Commit Graph

12179 Commits

Author SHA1 Message Date
Krishnan Parthasarathi
4a1edfd9aa
Different read quorum for tiered objects (#20115)
For a non-tiered object, MinIO requires that EcM (# of data blocks) of
xl.meta agree, corresponding to the number of data blocks needed to 
read this object.

OTOH, tiered objects have metadata in the hot tier and data in the 
warm tier. The data and its integrity are offloaded to the warm tier. This
allows us to reduce the read quorum from EcM (typically > N/2, where N -
erasure stripe width) to N/2 + 1. The simple majority of metadata
ensures consensus on what the object is and where it is
located.
2024-07-25 14:02:50 -07:00
Anis Eleuch
b7f319b62a
properly reload a fresh drive when found in a failed state during startup (#20145)
When a drive is in a failed state when a single node multiple drives
deployment is started, a replacement of a fresh disk will not be
properly healed unless the user restarts the node.

Fix this by always adding the new fresh disk to globalLocalDrivesMap. Also
remove globalLocalDrives for simplification, a map to store local node
drives can still be used since the order of local drives of a node is
not defined.
2024-07-24 16:30:33 -07:00
Anis Eleuch
33c101544d
kms: Expose API when bucket federation is enabled (#20143)
kms: Expose API available when bucket federation is enabled

When bucket federation feature is enabled, KMS API will not work, such
as `mc admin kms key list`

The commit will fix the issue by disabling bucket forwarding when this
is a KMS request.
2024-07-24 15:44:29 -07:00
Anis Eleuch
21cf29330e
grafana: Fix the unit in Open FDs panel (#20144) 2024-07-24 07:51:03 -07:00
Harshavardhana
3b21bb5be8
use unixNanoTime instead of time.Time in lockRequestorInfo (#20140)
Bonus: Skip Source, Quorum fields in lockArgs that are never
sent during Unlock() phase.
2024-07-24 03:24:01 -07:00
Harshavardhana
6fe2b3f901
avoid sendFile() for ranges or object lengths < 4MiB (#20141) 2024-07-24 03:22:50 -07:00
Taran Pelkey
b368d4cc13
Fix updateGroupMembershipsForLDAP behavior with unicode (#20137) 2024-07-23 19:10:03 -07:00
Klaus Post
0680af7414
Set O_NONBLOCK for reads and writes on unix (#20133)
Tracing syscalls, opening and reading an `xl.meta` looks like this:

```
openat(AT_FDCWD, "/mnt/drive1/ss8-old/testbucket/ObjSize4MiBThreads72/(554O51H/peTb(0iztdbTKw59.csv/xl.meta", O_RDONLY|O_NOATIME|O_CLOEXEC) = 34 <0.000>
fcntl(34, F_GETFL)                      = 0x48000 (flags O_RDONLY|O_LARGEFILE|O_NOATIME) <0.000>
fcntl(34, F_SETFL, O_RDONLY|O_NONBLOCK|O_LARGEFILE|O_NOATIME) = 0 <0.000>
epoll_ctl(4, EPOLL_CTL_ADD, 34, {events=EPOLLIN|EPOLLOUT|EPOLLRDHUP|EPOLLET, data={u32=3172471557, u64=8145488475984499461}}) = -1 EPERM (Operation not permitted) <0.000>
fcntl(34, F_GETFL)                      = 0x48800 (flags O_RDONLY|O_NONBLOCK|O_LARGEFILE|O_NOATIME) <0.000>
fcntl(34, F_SETFL, O_RDONLY|O_LARGEFILE|O_NOATIME) = 0 <0.000>
fstat(34, {st_mode=S_IFREG|0644, st_size=354, ...}) = 0 <0.000>
read(34, "XL2 \1\0\3\0\306\0\0\1P\2\2\1\304$\225\304\20\0\0\0\0\0\0\0\0\0\0\0"..., 354) = 354 <0.000>
close(34)                               = 0 <0.000>
```

Everything until `fstat` is the `os.Open` call.

Looking at the code: https://github.com/golang/go/blob/master/src/os/file_unix.go#L212-L243

It seems for every file it "tries" to see if it is pollable. This causes `syscall.SetNonblock(fd, true)` to be called. This is the first `F_SETFL`.

It then calls `f.pfd.Init("file", true)`. This will attempt to set it as pollable using `epoll_ctl`. This will always fail for files. It therefore calls `syscall.SetNonblock(fd, false)` resulting in the second `F_SETFL`.

If we set the `O_NONBLOCK` call on the initial open, we should avoid the 4 `fcntl` syscalls per file.

I don't see any way to avoid the `epoll_ctl` call, since kind is either `kindOpenFile` or `kindNonBlock`, so "pollable" will always be true. However avoiding 4 of 6 syscalls still seems worth it.

This should not have any effect, since files will end up with "nonblock" anyway.
2024-07-23 09:36:24 -07:00
Harshavardhana
91805bcab6
add optimizations to bring performance on unversioned READS (#20128)
allow non-inlined on disk to be inlined via
an unversioned ReadVersion() call, we only
need ReadXL() to resolve objects with multiple
versions only.

