Security Bugfix Release

Highlights

• This release fixes a security issue related to service accounts and their permissions for more details
• Performance improvement for large clusters with high IOPs requirements.

What's Changed

• fix: bucket metric of data usage for replication info will be ignore by @jiuker in #18895
• fix metrics panic in node metrics endpoint by @poornas in #18894
• Update list.md by @dvaldivia in #18907
• allow configuring maximum idle connections per host by @harshavardhana in #18908
• enable xattr capture by default by @harshavardhana in #18911
• reuse transports for callhome and remote tgt validation by @poornas in #18912
• exclude veeam virtual objects from replication by @poornas in #18918
• Add cgroup v2 support for memory limit by @vadmeste in #18905
• remove checkBucketExist check entirely to avoid fan-out calls by @harshavardhana in #18917
• Improve tracing & notification scalability by @klauspost in #18903
• use all the available nr_requests for NVMe by @harshavardhana in #18920
• add total usable capacity, free and used to DataUsageInfo() by @harshavardhana in #18921
• remove all the frivolous logs, that may or may not be actionable by @harshavardhana in #18922
• Update service file version in makefile by @donatello in #18925
• Correct small typo by @fwessels in #18923
• fix: permission checks for editing access keys by @donatello in #18928
• Keep ServiceV1 admin stop/restart API and mark as deprecated by @vadmeste in #18932

Full Changelog: RELEASE.2024-01-29T03-56-32Z...RELEASE.2024-01-31T20-20-33Z

Bugfix Release

Full Changelog: RELEASE.2024-01-28T22-35-53Z...RELEASE.2024-01-29T03-56-32Z

Highlights

• Fixes an ugly issue with runtime debug stack excessive logging regression from
  RELEASE.2024-01-28T22-35-53Z, users who upgraded to RELEASE.2024-01-28T22-35-53Z
  are advised to upgrade immediately.

Changelog

• 9987ff5 - avoid calling close for nil inbound/outblock channels (3 hours ago)
• cff8235 - remove getReplicationNodeMetrics() from peer metrics groups (4 hours ago)
• 9ef132c - remove excessive logging due to runtime.debugStack (5 hours ago)

Bugfix release

Highlights

• Fixes an important racy behavior upon deadlines to avoid sharing state()
  when disks are hung or misbehaving in PR #18861
• Pre-Allocate memory by default for all deployments. this increases performance
  in almost all the scenarios, from small to large objects
  NOTE: users who have limited memory can set CI=true to reduce pre-allocated
  memory from 1GiB/2GiB to 256MiB.
• Fixes a WebSocket grid panic upon nil channel close.
• Introduces new API for ServerUpdate and ServiceRestart that provide a more
  distributed view of the entire process.
• Startup sequence for large setups now can restart < 30 seconds even on a busy setup.

