Sam Batschelet
41061e56ad
Merge pull request #13139 from hexfusion/bp-12727
...
[release-3.4]: ClientV3: Ordering: Fix TestEndpointSwitchResolvesViolation test
2021-06-24 10:38:10 -04:00
Sam Batschelet
501d8f01ea
[release-3.4]: ClientV3: Ordering: Fix TestEndpointSwitchResolvesViolation test
...
Signed-off-by: Sam Batschelet <sbatsche@redhat.com>
2021-06-23 21:26:55 -04:00
Sam Batschelet
38669a0709
Merge pull request #13137 from hexfusion/track-modules
...
vendor: track vendor/modules.txt
2021-06-23 14:52:10 -04:00
Sam Batschelet
7489911d51
Merge pull request #13135 from serathius/actions-3.4
...
Migrate PR testing from travis to GitHub actions
2021-06-23 14:03:37 -04:00
Sam Batschelet
15b7954d03
vendor: track vendor/modules.txt
...
Signed-off-by: Sam Batschelet <sbatsche@redhat.com>
2021-06-23 13:56:39 -04:00
Marek Siarkowicz
6cc1345a0b
Migrate PR testing from travis to GitHub actions
2021-06-23 18:25:29 +02:00
Sam Batschelet
589a6993b8
Merge pull request #13101 from mrueg/backport-12864
...
[backport 3.4] fix check datascale command for https endpoints
2021-06-14 08:19:22 -04:00
Saeid Bostandoust
4bacd21e20
fix check datascale command for https endpoints
2021-06-11 11:58:20 +02:00
Sam Batschelet
0ecc337028
Merge pull request #13100 from tangcong/automated-cherry-pick-of-#13077-origin-release-3.4
...
[backport 3.4] embed: unlimit the recv msg size of grpc-gateway
2021-06-10 21:29:34 -04:00
spacewander
628fa1818e
embed: unlimit the recv msg size of grpc-gateway
...
Ensure the client which access etcd via grpc-gateway won't
be limited by the MaxCallRecvMsgSize. Here we choose the same
default value of etcdcli as grpc-gateway's MaxCallRecvMsgSize.
Fix https://github.com/etcd-io/etcd/issues/12576
2021-06-11 08:07:28 +08:00
Gyuho Lee
d19fbe541b
version: 3.4.16
...
Signed-off-by: Gyuho Lee <leegyuho@amazon.com>
v3.4.16
2021-05-12 01:52:43 +00:00
Piotr Tabor
6bbc85827b
Merge pull request #12917 from chaochn47/2021-05-03-backport-#12880
...
Backport-3.4 exclude alarms from health check conditionally
2021-05-06 10:21:09 +02:00
Chao Chen
dbde4f2d5e
Backport-3.4 exclude alarms from health check conditionally
2021-05-04 10:37:12 -07:00
Gyuho Lee
15715dcf1a
Merge pull request #12902 from MakDon/release-3.4
...
[Backport-3.4] etcdserver/mvcc: update trace.Step condition
2021-04-28 11:05:35 -07:00
makdon
963d3b9369
etcdserver/mvcc: update trace.Step condition
...
backport PR #12894 to release-3.4
2021-04-28 11:35:49 +08:00
Piotr Tabor
ba829044f5
Merge pull request #12888 from chaochn47/2021-04-22-cherry-pick-12871
...
Backport-3.4 etcdserver/util.go: reduce memory when logging range requests
2021-04-23 00:45:20 +02:00
Chao Chen
c4eb81af99
Backport-3.4 etcdserver/util.go: reduce memory when logging range requests
2021-04-22 15:07:44 -07:00
Piotr Tabor
ceafa1b33e
Merge pull request #12882 from lilic/bump-go-12
...
.travis,Makefile,functional: Bump go 1.12 version to v1.12.17
2021-04-20 23:33:23 +02:00
Lili Cosic
5890bc8bd6
.travis,Makefile,functional: Bump go 1.12 version to v1.12.17
...
This version was already used to build the release v3.4.15.
