Commit Graph

267 Commits

Author SHA1 Message Date
Vladimir Mihailenco 58f7149e38 feat: add ContextTimeoutEnabled to respect context timeouts and deadlines 2022-10-11 10:22:42 +03:00
Vladimir Mihailenco dd9a200427
fix: improve pipelines retry logic (#2232)
* fix: improve pipelines retry logic
2022-10-06 14:05:55 +03:00
Stephanie Hingtgen 6327c52e60
Add ParseURL function for cluster mode (#1924)
* feat: add ParseClusterURL to allow for parsing of redis cluster urls into cluster options
2022-10-06 13:33:37 +03:00
jianghang baa48a4415 feat(pubsub): support sharded pub/sub 2022-08-04 00:05:22 +08:00
Vladimir Mihailenco f6a8adc50c fix: remove conn reaper from the pool and uptrace option names 2022-07-28 15:17:59 +03:00
Vladimir Mihailenco 60081bdd55
Merge pull request #2148 from signalfx/cluster-pipeline-err
fix: additional node failures in clustered pipelined reads
2022-07-28 13:27:22 +03:00
Knut Zuidema e061db8c13
refactor: remove unused context attributes (#2154)
* refactor: remove unused context field
2022-07-14 13:43:42 +03:00
Ben Keith 03376a5d9c fix: additional node failures in clustered pipelined reads
This causes a node to be marked as failing if the client is in READONLY mode
and any error occurs that is either a LOADING error (existing behvaior) or
any non-redis error (e.g. connection error, timeout, etc.).  This prevents that
node from being tried again on the next attempt as usually non-redis errors are
a strong indication that that node is not operational or degraded in some way.
2022-07-07 15:09:48 -04:00
Jacob Hoffman-Andrews fa4d1ea839 fix: log errors from cmdsInfoCache
Since the error gets swallowed in cmdInfo(), it's important to log it so
it's possible for the user to see what the error was.
2022-07-06 14:58:53 -07:00
Ben Keith 64f972fbea fix: disregard failed pings in updateLatency() for cluster nodes
It is possible that a ping command might fail very quickly due to an
immediate connection refused message or some other server failure
condition. In this case, the derived latency is not reliable and should
not be used to compare this node to other nodes in the cluster.

This change will only count successful Ping commands in the average
latency and in the case of no successful Pings, set the latency to an
arbitrarily high value to make this node have the least priority when
routing by latency.
2022-07-01 11:04:26 -04:00
Vladimir Mihailenco 4ddd7d1803 chore: remove WithContext 2022-06-05 09:58:05 +03:00
Vladimir Mihailenco a18fad5bd3 chore: v9 2022-06-04 17:42:06 +03:00
Monkey 335d946cd6
chore: sync master (#2051)
* Upgrade redis-server version (#1833)

* Upgrade redis-server version

Signed-off-by: monkey <golang@88.com>

* XAutoClaim changed the return value

Signed-off-by: monkey <golang@88.com>

* add cmd: geosearch, geosearchstore (#1836)

* add cmd: geosearch, geosearchstore

Signed-off-by: monkey92t <golang@88.com>

* GeoSearchQuery and GeoSearchLocationQuery changed to pointer passing

Signed-off-by: monkey92t <golang@88.com>

* Added missing method XInfoStreamFull to Cmdable interface

* Run go mod tidy in redisotel

Signed-off-by: Bogdan Drutu <bogdandrutu@gmail.com>

* Revert "ConnPool check  fd for bad conns (#1824)" (#1849)

This reverts commit 346bfafddd.

* Automate release process (#1852)

* Bump github.com/onsi/gomega from 1.10.5 to 1.14.0 (#1832)

* Bump github.com/onsi/gomega from 1.10.5 to 1.14.0

Bumps [github.com/onsi/gomega](https://github.com/onsi/gomega) from 1.10.5 to 1.14.0.
- [Release notes](https://github.com/onsi/gomega/releases)
- [Changelog](https://github.com/onsi/gomega/blob/master/CHANGELOG.md)
- [Commits](https://github.com/onsi/gomega/compare/v1.10.5...v1.14.0)

---
updated-dependencies:
- dependency-name: github.com/onsi/gomega
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

* Upgrade gomega to v1.15.0

Signed-off-by: monkey92t <golang@88.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: monkey92t <golang@88.com>

