42 Commits

Author SHA1 Message Date
Venkata, Krishna (kv988c)
1ad5857957 Fetch Total_OSD value from
ceph-client/values.yaml

Change-Id: I6af79af0627f85550655f5066dd6278bf0b5365a
2019-05-21 17:15:48 -05:00
Renis Makadia
5985b61286 Ceph-Client: Update, Enable and Cleanup helm tests
- Update ceph-client chart to
1) By default, enable ceph-client helm test. Update enabler
key in values.yaml to follow pattern as in other charts
2) Add needed dependancy for ceph-client helm tests
3) Update helm test script to reduce output and update
error msgs
4) Removed unwanted ENV variables SPECS and EXPECTED_POOLMINSIZE
- Update gate scripts to run helm test command

Change-Id: I6a0e4f5107e49dac081ac2037bcc0f9c0864793f
2019-05-18 03:09:45 +00:00
Roy Tang (rt7380)
85bd731562 Expose Anti-Affinity Weight Setting
This ps exposes the anti-affinity weight value, including
default, that will be consumed by the updated htk function.

Change-Id: Id8eb303674764ef8b0664f62040723aaf77e0a54
2019-05-14 17:04:52 -05:00
Dmitrii Kabanov
b7c07a595c Update image links for Ceph.
Updated the links according to the decision here [1].
The convention for images is <imagename>:<tagging>
where <tagging> takes the format <version>-<distro>.

[1]: http://eavesdrop.openstack.org/irclogs/%23openstack-helm/%23openstack-helm.2019-03-07.log.html#t2019-03-07T17:50:58

Change-Id: I84f8ce731e428f8ef035c008ff615e712c78a1f4
2019-04-24 14:53:42 +00:00
Pete Birley
f569cf0d5c Ceph-Client: Add security context to chart and read-only-fs
This PS adds the security context macros to the ceph-client chart,
and moves the default to read-only-rootfs for all containers.

Change-Id: I2fe03f31cc59e1cda2bf0396ae6e3aca5c440a16
Signed-off-by: Pete Birley <pete@port.direct>
2019-04-21 19:06:24 +00:00
Pete Birley
dece008337 Ceph: Make /etc/ceph and /run emptydirs uniformly across all pods
This PS updates the ceph charts to make /etc/ceph an emptydir
uniformly across all charts, both ensuring no default config is loaded,
and also permitting read-only filesystems to back the containers.

Additionally /run is uniformly applied across all long running pods
as a memory backed emptydir.

Change-Id: I00d1b15758b7eb4476fb950ddcb38db9a5149ad0
Signed-off-by: Pete Birley <pete@port.direct>
2019-04-21 19:06:18 +00:00
Pete Birley
2abf62ff4d OSH-Infra: Add emptydirs for tmp
This PS adds emptydirs backing the /tmp directory in pods, which
is required in most cases for full operation when using a read only
filesystem backing the container.

Additionally some yaml indent issues are resolved.

Change-Id: I8b7f1614da059783254aa6efc09facf23fca3cad
Signed-off-by: Pete Birley <pete@port.direct>
2019-04-20 20:50:59 +00:00
Steve Wilkerson
84f30ec103 Add release-annotation to pod spec, add missing annotations
This adds the release-annotation to the pod spec for the charts in
openstack-helm-infra. This also adds missing configmap annotations
to charts in openstack-helm-infra

Change-Id: Ie23f0c16a7a21d3929e98928db2bbcef69ae6490
2019-03-21 09:10:48 -05:00
Steve Taylor
65de349d58 Move ceph-mon's checkPGs cron job to ceph-client
- Move the cronjob from ceph-mon to ceph-client
- Adding ceph-rbd-pool job as dependencies for cronjob
- checkPGs manifest set to true so it will always run
in gate.

Co-Authored-By: Chinasubbareddy Mallavarapu <cr3938@att.com>,
                Renis Makadia <renis.makadia@att.com>

Change-Id: I9855d8d22265e78c7e2f5fa7ece69c9ff532ecb2
2019-03-19 20:53:08 +00:00
Zuul
d3531f1fba Merge "[CEPH] Enable iostat mgr module" 2019-02-04 09:45:13 +00:00
Matthew Heler
5de299b9b1 [CEPH] Enable iostat mgr module
Enable the iostat mgr module for Ceph. This module show the
current throughput and IOPS done on a Ceph cluster.

Change-Id: I2fe5b47401c15e349a49f345bacd99da39889373
2019-02-02 16:44:58 -06:00
Pete Birley
6ea6a85198 Ceph: Update default to use OSH image
This PS udpates the default image in the chart to the latest OSH image.