What's Changed

• fix: remove policy mapping file when empty by @donatello in #18828
• grid: Return rejection reason by @klauspost in #18834
• introduce new ServiceV2 API to handle guided restarts by @harshavardhana in #18826
• add new drive I/O waiting/tokens metric by @harshavardhana in #18836
• improve server update behavior by re-using memory properly by @harshavardhana in #18831
• Distribution bucket metrics with site replication by @shtripat in #18841
• Add extra protection for grid reconnects by @klauspost in #18840
• compress binary while sending it to all the nodes by @harshavardhana in #18837
• avoid reload of 'format.json' over the network under normal conditions by @harshavardhana in #18842
• site replication: remove extraneous log for missing group by @poornas in #18785
• pre-allocate memory by default by @harshavardhana in #18846
• Helm: add cilium networkpolicy by @QuantumEnigmaa in #18650
• feat(chart): add support to set the display name of OpenID provider by @lentidas in #18781
• Helm: Add apiVersion and kind to the StatefulSets volumeClaimTemplates by @ItsKev in #18770
• Correct a mistake in the value.yaml of minio helm chart by @Albert-W in #18611
• add necessary protection err, fileInfo slice reads and writes by @harshavardhana in #18854
• Fix startup message when using single path by @fwessels in #18856
• Fix all racy use of NewDeadlineWorker by @klauspost in #18861
• Add separate grid reconnection mutex by @klauspost in #18862
• fix replication error logs to include target endpoint by @poornas in #18863
• fix: Re-use TCP connections for Kafka dials by @Praveenrajmani in #18860
• migrate bootstrap logic directly to websockets by @harshavardhana in #18855
• further bootstrap/startup optimization for reading 'format.json' by @harshavardhana in #18868
• Fix inspect format.json exclusion by @klauspost in #18871
• allow bucket import to be idempotent by @harshavardhana in #18873
• remove unnecessary logs in HealBucket() by @harshavardhana in #18875
• add new update v2 that updates per node, allows idempotent behavior by @harshavardhana in #18859
• avoid 'panic' on mc admin update for single drive setup by @harshavardhana in #18876
• Update list.md by @cesnietor in #18878
• fix: regression in internode bytes counting by @harshavardhana in #18880
• update() stale rebalance stats() object during pool expansion by @harshavardhana in #18882
• add missing proper error return in WalkDir() by @harshavardhana in #18884
• grid: Simpler reconnect logic by @klauspost in #18889
• avoid close 'nil' panics if any by @harshavardhana in #18890
• remove local disk metrics from cluster metrics by @harshavardhana in #18886
• fix: incomingBytes() to update via handleMessages() by @harshavardhana in #18891

New Contributors

• @QuantumEnigmaa made their first contribution in #18650
• @lentidas made their first contribution in #18781
• @ItsKev made their first contribution in #18770
• @Albert-W made their first contribution in #18611

Full Changelog: RELEASE.2024-01-18T22-51-28Z...RELEASE.2024-01-28T22-35-53Z

Bugfix release

What's Changed

• reject reference format from a different deployment by @harshavardhana in #18800
• fix: a typo in storeDataUsageInBackend() comment by @chienguo in #18778
• Deallocate huge read buffers by @klauspost in #18813
• fix: HealBucket regression for empty buckets, simplify it by @harshavardhana in #18815
• Fix mini typo by @fwessels in #18816
• remove all older unused APIs by @harshavardhana in #18769
• add codespell action by @harshavardhana in #18818
• Do not log volume not empty when healing dangling buckets by @vadmeste in #18822
• capture I/O in waiting and total tokens in diskMetrics by @harshavardhana in #18819
• Use +Inf label additionally for Histogram metrics by @shtripat in #18807

New Contributors

• @chienguo made their first contribution in #18778

Full Changelog: RELEASE.2024-01-16T16-07-38Z...RELEASE.2024-01-18T22-51-28Z

Bugfix release

What's Changed

• fix: listing SSE encrypted multipart objects by @harshavardhana in #18786
• adding a missing return case to fix GetObjectTagging by @zveinn in #18793
• xl-meta: Clean output by @klauspost in #18794

Full Changelog: RELEASE.2024-01-13T07-53-03Z...RELEASE.2024-01-16T16-07-38Z

Bugfix release

What's Changed

• Verify that remote target bucket is on MinIO server for bucket replication by @taran-p in #18656
• avoid disk monitoring leaks under various conditions by @harshavardhana in #18777
• xl: Remove wrong wording for errCorruptedFormat by @vadmeste in #18775
• Add more size intervals to obj size histogram by @krisis in #18772
• support proxying of tagging requests in replication by @poornas in #18649
• treat all localhost endpoints as local setup with same port by @harshavardhana in #18784

Full Changelog: RELEASE.2024-01-11T07-46-16Z...RELEASE.2024-01-13T07-53-03Z

Bugfix release

Highlights

• A possible CPU usage improvement with many nested folders when doing recursive
  listing moving to readdir() to be attempted via fd instead of using poll.FD, which
  seems to cause a good amount of CPU build-up.

• Performance improvements on bucket only heal on large dense setups, where bucket
  is now healed via each nodes providing 10x to 100x reduction in number of network calls
  depending on number of drives per node.