The choice of this block makes it to be dynamic
and chosen by the user via `mc admin config set`

Other bonus things

- Start measuring internode TTFB performance.
- Set TCP_NODELAY, TCP_CORK for low latency
2024-07-23 03:53:03 -07:00
Klaus Post
c0e2886e37
Tweak grid for less writes (#20129)
Use `runtime.Gosched()` if we have less than maxMergeMessages and the 
queue is empty.  Up maxMergeMessages to 50 to merge more messages into 
a single write.

Add length check for an early bailout on readAllInto when we know packet length.
2024-07-23 03:28:14 -07:00
Andreas Auernhammer
4f5dded4d4
fips: enforce FIPS-compliant TLS ciphers in FIPS mode (#20131)
This commit enforces FIPS-compliant TLS ciphers in FIPS mode
by importing the `fipsonly` module.

Otherwise, MinIO still accepts non-FIPS compliant TLS connections.
2024-07-23 03:11:25 -07:00
jiuker
b3a94c4e85
fix: Use xtime duration to parse batch job (#20117) 2024-07-23 00:05:53 -07:00
Harshavardhana
8e618d45fc
remove unnecessary LRU for internode auth token (#20119)
removes contentious usage of mutexes in LRU, which
were never really reused in any manner; we do not
need it.

To trust hosts, the correct way is TLS certs; this PR completely
removes this dependency, which has never been useful.

```
0  0%  100%  25.83s 26.76%  github.com/hashicorp/golang-lru/v2/expirable.(*LRU[...])
0  0%  100%  28.03s 29.04%  github.com/hashicorp/golang-lru/v2/expirable.(*LRU[...])
```

Bonus: use `x-minio-time` as a nanosecond to avoid unnecessary
parsing logic of time strings instead of using a more
straightforward mechanism.
2024-07-22 00:04:48 -07:00
Harshavardhana
3ef59d2821
do not set KMSSecretKey env from KMSSecretKeyFile (#20122)
fixes #20121
2024-07-21 14:39:15 -07:00
Harshavardhana
23db4958f5
fix tuned-adm command typo 2024-07-18 18:15:02 -07:00
Anis Eleuch
d9ee668b6d
s3: Fix wrong continuation token during listing with ILM enabled bucket (#20113) 2024-07-18 13:37:34 -07:00
Anis Eleuch
2e5d792f0c
batch-expiry: Save progress regularly in the drives and at the end (#20098)
- Also, fix failure reporting at the end.
- Also, avoid parsing report objects when listing or resuming jobs, this
does not cause any bugs, it is only printing, not useful errors.
2024-07-17 09:42:32 -07:00
Minio Trusted
b276651eaa Update yaml files to latest version RELEASE.2024-07-16T23-46-41Z 2024-07-17 15:26:12 +00:00
Poorna
3535197f99
replication: proxy only on missing object or read quorum err (#20101) 2024-07-16 16:46:41 -07:00
Frank Wessels
95f076340a
Update reedsolomon dependency with fix for Graviton4 processor (#20102) 2024-07-16 12:27:21 -07:00
Mark Theunissen
698bb93a46
Allow a KMS Action to specify keys in the Resources of a policy (#20079) 2024-07-16 07:03:03 -07:00
Minio Trusted
2584430141 Update yaml files to latest version RELEASE.2024-07-15T19-02-30Z 2024-07-15 22:10:04 +00:00
Klaus Post
ded373e600
Split handleMessages (cosmetic) (#20095)
Split the read and write sides of handleMessages into two separate functions

Cosmetic. The only non-copy-and-paste change is that `cancel(ErrDisconnected)` is moved 
into the defer on `readStream`.
2024-07-15 12:02:30 -07:00
Harshavardhana
e8c54c3d6c
add validation test for v3 metrics for all its endpoints (#20094)
add unit test for v3 metrics for all its exposed endpoints

Bonus:
  - support OpenMetrics encoding
  - adds boot time for prometheus
  - continueOnError is better to serve as
    much metrics as possible.
2024-07-15 09:28:02 -07:00
Shubhendu
f944a42886
Removed user and group details from logs (#20072)
Signed-off-by: Shubhendu Ram Tripathi <shubhendu@minio.io>
2024-07-14 11:12:07 -07:00
Harshavardhana
eff0ea43aa
fix: typo in BucketUsageMetrics group registration in v3 metrics (#20090)
```
curl http://localhost:9000/minio/metrics/v3/cluster/usage/buckets
```