* Add version.go

* Fix otel example

* Fix package name in release script

* More fixes for otel example

* And more

* Fix release.sh

* Release v8.11.3 (release.sh)

* Create an annotated tag to give release.yml chance to run

* Tweak tag.sh

* Add Cmd.Slice helper to cast to []interface{} (#1859)

* after the connection pool is closed, no new connections should be added (#1863)

* after the connection pool is closed, no new connections should be added

Signed-off-by: monkey92t <golang@88.com>

* remove runGoroutine

Signed-off-by: monkey92t <golang@88.com>

* pool.popIdle add p.closed check

Signed-off-by: monkey92t <golang@88.com>

* upgrade golangci-lint v1.42.0

Signed-off-by: monkey92t <golang@88.com>

* Bump github.com/onsi/gomega from 1.15.0 to 1.16.0 (#1865)

Bumps [github.com/onsi/gomega](https://github.com/onsi/gomega) from 1.15.0 to 1.16.0.
- [Release notes](https://github.com/onsi/gomega/releases)
- [Changelog](https://github.com/onsi/gomega/blob/master/CHANGELOG.md)
- [Commits](https://github.com/onsi/gomega/compare/v1.15.0...v1.16.0)

---
updated-dependencies:
- dependency-name: github.com/onsi/gomega
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* Add go 1.17 to the build matrix

* Remove go 1.15 from build matrix

* Add scan struct example (#1870)

* Replace release job

* Bump github.com/cespare/xxhash/v2 from 2.1.1 to 2.1.2 (#1872)

Bumps [github.com/cespare/xxhash/v2](https://github.com/cespare/xxhash) from 2.1.1 to 2.1.2.
- [Release notes](https://github.com/cespare/xxhash/releases)
- [Commits](https://github.com/cespare/xxhash/compare/v2.1.1...v2.1.2)

---
updated-dependencies:
- dependency-name: github.com/cespare/xxhash/v2
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* Fix tag script to push tag by tag

* Fix releasing.md

* Fix/pubsub ping mutex (#1878)

* Fix PubSub.Ping to hold the lock

* Fix PubSub.Ping to hold the lock

* add write cmd data-race test

Signed-off-by: monkey92t <golang@88.com>

Co-authored-by: monkey92t <golang@88.com>

* chore: cleanup OpenTelemetry example

* chore: gofmt all code

* Refactor TestParseURL

This is in preparation for supporting query parameters
in ParseURL:

- use an expected *Options instance to execute assertions on
- extract assertions into helper function
- enable parallel testing
- condense test table

* Add query parameter parsing to ParseURL()

Before this change, ParseURL would only accept a very restricted
set of URLs (it returned an error, if it encountered any parameter).

This commit introduces the ability to process URLs like

	redis://localhost/1?dial_timeout=10s

and similar.

Go programs which were providing a configuration tunable (e.g.
CLI flag, config entry or environment variable) to configure
the Redis connection now don't need to perform this task
themselves.

* chore: add links to readme

* chore: fix discussions link

* empty hooks.withContext removed

* chore: gofmt

* chore: use conventional commits and auto-generate changelog

* feat: add acl auth support for sentinels

* chore: swap to acl auth at the test-level

* Add support for BLMove command

* chore: update dependencies

* chore: update link

* feat: add SetVal method for each command

* feat: add Cmd.{String,Int,Float,Bool}Slice helpers and an example

* chore: tweak GH actions to run all jobs

* chore: add Lua scripting example

* Fix Redis Cluster issue during roll outs of new nodes with same addr (#1914)

* fix: recycle connections in some Redis Cluster scenarios

This issue was surfaced in a Cloud Provider solution that used for
rolling out new nodes using the same address (hostname) of the nodes
that will be replaced in a Redis Cluster, while the former ones once
depromoted as Slaves would continue in service during some mintues
for redirecting traffic.

The solution basically identifies when the connection could be stale
since a MOVED response will be returned using the same address (hostname)
that is being used by the connection. At that moment we consider the
connection as no longer usable forcing to recycle the connection.

