4d1c594570
Removing standby from tests
...
After a dedicated test-case for standby it is not required
in other test cases.
2019-06-20 16:54:14 +02:00
966d21c4b6
Cosmetic changes
2019-06-20 16:35:15 +02:00
147e89797c
Adding a separate manifest fro standby cluster
2019-06-20 16:03:31 +02:00
e90679c1c7
Avoid error messages in operator logs for syncing
...
users and databases
2019-06-20 15:21:34 +02:00
6daecf10e4
Merge branch 'master' of https://github.com/zalando/postgres-operator into standby
2019-06-20 15:03:27 +02:00
326b306283
Minor formatiing updates
2019-06-20 14:02:25 +02:00
12458f2d32
Do not sync roles for standby
...
Standby cluster is streaming from the master and does not require
to create or use databases of it's own. Hence, bypass the creation
of users or databases.
Though the standby cluster would work without this also, but will have
errors in logs, this is to avoid that.
2019-06-20 12:19:00 +02:00
93bfed3e75
Add secret mount to operator ( #535 )
...
* add secret mount to operator
2019-06-19 12:40:49 +02:00
0ed92ed04e
add deletecollection verb ( #589 )
...
Fixing privileges to execute `patronictl remove`.
You could/should have also just used the operator delete cluster flow (remove manifest). It is not really the plan to use patroni inside a pod to remove a existing cluster.
2019-06-19 10:47:27 +02:00
6918394562
Add PDB configuration toggle ( #583 )
...
* Don't create an impossible disruption budget for smaller clusters.
* sync PDB also on update
2019-06-18 10:48:21 +02:00
3553144cda
Support subPath in generated container ( #452 )
...
* mounted volumes now provide a subPath
2019-06-17 15:49:01 +02:00
c65a9baedf
specify ReadOnlyRootFilesystem: false for pod security policies ( #560 )
...
Explicitly specify ReadOnlyRootFilesystem: false so kubernetes can pick
a less restrictive policy the operator has access to.
2019-06-17 14:03:33 +02:00
cd2a713a97
Merge branch 'master' of https://github.com/zalando/postgres-operator into standby
2019-06-17 10:54:21 +02:00
44acd7e4db
Not being able to register CRD is not a fatal error ( #444 )
...
Operator proceeds to checking if CRD is present and ready,
and if not, only then it is a fatal error.
2019-06-14 16:08:29 +02:00
9af13ed0df
Minor updates
...
- Remove extra warning message while creating standby
- Limit the number of pods to 1 for standby cluster
2019-06-14 16:01:23 +02:00
e1d9395338
rbac: add user-facing clusterroles ( #585 )
...
* rbac: add user-facing clusterroles
2019-06-14 15:59:51 +02:00
028b834ea6
postgres-operator deployment template: run operator as non-root, and with readonly filesystem ( #582 )
2019-06-14 15:47:08 +02:00
35a1f2cff8
Add Rafia to code owners ( #588 )
2019-06-14 14:56:40 +02:00
6fbfee3903
decouple clusterrole name and serviceaccount name ( #581 )
...
Decouple clusterrole name and service account name.
2019-06-14 14:24:23 +02:00
bbf28c4df7
Add additional S3 settings for cloning ( #497 )
2019-06-14 12:28:00 +02:00
14570dbc02
Merge branch 'master' of https://github.com/zalando/postgres-operator into standby
2019-06-14 12:14:17 +02:00
70c0e59cdb
Rebase
2019-06-13 16:40:15 +02:00
f16f24d3b4
Multiple fixes
...
- update docs to be more decriptive and cosistent with terminology
- Remove name and uid for the standby cluster
- Error out in case empty wal-path is provided for standby cluster
2019-06-13 12:27:19 +02:00
ad5fec9bee
docs: add storageclass to complete-postgres-manifest example ( #586 )
2019-06-11 16:25:02 +02:00
742a43a879
Adding docs and tests
2019-06-07 12:10:07 +02:00
2886027516
Some typos/spelling mistakes fix ( #580 )
...
Harmless typos fix.
2019-06-06 14:20:15 +02:00
69af2d60e5
Implement runner for e2e tests ( #548 )
...
* implement a runner for e2e tests
* move e2e tests to a Docker container
* integrate e2e tests into build pipelines
* add tests for multi-namespace support and logical backup jobs
* @FxKu implement the first e2e test for failovers
2019-06-05 17:07:27 +02:00
ec5b1d4d58
StatefulSet fsGroup config option to allow non-root spilo ( #531 )
...