Change-Id: Ib8d2a72ad48049fe02560dc4405f0088890b6f64
Signed-off-by: Pete Birley <pete@port.direct>
2019-02-01 21:25:13 +00:00
Chinasubbareddy M
1abcde851e Ceph : cleanup ceph charts
This is  to clean ceph charts for unused variables and left over from
ceph chart split

Change-Id: Iec50599a031ae7acacc8eb0504f7146647450306
2019-01-11 19:53:36 +00:00
Matthew Heler
c0d028e245 Uplift Ceph charts to the Mimic release
Change the release of Ceph from 12.2.3 (Luminous) to latest 13.2.2
(Mimic). Additionally use supported RHEL/Centos Images rather then
Ubuntu images, which are now considered deprecated by Redhat.

- Uplift all Ceph images to the latest 13.2.2 ceph-container images.
- RadosGW by default will now use the Beast backend.
- RadosGW has relaxed settings enabled for S3 naming conventions.
- Increased RadosGW resource limits due to backend change.
- All Luminous specific tests now test for both Luminous/Mimic.
- Gate scripts will remove all none required ceph packages. This is
required to not conflict with the pid/gid that the Redhat container
uses.

Change-Id: I9c00f3baa6c427e6223596ade95c65c331e763fb
2019-01-05 14:38:38 +00:00
Chris Wedgwood
0c4e37391f 'NOP' cleanup for more consistent white-space use in charts
Where we have the style '{{ ...' we should use the style '... }}'.

Change-Id: Ic3e779e4681370d396f95d3804ca27db5b9d3642
2019-01-03 22:45:49 +00:00
Zuul
1b0d47bb01 Merge "[Ceph] Update ceph helm tests" 2018-12-17 18:23:51 +00:00
Matthew Heler
de69c68365 [Ceph] Update ceph helm tests
- Ensure the helm tests are logging all commands and variables

Change-Id: I4f4c553a3fbb4d77e9d1ab41c1c0c763c963cfd3
2018-12-15 13:47:43 -06:00
Pete Birley
5695a15f93 Ceph: Allow multiple test pods for ceph-client to be present in clusters
This ps allows multiple ceph test pods to be present in cluster with
more than one ceph deployment.

Change-Id: Ib8be8fc58e3a374dfcf6845988668433cf43655a
Signed-off-by: Pete Birley <pete@port.direct>
2018-12-12 10:03:15 -06:00
Renis Makadia
5aaa59f23f Disable Ceph helm test while being updated.
Change-Id: I6011ee6f41044647dea43fa2848058d723e194b8
2018-12-06 15:47:08 +00:00
Matthew Heler
35cce6cb43 Switch Ceph to IPs when DNS is down
Add helper scripts that are called by a POD to switch
Ceph from DNS to IPs. This POD will loop every 5 minutes
to catch cases where the DNS might be unavailable.

On a POD's Service start switch ceph.conf to using IPs rather
then DNS.

Change-Id: I402199f55792ca9f5f28e436ff44d4a6ac9b7cf9
2018-12-03 10:51:37 -06:00
Renis Makadia
b1005b23b4 Helm tests for Ceph-OSD and Ceph-Client charts
Change-Id: If4a846f0593b8679558662205a8560aa3cbb18ae
2018-12-01 08:08:00 +00:00
Matthew Heler
6e8c289c13 Add failure domains, and device classes for custom CRUSH rules
Largely inspired and taken from Kranthi's PS.

 - Add support for creating custom CRUSH rules based off of failure
domains and device classes (ssd & hdd)
- Basic logic around the PG calculator to autodetect the number of
OSDs globally and per device class (required when using custom crush
rules that specify device classes).

Change-Id: I13a6f5eb21494746c2b77e340e8d0dcb0d81a591
2018-11-27 09:37:30 -06:00
Matthew Heler
5ce9f2eb3b Enable Ceph charts to be rack aware for CRUSH
Add support for a rack level CRUSH map. Rack level CRUSH support is
enabled by using the "rack_replicated_rule" crush rule.

Change-Id: I4df224f2821872faa2eddec2120832e9a22f4a7c
2018-11-20 09:07:36 -06:00
Zuul
fca344900f Merge "Enable the mgr balancer module by default." 2018-11-02 22:36:13 +00:00
Matthew Heler
a79562a28b Enable the mgr balancer module by default.
The balancer module will distribute PGs more evenly across OSDs. 
While CRUSH does a good job at this, it is not perfect and hot spots
(where an OSD has more PGs then it's peers) can occur.

Change-Id: Ic45a6bf745bdd09a3f5782e9e8bda89c3d3da2aa
2018-11-01 15:52:51 +00:00
Matthew Heler
6ef48d3706 Further performance tuning changes for Ceph
- Throttle down snap trimming as to lessen it's performance impact
(Setting just osd_snap_trim_priority isn't effective enough to throttle
down the impact)
osd_snap_trim_sleep: 0.1 (default 0)
osd_pg_max_concurrent_snap_trims: 1 (default 2)

- Align filestore_merge_threshold with upstream Ceph values
(A negative number disables this function, no change in behavior)
filestore_merge_threshold: -10 (formerly -50, default 10)

- Increase RGW pool thread size for more concurrent connections
rgw_thread_pool_size: 512 (default 100)

- Disable in-memory logs for the ms subsytem.
debug_ms: 0/0 (default 0/5)

- Formating cleanups

Change-Id: I4aefcb6e774cb3e1252e52ca6003cec495556467
2018-10-26 15:10:50 +00:00
Pete Birley
1144ccbbb2 Ceph: Update MGR check to allow use on hosts with fqdns defined
This PS updates the mgr check to allow use on hosts with fqdns
defined.