Did not work as documented, due to the fact that there was a typo
in the bucket usage metrics registration group. This endpoint is
a cluster endpoint and does not require any `buckets` argument.
2024-07-14 11:11:42 -07:00
Minio Trusted
3b602bb532 Update yaml files to latest version RELEASE.2024-07-13T01-46-15Z 2024-07-13 02:08:28 +00:00
Alex
459985f0fa
Update Console to v1.6.3 (#20084)
Signed-off-by: Benjamin Perez <benjamin@bexsoft.net>
2024-07-12 18:46:15 -07:00
Harshavardhana
d0080046c2 allow sysfs tuning from tuned 2024-07-12 16:31:18 -07:00
Harshavardhana
7fcb428622
do not print unexpected logs (#20083) 2024-07-12 13:51:54 -07:00
dependabot[bot]
4ea6f94ed8
Bump github.com/nats-io/nats-streaming-server from 0.24.3 to 0.24.6 (#20082)
Bumps [github.com/nats-io/nats-streaming-server](https://github.com/nats-io/nats-streaming-server) from 0.24.3 to 0.24.6.
- [Release notes](https://github.com/nats-io/nats-streaming-server/releases)
- [Changelog](https://github.com/nats-io/nats-streaming-server/blob/main/.goreleaser.yml)
- [Commits](https://github.com/nats-io/nats-streaming-server/compare/v0.24.3...v0.24.6)

---
updated-dependencies:
- dependency-name: github.com/nats-io/nats-streaming-server
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2024-07-12 11:45:09 -07:00
Klaus Post
83adc2eebf
Fix ListObjects aborting after 3 minute on async request (#20074)
When creating the async listing, if the first request does not return within 3 
minutes, it is stopped, since it isn't being kept alive.

Keep updating `lastHandout` while we are waiting for the initial request to be fulfilled.
2024-07-12 09:23:16 -07:00
Poorna
989c318a28
replication: make large workers configurable (#20077)
This PR also improves throttling by reducing tokens requested
from rate limiter based on available tokens to avoid exceeding
throttle wait deadlines
2024-07-12 07:57:31 -07:00
Frank Wessels
ef802f2b2c
Updated dependencies for ARM SVE support (#20081) 2024-07-12 07:36:29 -07:00
Taran Pelkey
f5d2fbc84c
Add DecodeDN and QuickNormalizeDN functions to LDAP config (#20076) 2024-07-11 18:04:53 -07:00
Allan Roger Reid
e139673969
Audit failure in batch job key rotate (#20073) 2024-07-11 16:13:15 -07:00
Harshavardhana
a8c6465f22
hide some deprecated fields from 'get' output (#20069)
also update wording on `subnet license="" api_key=""`
2024-07-10 13:16:44 -07:00
Minio Trusted
27538e2d22 Update yaml files to latest version RELEASE.2024-07-10T18-41-49Z 2024-07-10 19:27:17 +00:00
Taran Pelkey
6c6f0987dc
Add groups to policy entities (#20052)
* Add groups to policy entities

* update comment

---------

Co-authored-by: Harshavardhana <harsha@minio.io>
2024-07-10 11:41:49 -07:00
Austin Chang
5f64658faa
clarify error message for root user credential (#20043)
Signed-off-by: Austin Chang <austin880625@gmail.com>
2024-07-10 09:57:01 -07:00
Anis Eleuch
ce183cb2b4
heal: List and heal again for any listing error (#19999)
When a fresh drive healing is finished, add more checks for the drive listing
errors. If any, re-list and heal again. Although this is an infrequent use
case to have listPathRaw() returning nil when minDisks is set to 1, we
still need to handle all possible use cases to avoid missing healing
any object.

Also, check for HealObject result to decide of an object is healed in the
fresh disk since HealObject returns nil if an object is healed in any
disk, and not in the new fresh drive.
2024-07-10 09:55:36 -07:00
Klaus Post
b3bac73c0f
Clarify post policy error message (#20067)
It is not really clear that the listed keys are missing.

Clarify the error
2024-07-10 07:18:44 -07:00
Anis Eleuch
e726d8ff0f
list: Hide objects/versions with pending/failed replicated deletion (#20047)
In regular listing, this commit will avoid showing an object when its
latest version has a pending or failed deletion. In replicated setup.
It will also prevent showing older versions in the same case.
2024-07-09 15:26:42 -07:00
Shubhendu
f4230777b3
Log replication errors once (#20063)
Also, sort the error map for multiple sites in ascending order
of deployment IDs, so that the error message generated is always
definitive order and same.

Signed-off-by: Shubhendu Ram Tripathi <shubhendu@minio.io>
2024-07-09 10:10:31 -07:00
Krishnan Parthasarathi
380233d646
batch: Update job info object on success (#20053) 2024-07-08 18:45:54 -07:00
Allan Roger Reid
d592bc0c1c
Fix documentation for removal of delete markers ILM rule (#20056) 2024-07-08 18:45:38 -07:00
Klaus Post
0d0b0aa599
Abstract grid connections (#20038)
Add `ConnDialer` to abstract connection creation.

- `IncomingConn(ctx context.Context, conn net.Conn)` is provided as an entry point for 
   incoming custom connections.

- `ConnectWS` is provided to create web socket connections.
2024-07-08 14:44:00 -07:00
Anis Eleuch
b433bf14ba
Add typos check to Makefile (#20051) 2024-07-08 14:39:49 -07:00
Minio Trusted
cf371da346 Update yaml files to latest version RELEASE.2024-07-04T14-25-45Z 2024-07-04 14:58:08 +00:00
Klaus Post
107d951893
Log ILM failed object name (#20040)
Log so we know which object we are dealing with.

Log each object once.
2024-07-04 07:25:45 -07:00