2021-04-20 14:00:44 +02:00
Piotr Tabor
c274aa5ea4
Merge pull request #12849 from lilic/test-go-1-15
...
[release-3.4]: .travis.yml: Test with go v1.15.11
2021-04-19 18:17:05 +02:00
Piotr Tabor
276ee962ec
integration: Fix 'go test --tags cluster_proxy --timeout=30m -v ./integration/...'
...
grpc proxy opens additional 2 watching channels. The metric is shared
between etcd-server & grpc_proxy, so all assertions on number of open
watch channels need to take in consideration the additional "2"
channels.
2021-04-19 16:41:28 +02:00
Lili Cosic
8d1b8335e3
pkg/tlsutil: Adjust cipher suites for go 1.12
...
Cherry-pick of 60e44286fa3c0c0 from master branch does not work due to
missing `tls.CipherSuites()` function. We work around by using go build
tags for both the building and tests.
2021-04-19 11:49:13 +02:00
Piotr Tabor
c3f447a698
Fix pkg/tlsutil (test) to not fail on 386.
...
In fact this commit rewrites the functionality to use upstream list of
ciphers instead of checking whether the lists are in sync using ast
analysis.
2021-04-19 11:49:13 +02:00
Lili Cosic
85e037d9c6
bill-of-materials.json: Update golang.org/x/sys
2021-04-19 11:49:13 +02:00
Lili Cosic
a1691be1bd
.travis,test: Turn race off in Travis for go version 1.15
...
Currently with race it fails, we can enable this at a later point.
2021-04-19 11:49:13 +02:00
Vimal K
df35086b6a
integration : fix TestTLSClientCipherSuitesMismatch in go1.13
...
In go1.13, the TLS13 is enablled by default, and as per go1.13 release notes :
TLS 1.3 cipher suites are not configurable. All supported cipher suites are safe,
and if PreferServerCipherSuites is set in Config the preference order is based
on the available hardware.
Fixing the test case for go1.13 by limiting the TLS version to TLS12
2021-04-19 11:18:14 +02:00
Lili Cosic
eeefd614c8
vendor: Run go mod vendor
2021-04-19 11:18:14 +02:00
Lili Cosic
4276c33026
go.mod,go.sum: Bump github.com/creack/pty that includes patch
...
This patch is needed due to go 1.15 erroring on:
"Setctty set but Ctty not valid in child".
2021-04-19 11:18:13 +02:00
Lili Cosic
cfc08e5f06
go.mod,go.sum: Comply with go v1.15
2021-04-19 11:18:13 +02:00
Lili Cosic
0b7e4184e8
etcdserver,wal: Convert int to string using rune()
2021-04-19 11:18:13 +02:00
Lili Cosic
35bd924596
integration,raft,tests: Comply with go v1.15 gofmt
2021-04-19 11:18:13 +02:00
Lili Cosic
62596faeed
.travis.yml: Test with go v1.15.11
...
Currently in CI the tests are only run with go v1.12, this adds also go
v1.15.11.
Excludes certain variants for v1.15.
2021-04-19 11:18:13 +02:00
Piotr Tabor
b7e5f5bc12
Merge pull request #12839 from lilic/fix-go-version
...
[release-3.4]: Pin go version in go.mod to 1.12
2021-04-07 17:52:05 +02:00
Lili Cosic
91bed2e01f
pkpkg/testutil/leak.go: Allowlist created by testing.runTests.func1
2021-04-07 17:20:52 +02:00
Lili Cosic
b19eb0f339
vendor: Run go mod vendor
2021-04-07 15:25:32 +02:00
Lili Cosic
8557cb29ba
go.sum, go.mod: Run go mod tidy with go 1.12
2021-04-07 15:25:08 +02:00
Lili Cosic
ef415e3fe1
go.mod: Pin go to 1.12 version
...
As go 1.12.2 is what is tested in CI as well as recommended to be built
with 1.12.2 we should also pin to this in the go directive version.