Change-Id: If1cb740e8093fbcafce846234c96db931409b436
Signed-off-by: Pete Birley <pete@port.direct>
2018-10-24 00:57:12 +00:00
Chinasubbareddy M
e23e372120 ceph-mgr: make prometheus module port configurable
this is to give example for prometheus module port configurable

Change-Id: I66844bb8ee59a58f7bfd3e3002a183779810e881
2018-10-23 15:40:43 -05:00
Matthew Heler
0de1d23895 Fix grep logic around weighting OSDs during ceph-client chart.
Change-Id: I7831ac07a53b9aaf3000e9f64bf8c17344723a8f
2018-10-17 15:58:24 -05:00
Chinasubbareddy M
793b3631b5 Ceph-mgr: make liveness to check through admin scoket
This is to update the mgr liveness script to use admin socket
instead of resolving ceph  mon fqdn

Change-Id: Id95f78afef44103a834312d0667d49947ee803a4
Co-Authored-By: Jean-Charles Lopez <jl970p@att.com>
2018-10-17 14:40:42 +00:00
Jean-Charles Lopez
55f1d2db57 Secure pool during deployment
Change-Id: Ifbeb956ab2c015deaed501ee4bff22dfc1e0404f
2018-10-17 04:53:53 +00:00
Zuul
570355b1d9 Merge "Initialize OSDs with a crush weight of 0 to prevent automatic rebalancing." 2018-10-17 02:45:45 +00:00
Matthew Heler
5efac315f7 Initialize OSDs with a crush weight of 0 to prevent automatic rebalancing.
Weight the OSDs based on reported disk size when ceph-client chart runs.

Change-Id: I9f4080a9843f1a63564cf71154841b351382bfe2
2018-10-16 21:33:49 +00:00
Steve Wilkerson
f3d8bda9d6 Grafana: Support multiple Ceph clusters with dashboards
This updates the Grafana Ceph dashboards to use templating to
determine which ceph-mgr to use for displaying ceph related
metrics.  This required setting the appropriate labels on the
ceph-mgr service to be able to distinguish between releases

Change-Id: Id2eceacadc5b6366d7bc6668bc16ccf5ba878e4a
2018-10-16 21:32:13 +00:00
Chinasubbareddy M
616aecd80a Ceph-client: make pool creation depedent on ceph-mgr service
This is to add dependency for pool creation untill ceph-mgr fully up.

Change-Id: Id3111810a855bedff62970091b225358c269cecd
2018-10-15 10:00:27 -05:00
Chinasubbareddy M
2f2cb7d567 Ceph: Add configmap hash as annotation
adding configmap  hash to following ds/deployments to trigger
rolling updates if there are any update for configmap

- ceph-mon
- ceph-mds
- ceph-mgr
- ceph-rgw

Change-Id: I4173cb12c18640c9b1a0e5a698d48f4735e250fb
2018-09-22 07:26:52 +00:00
Zuul
c10f9ce59e Merge "Modify Ceph default settings for improved performance" 2018-09-20 22:44:11 +00:00
Pete Birley
c80e65cadb Revert "Make it possible to secure pool during deployment"
This reverts commit 14247c334b6fe8a20caba47efba99ef24298106f.

As it blocks the Airship projects genesis process.

Change-Id: I3b13f4e8f397ee3bc59f85e9a38d8581306deb2d
2018-09-18 21:01:56 +00:00
Jean-Charles Lopez
14247c334b Make it possible to secure pool during deployment
Change-Id: If5e19720f538b506dbe8a551b79a972b4bf25020
2018-09-14 23:02:42 +00:00
Jean-Charles Lopez
c6cad19d11 Modify Ceph default settings for improved performance
Change-Id: Ia0d856e53f3bfdc1414264b468b576003dc23b6e
2018-09-13 07:47:42 -07:00
Pete Birley
bb3ff98d53 Add release uuid to pods and rc objects
This PS adds the ability to attach a release uuid to pods and rc
objects as desired. A follow up ps will add the ability to add arbitary
annotations to the same objects.

Change-Id: Iceedba457a03387f6fc44eb763a00fd57f9d84a5
Signed-off-by: Pete Birley <pete@port.direct>
2018-09-13 05:35:35 +00:00
Steve Wilkerson
25bc83b580 Ceph: Move Ceph charts to openstack-helm-infra
This continues the work of moving infrastructure related services
out of openstack-helm, by moving the ceph charts to openstack
helm infra instead.

Change-Id: I306ccd9d494f72a7946a7850f96d5c22f36eb8a0
2018-08-28 15:03:35 -05:00