What's Changed

• remove outdated notes by @jiuker in #18755
• fix: prevent queries from starting before initialization by @jiuker in #18756
• Add line, col to types used in batch-expire by @krisis in #18747
• fix: s3 sql parse error for colums as with quotes by @jiuker in #18765
• avoid frivolous logs for expired credentials by @harshavardhana in #18767
• Node level bucket heal by @shtripat in #18612
• optimize readdir() open calls to be dealt with directly via 'fd' by @harshavardhana in #18762
• fix: prevent queries from starting before initialization by @jiuker in #18766
• Workflow to add issues to Tracking project by @dvaldivia in #18771
• update all dependencies and console v0.45.0 by @harshavardhana in #18773

Full Changelog: RELEASE.2024-01-05T22-17-24Z...RELEASE.2024-01-11T07-46-16Z

Bugfix release

What's Changed

• Added list of scanner metrics to document by @shtripat in #18731
• scanner: Allow full throttle if there is no parallel disk ops by @vadmeste in #18109
• fix: an odd crash when deleting null DEL markers by @harshavardhana in #18727
• prom: Add read quorum per erasure set metric by @vadmeste in #18736
• scanner: Add a config to disable short sleep between objects scan by @vadmeste in #18734
• NEW API: GetObjectAttributes by @zveinn in #18732
• simplify success response for GetObjectAttributes() by @harshavardhana in #18746
• audit: Add the count of drives from where the dangling object is removed by @vadmeste in #18737

Full Changelog: RELEASE.2024-01-01T16-36-33Z...RELEASE.2024-01-05T22-17-24Z

Feature release

Highlights

• A significant performance improvement feature to optimize ListObjects() is part of this release.
  MinIO manages some form of drive-level information about the validity of the drives, which
  It optimizes the number of disks used in a List() call per erasure set. This dramatically
  improves ListObjects() operations per second on a dense cluster. Refer to PR #18084
  for more information. This is not enabled by default but will eventually become the
  default in future releases.

• New console HTTP security headers are fully customizable now for specific needs, refer #18631

    * csp_policy
    * hsts_seconds
    * hsts_include_subdomains
    * hsts_preload
    * referrer_policy

What's Changed

• upgrade Go mod dependencies by @harshavardhana in #18710
• Upgrade Console to v0.44.0 by @dvaldivia in #18717
• clarify maximum number of buckets by @harshavardhana in #18718
• Adding ~ to match job when multiple jobs by @marioanton in #18706
• audit: Fix merrs and derrs object dangling message by @vadmeste in #18714
• re-use globalLocalDrives properly by @harshavardhana in #18721
• feat: introduce list_quorum="auto" to prefer quorum drives by @harshavardhana in #18084
• Make enableSha256 configurable in mc perf test by @moting9 in #18712
• [doc] Fix the name of RoleArn param in AssumeRoleWithWebIdentity requests by @tsuzu in #18711
• MinIO Console security headers by @pjuarezd in #18631

New Contributors

• @marioanton made their first contribution in #18706
• @moting9 made their first contribution in #18712
• @tsuzu made their first contribution in #18711
• @pjuarezd made their first contribution in #18631

Full Changelog: RELEASE.2023-12-23T07-19-11Z...RELEASE.2024-01-01T16-36-33Z

Bugfix release

What's Changed

• Export tier metrics by @krisis in #18678
• allow pre-allocating buffers to reduce frequent GCs during growth by @harshavardhana in #18686
• prom: Add online and healing drives metrics per erasure set by @vadmeste in #18700
• send proper IPv6 names avoid bracketing notation by @harshavardhana in #18699
• rename object_size -> block_size for cache subsystem by @harshavardhana in #18694
• tier: Allow edit of the new Azure and AWS auth params by @vadmeste in #18690
• fix: reject clients that do not send proper payload by @harshavardhana in #18701
• Graph cluster wide where applicable by @shtripat in #18705
• server-info: Avoid initializing audit/log http/kafka targets by @vadmeste in #18703
• Fix precendence bug in S3Select SQL IN clauses by @donatello in #18708

Full Changelog: RELEASE.2023-12-20T01-00-02Z...RELEASE.2023-12-23T07-19-11Z