2021-04-07 15:21:42 +02:00
Sam Batschelet
82eae9227c
Merge pull request #12803 from cwedgwood/metrics-3.4
...
etcdserver: fix incorrect metrics generated when clients cancel watches
2021-04-01 08:17:37 -04:00
Chris Wedgwood
656dc63eab
etcdserver: fix incorrect metrics generated when clients cancel watches
...
Manual cherry-pick of 9571325fe85173a60c89d6ac6ce3491c7b1ec7a4 for
release-3.4.
2021-03-31 22:59:29 -07:00
Piotr Tabor
30799c97be
Merge pull request #12815 from dbavatar/release-3.4-peervalidation
...
etcdserver: Fix PeerURL validation
2021-03-30 12:54:32 +02:00
Piotr Tabor
16fe9a89ff
Merge pull request #12816 from cwedgwood/3.4-relax-gate-timeout
...
integration: relax leader timeout from 3s to 4s
2021-03-30 12:53:27 +02:00
Chris Wedgwood
c499d9b047
integration: relax leader timeout from 3s to 4s
...
The integration jobs fail with timeouts slightly over 3s, increase
this marginally so false failures are less prevalent.
2021-03-29 10:17:44 -07:00
Piotr Tabor
2702f9e5f2
Merge pull request #12751 from cwedgwood/nofsyncdowrite
...
When using --unsafe-no-fsync still write out the data
2021-03-07 11:52:33 +01:00
Chris Wedgwood
94634fc258
etcdserver: when using --unsafe-no-fsync write data
...
There are situations where we don't wish to fsync but we do want to
write the data.
Typically this occurs in clusters where fsync latency (often the
result of firmware) transiently spikes. For Kubernetes clusters this
causes (many) elections which have knock-on effects such that the API
server will transiently fail causing other components fail in turn.
By writing the data (buffered and asynchronously flushed, so in most
situations the write is fast) and avoiding the fsync we no longer
trigger this situation and opportunistically write out the data.
Anecdotally:
Because the fsync is missing there is the argument that certain
types of failure events will cause data corruption or loss, in
testing this wasn't seen. If this was to occur the expectation is
the member can be readded to a cluster or worst-case restored from a
robust persisted snapshot.
The etcd members are deployed across isolated racks with different
power feeds. An instantaneous failure of all of them simultaneously
is unlikely.
Testing was usually of the form:
* create (Kubernetes) etcd write-churn by creating replicasets of
some 1000s of pods
* break/fail the leader
Failure testing included:
* hard node power-off events
* disk removal
* orderly reboots/shutdown
In all cases when the node recovered it was able to rejoin the
cluster and synchronize.
2021-03-05 10:09:52 -08:00
Sam Batschelet
afd6d8a40d
Merge pull request #12740 from hexfusion/cp-12448--release-3.4
...
Manual cherry pick of #12448 on release 3.4
2021-03-03 13:37:20 -05:00
Sam Batschelet
9aeabe447d
server: Added config parameter experimental-warning-apply-duration
...
Signed-off-by: Sam Batschelet <sbatsche@redhat.com>
2021-03-03 12:14:30 -05:00
Gyuho Lee
aa7126864d
version: 3.4.15
...
Signed-off-by: Gyuho Lee <leegyuho@amazon.com>
v3.4.15
2021-02-26 22:08:24 +00:00
Gyuho Lee
3be9460ddc
Merge pull request #12679 from chaochn47/backport_3.4_#12677
...
[Backport-3.4] etcdserver/api/etcdhttp: log successful etcd server side health check in debug level
2021-02-09 15:01:19 -08:00
Chao Chen
f27ef4d343
[Backport-3.4] etcdserver/api/etcdhttp: log successful etcd server side health check in debug level
...
ref. #12677
ref. 0b9cfa8677
2021-02-08 21:44:44 -08:00
Piotr Tabor
a1c5f59b59
Merge pull request #12402 from vitalif/release-3.4
...
etcdserver: Fix 64 KB websocket notification message limit
2021-02-03 09:19:21 +01:00