* chore: lazy reload when moved or ask

* chore: use conv commit message

* chore: release v8.11.4 (release.sh)

* fix: add whitespace for avoid unlikely colisions

* fix: format

* chore: fix links

* chore: use ctx parameter in cmdInfo

* Bump github.com/onsi/ginkgo from 1.16.4 to 1.16.5 (#1925)

Bumps [github.com/onsi/ginkgo](https://github.com/onsi/ginkgo) from 1.16.4 to 1.16.5.
- [Release notes](https://github.com/onsi/ginkgo/releases)
- [Changelog](https://github.com/onsi/ginkgo/blob/master/CHANGELOG.md)
- [Commits](https://github.com/onsi/ginkgo/compare/v1.16.4...v1.16.5)

---
updated-dependencies:
- dependency-name: github.com/onsi/ginkgo
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>

Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* feat: add support for time.Duration write and scan

* test: add test case for setting and scanning durations

* chore: fix linter

* fix(extra/redisotel): set span.kind attribute to client

According to the opentelemetry specification this should always be set to client for database client
libraries.

I've also removed the SetAttributes call and instead set the attributes during creation of the span.
This is what the library SHOULD be doing according to the opentelemetry api specification.

* chore: update otel example

* fix: update some argument counts in pre-allocs

In some cases number of pre-allocated places in
argument array is missing 1 or 2 elements,
which results in re-allocation of twice as large array

* chore: add example how to delete keys without a ttl

* chore: don't enable all lints

* chore(deps): bump github.com/onsi/gomega from 1.16.0 to 1.17.0

Bumps [github.com/onsi/gomega](https://github.com/onsi/gomega) from 1.16.0 to 1.17.0.
- [Release notes](https://github.com/onsi/gomega/releases)
- [Changelog](https://github.com/onsi/gomega/blob/master/CHANGELOG.md)
- [Commits](https://github.com/onsi/gomega/compare/v1.16.0...v1.17.0)

---
updated-dependencies:
- dependency-name: github.com/onsi/gomega
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

* feat: Add redis v7's NX, XX, GT, LT expire variants

* chore: add missing readme

* chore: tweak feature links

* chore: remove Discord

* fix: set timeout for WAIT command. Fixes #1963

* build: update `go` directive in `go.mod` to 1.17

This commit enables support for module graph pruning and lazy module
loading for projects that are at Go 1.17 or higher.

Reference: https://go.dev/ref/mod#go-mod-file-go
Reference: https://go.dev/ref/mod#graph-pruning
Reference: https://go.dev/ref/mod#lazy-loading
Signed-off-by: Eng Zer Jun <engzerjun@gmail.com>

* chore: update link

* chore: export cmder.SetFirstKeyPos to support build module commands

* feat(redisotel): ability to override TracerProvider (#1998)

* fix: add missing Expire methods to Cmdable

This is a followup to https://github.com/go-redis/redis/pull/1928

* chore(deps): bump github.com/onsi/gomega from 1.17.0 to 1.18.1

Bumps [github.com/onsi/gomega](https://github.com/onsi/gomega) from 1.17.0 to 1.18.1.
- [Release notes](https://github.com/onsi/gomega/releases)
- [Changelog](https://github.com/onsi/gomega/blob/master/CHANGELOG.md)
- [Commits](https://github.com/onsi/gomega/compare/v1.17.0...v1.18.1)

---
updated-dependencies:
- dependency-name: github.com/onsi/gomega
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>

* Update README.md (#2011)

chore: add fmt library in example code

* chore: instrumentation name and version (#2012)

* fix: invalid type assert in stringArg

* chore: cleanup

* fix: example/otel compile error (#2028)

* fix: rename Golang to Go (#2030)

https://go.dev/doc/faq#go_or_golang

* feat: add support for passing extra attributes added to spans

* feat: set net.peer.name and net.peer.port in otel example

* chore: tweak Uptrace copy

* feat: add support for COPY command (#2016)

* feat: add support for acl sentinel auth in universal client

* chore(deps): bump actions/checkout from 2 to 3

Bumps [actions/checkout](https://github.com/actions/checkout) from 2 to 3.
- [Release notes](https://github.com/actions/checkout/releases)
- [Changelog](https://github.com/actions/checkout/blob/main/CHANGELOG.md)
- [Commits](https://github.com/actions/checkout/compare/v2...v3)

---
updated-dependencies:
- dependency-name: actions/checkout
  dependency-type: direct:production
  update-type: version-update:semver-major
...

