 
tools: Playbook to update Intel NVMe firmware 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Wed, 17 Jan 2018 18:06:07 +0000  (13:06 -0500)] 
 
testnode: Check for and mark down systems with missing NVMe devices 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Tue, 16 Jan 2018 21:40:39 +0000  (14:40 -0700)] 
 
Merge pull request #369 from ceph/wip-rclocal-break 
 
cobbler: Remove DHCP config for NICs if ifup fails in rc.local 
 
David Galloway  [Wed, 10 Jan 2018 21:17:55 +0000  (16:17 -0500)] 
 
cobbler: Remove DHCP config for NICs if ifup fails in rc.local 
 
An issue was discovered where rc.local bails if a testnode has multiple 
NICs cabled but each NIC doesn't have a DHCP reservation.  For example, 
some of the magnas have a second NIC cabled but are cabled to a tagged 
port on the switch so they can pass traffic via multiple VLANs. 
 
Fixes: http://tracker.ceph.com/issues/22651  
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Wed, 20 Dec 2017 22:29:45 +0000  (15:29 -0700)] 
 
Merge pull request #367 from ceph/wip-rhel75 
 
cobbler: Profile for RHEL7.5 
 
David Galloway  [Wed, 20 Dec 2017 21:05:39 +0000  (16:05 -0500)] 
 
cobbler: Profile for RHEL7.5 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Wed, 20 Dec 2017 18:14:02 +0000  (11:14 -0700)] 
 
Merge pull request #365 from ceph/wip-import-tasks 
 
all: Use import_tasks vs include_tasks so tags work again 
 
Zack Cerza  [Wed, 20 Dec 2017 17:38:03 +0000  (10:38 -0700)] 
 
Merge pull request #366 from ceph/wip-addl-nic 
 
testnode: Support for configuring a second NIC 
 
Zack Cerza  [Wed, 20 Dec 2017 17:35:28 +0000  (10:35 -0700)] 
 
Merge pull request #364 from ceph/wip-no-break 
 
cobbler: Don't break after first NIC is configured 
 
David Galloway  [Wed, 20 Dec 2017 01:05:47 +0000  (20:05 -0500)] 
 
cobbler: Don't break after first NIC is configured 
 
In case the system has multiple NICs cabled (like Octo's bruuni and 
pluto) 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Tue, 19 Dec 2017 00:59:47 +0000  (19:59 -0500)] 
 
testnode: Support for configuring a second NIC 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Tue, 19 Dec 2017 01:17:05 +0000  (20:17 -0500)] 
 
all: Use import_tasks vs include_tasks so tags work again 
 
See https://github.com/ansible/ansible/pull/33468 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Mon, 18 Dec 2017 20:35:48 +0000  (13:35 -0700)] 
 
Merge pull request #363 from ceph/wip-bmc-network 
 
tools: Playbook for setting BMC settings 
 
Zack Cerza  [Mon, 18 Dec 2017 20:35:14 +0000  (13:35 -0700)] 
 
Merge pull request #358 from ceph/wip-include 
 
all: s/include:/include_tasks:/g prepping for deprecation in v2.8 
 
vasukulkarni  [Thu, 7 Dec 2017 21:51:53 +0000  (13:51 -0800)] 
 
Merge pull request #362 from ceph/wip-easy-lvm 
 
testnode: Add support to quickly create LVs 
 
David Galloway  [Thu, 7 Dec 2017 19:21:46 +0000  (14:21 -0500)] 
 
tools: Playbook for setting BMC settings 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Wed, 6 Dec 2017 22:20:27 +0000  (17:20 -0500)] 
 
testnode: Add support to quickly create LVs 
 
Fixes: http://tracker.ceph.com/issues/22328  
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Wed, 6 Dec 2017 22:26:57 +0000  (15:26 -0700)] 
 
Merge pull request #361 from ceph/wip-also-ram 
 
testnode: Also stop zapping ram# devices 
 
David Galloway  [Wed, 6 Dec 2017 22:22:01 +0000  (17:22 -0500)] 
 
testnode: Also stop zapping ram# devices 
 
Saw this on smithi while testing new LVM changes 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Wed, 6 Dec 2017 18:56:33 +0000  (11:56 -0700)] 
 
Merge pull request #360 from ceph/wip-zap-loop 
 
testnode: Stop zapping loop devices 
 
David Galloway  [Wed, 6 Dec 2017 18:51:12 +0000  (13:51 -0500)] 
 
testnode: Stop zapping loop devices 
 
Because it's a waste of time 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Tue, 5 Dec 2017 20:58:39 +0000  (13:58 -0700)] 
 
Merge pull request #359 from ceph/wip-kopts-post 
 
cobbler_systems: Allow setting post install kernel options 
 
David Galloway  [Tue, 5 Dec 2017 20:00:08 +0000  (15:00 -0500)] 
 
cobbler_systems: Allow setting post install kernel options 
 
If defined in group_vars, kernel_options_post will be used.  Otherwise, 
it defaults to the kernel_options_post in defaults/main.yml. 
 
