public/k3s
1
0
Fork 0
mirror of https://git.vectorsigma.ru/public/k3s.git synced 2026-07-30 12:09:38 +00:00
Code Issues Packages Projects Releases Wiki Activity
3,084 Commits 31 Branches 949 Tags
3287cab8e5eecac00a4b8cf3a2de7c11603b4e91
Commit Graph

160 Commits

Author SHA1 Message Date
Brad Davidson
a75951ba29 Send error response if member list cannot be retrieved 
Prevents joining nodes from being stuck with bad initial member list if there is a transient failure, or if they try to join themselves

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit d7cdbb7d4d)
 2024-04-11 10:01:23 -07:00
Vitor Savian
9176d7f68a Add tls for kine 
* Bump kine
* Add integration tests for kine with tls

Signed-off-by: Vitor Savian <vitor.savian@suse.com>
 2024-04-02 11:40:16 -03:00
Derek Nola
a81bc3325b Transition from deprecated pointer library to ptr (#9801) (#9825) 
Signed-off-by: Derek Nola <derek.nola@suse.com>
 2024-03-30 21:30:05 -07:00
Brad Davidson
87eae0a664 Clean up snapshotDir create/exists logic 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 3576ed4327)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-03-07 16:37:15 -08:00
Brad Davidson
135e76d272 Fix issue with etcd node name missing hostname 
* Set ServerNodeName in snapshot CLI setup
* Raise errer if ServerNodeName ends up empty some other way
* Fix status controller to use etcd node name annotation instead of prefix checking

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 82432a2df7)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-03-07 16:37:15 -08:00
Derek Nola
5efbb06874 Rename AgentReady to ContainerRuntimeReady for better clarity 
Signed-off-by: Derek Nola <derek.nola@suse.com>
 2024-02-21 13:50:47 -08:00
Brad Davidson
f6303cf14d Bump kine and set NotifyInterval to what the apiserver expects 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit de825845b2)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-02-10 00:49:41 -08:00
Brad Davidson
0089772cc7 Fix ipv6 endpoint address selection for on-demand snapshots 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 8224a3a7f6)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-02-10 00:49:41 -08:00
Brad Davidson
10dfb09383 Add check for etcd-snapshot-dir and fix panic in Walk 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 6ec1926f88)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-02-10 00:49:41 -08:00
Brad Davidson
9ac6519f5c Fix excessive retry on snapshot reconcile 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 4005600d4e)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-02-10 00:49:41 -08:00
Vitor Savian
4e80dccc92 Changed how lastHeartBeatTime works in the etcd condition 
Signed-off-by: Vitor Savian <vitor.savian@suse.com>
 2024-02-08 04:53:51 -03:00
Vitor Savian
5564a951d6 Error getting node in setEtcdStatusCondition 
Signed-off-by: Vitor Savian <vitor.savian@suse.com>

Added retry and changed nodes for

Signed-off-by: Vitor Savian <vitor.savian@suse.com>
 2024-01-12 18:12:16 -03:00
Brad Davidson
fed2f71417 Fix nil map in full snapshot configmap reconcile 
If a full reconcile wins the race against sync of an individual snapshot resource, or someone intentionally deletes the configmap, the data map could be nil and cause a crash.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 319dca3e82)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-01-11 16:30:56 -08:00
Brad Davidson
6bbdf19a10 Fix the OTHER log message that prints the wrong variable 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 1e663622d2)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2024-01-11 16:30:56 -08:00
Vitor Savian
e9b2e59ff4 Handle etcd status condition when cluster reset and disable etcd 
Signed-off-by: Vitor Savian <vitor.savian@suse.com>

Set condition if node is unhealthy

Signed-off-by: Vitor Savian <vitor.savian@suse.com>
 2024-01-09 12:00:37 -03:00
Brad Davidson
1aa0eac285 Print key instead of file path in snapshot metadata log message 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-11-21 14:49:38 -08:00
Brad Davidson
0624d2a51f Don't apply s3 retention if S3 client failed to initialize 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-11-21 14:49:38 -08:00
Brad Davidson
0473839f16 Don't request metadata when listing objects 
While some implementations may support it, it appears that most don't,
and some may in fact return an error if it is requested.

We already stat the object to get the metadata anyway, so this was
unnecessary if harmless on most implementations.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-11-21 14:49:38 -08:00
Brad Davidson
f270db4bbb Reorder snapshot configmap reconcile to reduce log spew during initial startup 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-11-17 10:09:38 -08:00
Vitor Savian
fd676a5a64 Handle nil pointer when runtime core is not ready in etcd 
Signed-off-by: Vitor <vitor.savian@suse.com>
 2023-11-16 15:46:46 -08:00
Brad Davidson
6f489f08f9 Fix issue with snapshot metadata configmap 
Omit snapshot list configmap entries for snapshots without extra metadata; reduce log level of warnings about missing s3 metadata files.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 2088218c5f)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-11-16 09:55:41 -08:00
Vitor Savian
3377e9d809 Added etcd status condition 
* Added ADR for etcd status

Signed-off-by: Vitor <vitor.savian@suse.com>

* Added etcd status condition

Signed-off-by: Vitor <vitor.savian@suse.com>

---------

Signed-off-by: Vitor <vitor.savian@suse.com>
 2023-11-13 11:05:04 -08:00
Brad Davidson
b6f23014b2 Don't try to read token hash and cluster id during cluster-reset 
These fields are only necessary when saving snapshots to S3, and will block restoration if attempted