Signed-off-by: dependabot[bot] <support@github.com>

* chore: add hll example

* chore: tweak release script

* chore: release v8.11.5 (release.sh)

* chore: add discord back

Co-authored-by: Eugene Ponizovsky <ponizovsky@gmail.com>
Co-authored-by: Bogdan Drutu <bogdandrutu@gmail.com>
Co-authored-by: Vladimir Mihailenco <vladimir.webdev@gmail.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Kishan B <kishancs46@gmail.com>
Co-authored-by: Dominik Menke <dom@digineo.de>
Co-authored-by: Gökhan Özeloğlu <gozeloglu@gmail.com>
Co-authored-by: Justin Sievenpiper <justin@sievenpiper.co>
Co-authored-by: Алексей Романовский <aromanovsky@epiphan.com>
Co-authored-by: Stavros Panakakakis <stavrospanakakis@gmail.com>
Co-authored-by: Pau Freixes <pfreixes@gmail.com>
Co-authored-by: Ethan Hur <ethan0311@gmail.com>
Co-authored-by: Jackie <18378976+Pyrodash@users.noreply.github.com>
Co-authored-by: Kristinn Björgvin Árdal <kristinnardalsecondary@gmail.com>
Co-authored-by: ffenix113 <razerer@bigmir.net>
Co-authored-by: Bastien Penavayre <bastienPenava@gmail.com>
Co-authored-by: James3 Li(李麒傑) <james3_li@asus.com>
Co-authored-by: Eng Zer Jun <engzerjun@gmail.com>
Co-authored-by: gzjiangtao2014 <gzjiangtao2014@corp.netease.com>
Co-authored-by: Nelz <nelz9999@users.noreply.github.com>
Co-authored-by: Daniel Richter <Nexyz9@gmail.com>
Co-authored-by: Seyed Ali Ghaffari <ali.ghaffari@outlook.com>
Co-authored-by: lintanghui <lintanghui@bilibili.com>
Co-authored-by: hidu <duv123+github@gmail.com>
Co-authored-by: Jonas Lergell <jonas.lergell@volvocars.com>
Co-authored-by: Alex Kahn <alexanderkahn@gmail.com>
2022-03-19 12:40:31 +08:00
monkey92t 38d1749d56
adjust the code (#1842)
* Upgrade redis-server version (#1833)

* Upgrade redis-server version

Signed-off-by: monkey <golang@88.com>

* XAutoClaim changed the return value

Signed-off-by: monkey <golang@88.com>

* add cmd: geosearch, geosearchstore (#1836)

* add cmd: geosearch, geosearchstore

Signed-off-by: monkey92t <golang@88.com>

* GeoSearchQuery and GeoSearchLocationQuery changed to pointer passing

Signed-off-by: monkey92t <golang@88.com>

* adjust the code, and fix #1553, #1676

Signed-off-by: monkey92t <golang@88.com>
2021-08-02 19:01:01 +08:00
monkey 83454856c7 Merge branch 'master' into v9
Signed-off-by: monkey <golang@88.com>
2021-07-24 22:10:04 +08:00
Anatoly Rugalev 62fc2c821b
Check Failing() before serving random node (#1825)
* Check Failing() before serving random node

* Revert condition

* Addressed review comments

* Fallback to random failing node
2021-07-18 00:20:48 +08:00
ktaekwon000 ce40cd942a
Allow FIFO pool in redis client (#1820)
* Initial commit for FIFO pool

* Change PoolType string to PoolFIFO bool

* Remove redundant type

* add PoolFIFO option to all clients

Signed-off-by: monkey92t <golang@88.com>

Co-authored-by: Kim Tae Kwon <taekwon.kim@shopee.com>
Co-authored-by: monkey92t <golang@88.com>
2021-07-16 11:58:01 +08:00
monkey92t 916da5b2ce
sync v8.11.0 2021-07-02 20:56:25 +08:00
Alexey Makhov b7d23d446c
GOMAXPROCS usage instead of NumCPU (#1801) 2021-06-29 10:37:08 +08:00
monkey92t 8ad01240a4
Add support for resp3 protocol (#1739)
* support resp3 protocol