This fixes serial output during firstboot in Xenial or later (instead of 
the repeated "Ubuntu 16.04 [-1;-1f" garbage. 
 
Running `cobbler system edit` with --kopts-post="" is fine.  It just 
leaves that field blank if kernel_options_post is empty. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Fri, 1 Dec 2017 19:22:51 +0000  (14:22 -0500)] 
 
all: s/include:/include_tasks:/g prepping for deprecation in v2.8 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 1 Dec 2017 19:15:08 +0000  (12:15 -0700)] 
 
Merge pull request #357 from ceph/wip-update-xenial 
 
cobbler_profile: Replace 16.04 with 16.04.3 iso 
 
David Galloway  [Fri, 1 Dec 2017 18:55:42 +0000  (13:55 -0500)] 
 
cobbler_profile: Replace 16.04 with 16.04.3 iso 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 1 Dec 2017 16:50:49 +0000  (09:50 -0700)] 
 
Merge pull request #356 from ceph/wip-fog-csv 
 
tools: Add ansible role to generate FOG host csv file 
 
David Galloway  [Fri, 1 Dec 2017 16:44:40 +0000  (11:44 -0500)] 
 
tools: Add ansible role to generate FOG host csv file 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 1 Dec 2017 16:33:40 +0000  (09:33 -0700)] 
 
Merge pull request #355 from ceph/wip-ethtool 
 
cobbler: Install ethtool on reimage 
 
David Galloway  [Fri, 1 Dec 2017 16:31:32 +0000  (11:31 -0500)] 
 
cobbler: Install ethtool on reimage 
 
ethtool is used in the new rc.local hackery 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 1 Dec 2017 16:24:37 +0000  (09:24 -0700)] 
 
Merge pull request #354 from ceph/wip-ubuntu-pkgs 
 
Adjust Cobbler's Ubuntu preseed to work with Artful Aardvark and beyond 
 
David Galloway  [Fri, 1 Dec 2017 16:19:35 +0000  (11:19 -0500)] 
 
cobbler: Install grub to /dev/sda instead of whatever default is 
 
Apparently on some machines (namely bruuni in Octo), the default disk 
returns /dev/nvme0n1 which is not where the OS is installed. 
 
Since we already define which disk the OS should be installed on earlier 
in the preseed, it only makes sense to install grub on the same disk. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Fri, 1 Dec 2017 16:04:27 +0000  (11:04 -0500)] 
 
cobbler: Set package list for xenial and later instead of just xenial 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Mon, 27 Nov 2017 19:50:53 +0000  (11:50 -0800)] 
 
Merge pull request #346 from ceph/wip-set-hostname 
 
cobbler: Add logic to rc.local for FOG-provisioned machines 
 
Zack Cerza  [Mon, 27 Nov 2017 19:05:54 +0000  (11:05 -0800)] 
 
Merge pull request #353 from ceph/wip-delete-hostkeys 
 
tools: Delete ssh keys so FOG-imaged machines regenerate on firstboot 
 
David Galloway  [Mon, 27 Nov 2017 18:42:31 +0000  (13:42 -0500)] 
 
tools: Delete ssh keys so FOG-imaged machines regenerate on firstboot 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Mon, 27 Nov 2017 19:03:08 +0000  (14:03 -0500)] 
 
cobbler: Regenerate SSH host keys on Ubuntu firstboot if necessary 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Mon, 20 Nov 2017 23:56:23 +0000  (16:56 -0700)] 
 
Merge pull request #352 from ceph/wip-prep-fog 
 
Tools for FOG image prep 
 
David Galloway  [Mon, 20 Nov 2017 23:26:31 +0000  (18:26 -0500)] 
 
tools: Add script to modify next-server in DHCP 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Thu, 16 Nov 2017 19:36:05 +0000  (14:36 -0500)] 
 
tools: One-off playbook to prep a cobbler-imaged machine w/ FOG 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Fri, 17 Nov 2017 17:28:01 +0000  (12:28 -0500)] 
 
cobbler: Configure testnode NIC on firstboot after FOG image 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
David Galloway  [Fri, 10 Nov 2017 17:20:44 +0000  (12:20 -0500)] 
 
cobbler: Set hostname using DHCP every boot on testnodes 
 
This is for use with FOG.  When we create an OS image, the image 
inherits the testnode's hostname that was used to create the image. 
 
This rc.local snippet will make sure the testnode's hostname is set to 
whatever the IP's PTR record is on every boot. 
 