* StatefulSet fsGroup config option to allow non-root spilo
* Allow Postgres CRD to overide SpiloFSGroup of the Operator.
* Document FSGroup of a Pod cannot be changed after creation.
2019-06-04 16:38:26 +02:00
5a0e95ac45
Add CRD configuration to Helm chart values.yaml ( #559 )
...
* add templates for CRDs incl. crd-install hooks
* support both config styles in values.yaml
* fix ServiceAccount naming in values.yaml
2019-06-03 14:48:32 +02:00
ebda39368e
database.go: remove hardcoded .svc.cluster.local dns suffix ( #561 )
...
* database.go: substitute hardcoded .svc.cluster.local dns suffix with config parameter
Use the pod's configured dns search path, for clusters where .svc.cluster.local is not correct.
2019-05-31 16:32:00 +02:00
3ffc8ac5fa
Sample logical backup image. ( #572 )
...
* Sample logical backup image.
Based on the earlier work by Dmitry Dolgov @erthalion
2019-05-31 15:51:58 +02:00
e5849873df
Cosmetic changes
2019-05-31 10:45:10 +02:00
b619569e28
Improve cluster sidecar documentation ( #573 )
2019-05-27 15:31:52 +02:00
24d412a562
generate spilo config can return error (with test) ( #570 )
...
* fix: raise explicit error when failing to generate spilo config
Signed-off-by: Stephane Tang <hi@stang.sh>
2019-05-22 17:35:03 +02:00
cd829f07f0
Minor fix
2019-05-22 15:21:42 +02:00
b36bdbc8ef
Function to create required environment variables for standby
2019-05-22 14:43:36 +02:00
1f4267eb05
fix: remove headless service config when deleting cluster ( #567 )
...
see: https://github.com/zalando/postgres-operator/issues/566
Signed-off-by: Stephane Tang <hi@stang.sh>
2019-05-21 13:49:34 +02:00
76ce1004ca
Rethinking standby cluster
...
Instead of having a full-fledged standby cluster as in Patroni,
here we need lesser information, viz, S3 path, cluster name and uid.
It is more closer to clone now.
2019-05-20 09:53:42 +02:00
f3e1e80aaf
Add logical backup ( #442 )
...
* Add k8s cron job to spawn logical backups
* Minor doc updates
2019-05-16 15:52:01 +02:00
b449dd2429
Bug fix
2019-05-15 09:44:38 +02:00
2c02b371e2
fix statefulset sync ( #563 )
2019-05-14 11:15:47 +02:00
4b9e6058e1
add update for CRD to RBAC ( #564 )
2019-05-13 17:36:15 +02:00
892b2c330e
Adding the support for standby cluster
2019-05-10 16:45:47 +02:00
f29bdaf96a
Override clone s3 bucket path ( #487 )
...
Override clone s3 bucket path
Add possibility to use a custom s3 bucket path for cloning a cluster
from an arbitrary bucket (e.g. from another k8s cluster). For that
a new config options is introduced `s3_wal_path`, that should point
to a location that spilo would understand.
2019-05-10 12:52:42 +02:00
ad0b250b5b
patch CRD on operator update ( #558 )
...
* patch existing CRD each time there is an operator update
2019-05-09 12:35:15 +02:00
0fbfbb23bb
Use /status subresource instead of plain manifest field ( #534 )
...
* turns PostgresStatus type into a struct with field PostgresClusterStatus
* setStatus patch target is now /status subresource
* unmarshalling PostgresStatus takes care of previous status field convention
* new simple bool functions status.Running(), status.Creating()
2019-05-07 12:01:45 +02:00
25e02ad755
Minor change to helm section in quickstart guide ( #549 )
...
* updated helm section in quickstart
2019-04-24 18:25:57 +02:00
c1d108a832
Fix CRD-based operator configuration ( #541 )
...
* Fix CRD-based operator configuration
* add inherited labels, update docker image
2019-04-15 13:52:38 +02:00
32f106990a
Fix build ( #542 )
...
* bump up golang version
* gcc fix
2019-04-15 11:52:07 +02:00
433bd8f500
Add paragraph on how to extend operator configuration in dev docs ( #538 )
...
* add paragraph on how to extend operator configuration in dev docs
2019-04-09 11:23:07 +02:00
Rafia Sabih
Markus
Taehyun Kim
Felix Kunde
Maxim Ivanov
Erik Inge Bolsø
teuto.net Netzdienste GmbH
Rafia Sabih
Sergey Dudoladov
Aaron Miller
Stephane T
Dmitry Dolgov