Signed-off-by: monkey <golang@88.com>

* Upgrade mod version limit go1.14

https://github.com/go-redis/redis/issues/1715#issuecomment-820685614

Signed-off-by: monkey <golang@88.com>

* Remove the redundant check of ReadReply

Signed-off-by: monkey <golang@88.com>

* fix the problem

Signed-off-by: monkey <golang@88.com>

* workflows add v9

Signed-off-by: monkey <golang@88.com>

* update StringStringMapCmd to MapStringStringCmd

Signed-off-by: monkey <golang@88.com>
2021-04-27 10:04:46 +03:00
Vladimir Mihailenco 02a9c81ef1 Fix golangci-lint check 2021-03-23 11:02:29 +02:00
Vladimir Mihailenco b267a03f18 Don't accept context in lazy reload 2021-03-05 13:53:15 +02:00
Vladimir Mihailenco e4b2ca97cd Fix ReadOnly option in ClusterClient 2021-01-22 09:33:09 +02:00
Vladimir Mihailenco 76fd0eac61 Fix Tx pipeline hook 2021-01-09 09:57:46 +02:00
Ben Keith 74fdad907c ClusterSlots Config and ReadOnly
You have to set the ReadOnly flag on the ClusterOptions to make it fall back to
non-master nodes on reads. For non-clustered situations where you have multiple
nodes in a single ClusterSlot, it made the non-master node useless because it would
never route requests to it and if you set ReadOnly=true it would issue a READONLY
command to it, which would fail.

See comment in code for more details.
2021-01-06 12:14:04 -05:00
Vladimir Mihailenco 143859e345 Rename 2020-12-16 17:45:06 +02:00
Prathik Rajendran M fd6643daa1 Add support to get cluster replica node for a given key 2020-12-16 16:55:40 +05:30
Dimitrij Denissenko 3fc96195a4 Reduce dependency chain 2020-09-17 14:32:08 +01:00
Vladimir Mihailenco c5d4b71f66 Fix race 2020-09-17 12:36:10 +03:00
GreenHedgehog 1a65d677b3
Fix not applying updated ClusterClient context after calling WithContext method (#1480)
pass context to command info cache init call
2020-09-14 15:30:50 +03:00
Vladimir Mihailenco cbce5dbfa2 Add Sentinel RouteBy 2020-09-11 16:06:08 +03:00
Vladimir Mihailenco afb0064872
Merge pull request #1484 from go-redis/fix/cluster-node-latency
Periodically update Cluster node latency
2020-09-11 14:29:38 +03:00
Vladimir Mihailenco 46e04ce67e Fix build 2020-09-11 14:17:23 +03:00
Vladimir Mihailenco 83d698027f Periodically update Cluster node latency 2020-09-11 12:33:03 +03:00
Vladimir Mihailenco b657760cca Tweak number of retries 2020-09-11 12:13:27 +03:00
Ehsan Afzali 6c85ee767a Add NewFailoverClusterClient 2020-09-09 15:37:52 +03:00
Pavlov Aleksey 8a5db20d32 fix nil node dereference when use RouteByLatency in cluster 2020-08-06 18:04:04 +03:00
Vladimir Mihailenco 21a1f58caf Retry timeout and retryable error 2020-07-24 15:52:15 +03:00
Vladimir Mihailenco 862dcbadae Try 3 random addrs to load commands 2020-07-19 09:51:41 +03:00
刘青 8a3f304b25
Log with context to associate all log by traceID (#1413)
* Log with context to associate all log by traceID
2020-07-18 09:04:36 +03:00
Vladimir Mihailenco 38c87c1f84 Fix golangci 2020-07-16 10:01:27 +03:00
Vladimir Mihailenco d66af70ae8 Rename clusterAddrs to activeAddrs 2020-07-09 11:35:19 +03:00
Vladimir Mihailenco f108b8bce3
Merge pull request #1377 from davidfu13/master
Make cluster client get updated nodes' addrs
2020-07-09 11:27:43 +03:00
Vladimir Mihailenco 1831363467 Add RecordError 2020-07-09 10:39:46 +03:00
Vladimir Mihailenco dc52593c8c Reload state on read only error 2020-06-30 11:13:15 +03:00
Vladimir Mihailenco 64995f9a51 Tweak auto-reloads 2020-06-29 17:51:47 +03:00
Vladimir Mihailenco 0ffefcde98 Use random node to load cluster slots 2020-06-29 17:26:11 +03:00
David Fu bb6761fe83 Make cluster client get updated nodes' addrs
During loading state, if all the nodes can't be connected, it's
possible that all the nodes' IP addresses have changed at the same
time. Clear the clusterAddrs so that the cluster client has chance
to get updated nodes' IP addresses by using the initial setting of
cluster addresses.
2020-06-15 15:24:18 +08:00
Vladimir Mihailenco 736fa28659 Cleanup 2020-06-10 15:04:12 +03:00
Vladimir Mihailenco ef82e3705c Add OnConnect context 2020-06-10 10:36:22 +03:00