From f25eedfe7049b449c49146e56b79fc6ed1743d4b Mon Sep 17 00:00:00 2001 From: "mergify[bot]" <37929162+mergify[bot]@users.noreply.github.com> Date: Tue, 20 Jun 2023 16:43:29 -0400 Subject: [PATCH] Fix panic when MaxRetryInterval is specified, but RetryInterval is not (#35820) (#35839) * Fix panic when MaxRetryInterval is specified, but RetryInterval is not * add changelog (cherry picked from commit 3b730034627968491d7df0381767a95f48b74820) Co-authored-by: Tibor Djurica Potpara --- CHANGELOG.next.asciidoc | 21 +++++++++++++++++++++ libbeat/publisher/queue/diskqueue/config.go | 2 +- 2 files changed, 22 insertions(+), 1 deletion(-) diff --git a/CHANGELOG.next.asciidoc b/CHANGELOG.next.asciidoc index 79d07f7eaf09..602490f73e62 100644 --- a/CHANGELOG.next.asciidoc +++ b/CHANGELOG.next.asciidoc @@ -43,6 +43,27 @@ https://github.com/elastic/beats/compare/v8.2.0\...main[Check the HEAD diff] - Add checks to ensure reloading of units if the configuration actually changed. {pull}34346[34346] +- Fix namespacing on self-monitoring {pull}32336[32336] +- Fix race condition when stopping runners {pull}32433[32433] +- Fix concurrent map writes when system/process code called from reporter code {pull}32491[32491] +- Log errors from the Elastic Agent V2 client errors channel. Avoids blocking when error occurs communicating with the Elastic Agent. {pull}34392[34392] +- Only log publish event messages in trace log level under elastic-agent. {pull}34391[34391] +- Fix issue where updating a single Elastic Agent configuration unit results in other units being turned off. {pull}34504[34504] +- Fix dropped events when monitor a beat under the agent and send its `Host info` log entry. {pull}34599[34599] +- Fix namespacing on self-monitoring {pull}32336[32336] +- Fix race condition when stopping runners {pull}32433[32433] +- Fix concurrent map writes when system/process code called from reporter code {pull}32491[32491] +- Fix panics when a processor is closed twice {pull}34647[34647] +- Update elastic-agent-system-metrics to v0.4.6 to allow builds on mips platforms. {pull}34674[34674] +- The Elasticsearch output now splits large requests instead of dropping them when it receives a StatusRequestEntityTooLarge error. {pull}34911[34911] +- Fix Beats started by agent do not respect the allow_older_versions: true configuration flag {issue}34227[34227] {pull}34964[34964] +- Fix performance issues when we have a lot of inputs starting and stopping by allowing to disable global processors under fleet. {issue}35000[35000] {pull}35031[35031] +- In cases where the matcher detects a non-string type in a match statement, report the error as a debug statement, and not a warning statement. {pull}35119[35119] +- 'add_cloud_metadata' processor - add cloud.region field for GCE cloud provider +- 'add_cloud_metadata' processor - update azure metadata api version to get missing `cloud.account.id` field +- Make sure k8s watchers are closed when closing k8s meta processor. {pull}35630[35630] +- Upgraded apache arrow library used in x-pack/libbeat/reader/parquet from v11 to v12.0.1 in order to fix cross-compilation issues {pull}35640[35640] +- Fix panic when MaxRetryInterval is specified, but RetryInterval is not {pull}35820[35820] *Auditbeat* diff --git a/libbeat/publisher/queue/diskqueue/config.go b/libbeat/publisher/queue/diskqueue/config.go index 08d77229dcc9..33af260fc85f 100644 --- a/libbeat/publisher/queue/diskqueue/config.go +++ b/libbeat/publisher/queue/diskqueue/config.go @@ -166,7 +166,7 @@ func SettingsForUserConfig(config *config.C) (Settings, error) { settings.RetryInterval = *userConfig.RetryInterval } if userConfig.MaxRetryInterval != nil { - settings.MaxRetryInterval = *userConfig.RetryInterval + settings.MaxRetryInterval = *userConfig.MaxRetryInterval } return settings, nil