`rclocal_nameserver` should be set in the inventory groupvars. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Wed, 15 Nov 2017 16:04:06 +0000  (09:04 -0700)] 
 
Merge pull request #350 from ceph/wip-front-query 
 
vmhost: Allow task to fail but ignore errors 
 
David Galloway  [Wed, 15 Nov 2017 15:53:04 +0000  (10:53 -0500)] 
 
vmhost: Allow task to fail but ignore errors 
 
With a recent update to ansible, the changed task would never return a 
'failed' result with `failed_when` set.  `ignore_errors` is what we want 
so the task fails but the playbook proceeds. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Dan Mick  [Tue, 14 Nov 2017 01:09:24 +0000  (17:09 -0800)] 
 
Merge pull request #349 from ceph/wip-xfs-fix 
 
Omit XFS options on older OSes 
 
Reviewed-by: Dan Mick <dmick@redhat.com>  
Zack Cerza  [Tue, 14 Nov 2017 01:05:06 +0000  (18:05 -0700)] 
 
Omit XFS options on older OSes 
 
Signed-off-by: Zack Cerza <zack@redhat.com>  
Dan Mick  [Mon, 13 Nov 2017 23:28:49 +0000  (15:28 -0800)] 
 
Merge pull request #348 from ceph/wip-var-lib-nofail 
 
testnode: Avoid an unmountable /var/lib/ceph 
 
Reviewed-by: Dan Mick <dmick@redhat.com>  
Zack Cerza  [Mon, 13 Nov 2017 22:41:44 +0000  (15:41 -0700)] 
 
Don't write a version 5 XFS superblock 
 
Signed-off-by: Zack Cerza <zack@redhat.com>  
Zack Cerza  [Mon, 13 Nov 2017 22:08:33 +0000  (15:08 -0700)] 
 
Boot even if /var/lib partition won't mount 
 
Signed-off-by: Zack Cerza <zack@redhat.com>  
David Galloway  [Fri, 10 Nov 2017 17:48:27 +0000  (12:48 -0500)] 
 
Merge pull request #347 from ceph/wip-add-apt-repo 
 
testnode: Install software-properties-common 
 
Zack Cerza  [Fri, 10 Nov 2017 17:46:39 +0000  (10:46 -0700)] 
 
testnode: Install software-properties-common 
 
... on Ubuntu, for add-apt-repository 
 
Signed-off-by: Zack Cerza <zack@redhat.com>  
Zack Cerza  [Fri, 3 Nov 2017 16:25:17 +0000  (10:25 -0600)] 
 
Merge pull request #340 from ceph/wip-find-fail 
 
testnode: Avoid job failure due to fs loop 
 
David Galloway  [Fri, 29 Sep 2017 22:50:20 +0000  (18:50 -0400)] 
 
testnode: Avoid job failure due to fs loop 
 
Example: 
 
failed: [smithi205.front.sepia.ceph.com] (item=/var/run/) => {"changed": false, "cmd": "find /var/run/ -name \"*ceph*\"", "delta": "0:00:00.004976", "end": "2017-09-29 18:42:23.050412", "failed": true, "failed_when_result": true, "item": "/var/run/", "rc": 1, "start": "2017-09-29 18:42:23.045436", "stderr": "find: File system loop detected; '/var/run/rpc_pipefs/gssd' is part of the same file system loop as '/var/run/rpc_pipefs'.", "stderr_lines": ["find: File system loop detected; '/var/run/rpc_pipefs/gssd' is part of the same file system loop as '/var/run/rpc_pipefs'."], "stdout": "", "stdout_lines": []} 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Tue, 31 Oct 2017 20:20:18 +0000  (14:20 -0600)] 
 
Merge pull request #345 from ceph/wip-21989 
 
testnode: Be super aggressive with wiping LVM data 
 
David Galloway  [Tue, 31 Oct 2017 19:02:33 +0000  (15:02 -0400)] 
 
testnode: Be super aggressive with wiping LVM data 
 
Fixes: http://tracker.ceph.com/issues/21989  
Fixes: http://tracker.ceph.com/issues/21935 (hopefully)  
Signed-off-by: David Galloway <dgallowa@redhat.com>  
Zack Cerza  [Fri, 6 Oct 2017 17:33:23 +0000  (11:33 -0600)] 
 
Merge pull request #344 from ceph/wip-named-ddos 
 
nameserver: Double max amount of concurrent connections 
 
David Galloway  [Fri, 6 Oct 2017 17:29:04 +0000  (13:29 -0400)] 
 
nameserver: Double max amount of concurrent connections 
 
I observed an unintentional DoS on ns1.front last night right as most of 
the nightly scheduled jobs started up.  Lots of "nf_conntrack: table 
full, dropping packet" messages in the syslog. 
 
Doubling it should be safe. 
 
Signed-off-by: David Galloway <dgallowa@redhat.com>