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-27 15:06:50 -07:00
Brad Davidson
0f6e77feaa Manually requeue configmap reconcile when no nodes have reconciled snapshots 
Silences error message from lasso - this is a normal startup condition
when no snapshots exist so we shouldn't log nasty looking errors.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-18 15:09:39 -07:00
Brad Davidson
edc3fb2804 Re-enable etcd endpoint auto-sync 
Removing this in 002e6c43ee regressed
control-plane-only nodes, as we rely on the etcd client to update its
endpoint list internally so that we can use it to sync the load-balancer
address list.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-18 15:09:39 -07:00
Brad Davidson
78e57bc970 Start etcd client before ensuring self removal 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-14 00:38:56 -07:00
Brad Davidson
52c7d6c91d Add server token hash to CR and S3 
This required pulling the token hash stuff out of the cluster package, into util.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit d885162967)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
e7014cf859 Switch to managing ETCDSnapshotFile resources 
Reconcile snapshot CRs instead of ConfigMap; manage ConfigMap downstream from CR list

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 550ab36ab7)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
22c0916abd Move snapshot delete into local/s3 functions 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 5cd4f69bfa)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
df0fd0de49 Store extra metadata and cluster ID for snapshots 
Write the extra metadata both locally and to S3. These files are placed such that they will not be used by older versions of K3s that do not make use of them.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 7464007037)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
7375740ed4 Move s3 snapshot list functionality to s3.go 
Also, don't list ONLY s3 snapshots if S3 is enabled.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 80f909d0ca)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
487b3948a3 Consistently set snapshotFile timestamp 
Attempt to use timestamp from creation or filename instead of file/object modification times

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 8d47645312)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
d270e97825 Tidy s3 upload functions 
Consistently refer to object keys as such, simplify error handling.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit f1afe153a3)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
7a0eecd601 Elide old snapshot data when apiserver rejects configmap with ErrRequestEntityTooLarge 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 2b0e2e8ada)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
eed767ea74 Move etcd snapshot code into separate file 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 676b00aa0e)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Brad Davidson
9826b553c9 Disable HTTP on main etcd client port 
Fixes performance issue under load, ref: https://github.com/etcd-io/etcd/issues/15402 and https://github.com/kubernetes/kubernetes/pull/118460

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
(cherry picked from commit 8c73fd670b)
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-13 11:09:28 -07:00
Derek Nola
f294e2b22e [Release-1.27] Clear remove annotations on cluster reset (#8587) 
* Use admin kubeconfig instead of supervisor for etcd snapshot CLI

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>

* Skip creating CRDs and setting up event recorder for CLI controller context

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>

* Don't export functions not needed outside the etcd package

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>

* Reorganize Driver interface and etcd driver to avoid passing context and config into most calls

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>

* Clear remove annotations on cluster reset; refuse to delete last member from cluster

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>

---------

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
Co-authored-by: Brad Davidson <brad.davidson@rancher.com>
 2023-10-11 16:17:50 -07:00
Vitor Savian
e83b1ba4aa Fixed the etcd retention to delete orphaned snapshots based on the date (#8177) 
* Fix retention using name instead of date

Signed-off-by: Vitor <vitor.savian@suse.com>
 2023-08-14 18:48:59 -03:00
Ian Cardoso
e551308db8 fix for etcd-snapshot delete with --etcd-s3 flag (#8110) 
k3s etcd-snapshot save --etcd-s3 ... is creating a local snapshot and uploading it to s3 while k3s etcd-snapshot delete --etcd-s3 ... was deleting the snapshot only on s3 buckets, this commit change the behavior of delete to do it locally and on s3

Signed-off-by: Ian Cardoso <osodracnai@gmail.com>
 2023-08-04 14:26:32 -03:00
Vitor Savian
ca7aeed090 Etcd snapshots retention when node name changes (#8099) 
Fixed the etcd retention to delete orphaned snapshots

Signed-off-by: Vitor <vitor.savian@suse.com>
 2023-08-03 10:54:40 -03:00
Brad Davidson
aa76942d0f Add FilterCN function to prevent SAN Stuffing 
Wire up a node watch to collect addresses of server nodes, to prevent adding unauthorized SANs to the dynamiclistener cert.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-08-02 11:15:39 -07:00
Brad Davidson
e61fde93c1 Fix MemberList error handling and incorrect etcd-arg passthrough 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-04-28 22:04:30 -07:00
Brad Davidson
91afb38799 Retry cluster join on "too many learners" error 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-04-28 11:28:33 -07:00
Brad Davidson
d95980bba3 Lock bootstrap data with empty key to prevent conflicts 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-04-05 10:56:57 -07:00
Brad Davidson
b010db0cff Ensure that loopback is used for the advertised address when resetting 
Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-04-03 17:01:43 -07:00
Brad Davidson
0c302f4341 Fix etcd member deletion 
Turns out etcd-only nodes were never running **any** of the controllers,
so allowing multiple controllers didn't really fix things.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-02-14 09:39:41 -08:00
Brad Davidson
3d146d2f1b Allow for multiple sets of leader-elected controllers 
Addresses an issue where etcd controllers did not run on etcd-only nodes

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-02-10 10:46:48 -08:00
Brad Davidson
a298bfdb18 Add jitter to scheduled snapshots and retry harder on conflicts 
Also ensure that the snapshot job does not attempt to trigger multiple concurrent runs, as this is not supported.

Signed-off-by: Brad Davidson <brad.davidson@rancher.com>
 2023-01-11 14:32:03 -08:00
Derek Nola
06d81cb936 Replace deprecated ioutil package (#6230) 
* Replace ioutil package
* check integration test null pointer
* Remove rotate retries

Signed-off-by: Derek Nola <derek.nola@suse.com>
 2022-10-07 17:36:57 -07:00
Derek Nola
4c0bc8c046 Update etcd error to match correct url (#5909) 
Signed-off-by: Derek Nola <derek.nola@suse.com>
 2022-07-29 09:40:53 -07:00