Commit Graph

4545 Commits

Author SHA1 Message Date
Harshavardhana
32b2f6117e
fix: do not pass around sync.Map (#15250)
it is not safe to pass around sync.Map
through pointers, as it may be concurrently
updated by different callers.

this PR simplifies by avoiding sync.Map
altogether, we do not need sync.Map
to keep object->erasureMap association.

This PR fixes a crash when concurrently
using this value when audit logs are
configured.

```
fatal error: concurrent map iteration and map write

goroutine 247651580 [running]:
runtime.throw({0x277a6c1?, 0xc002381400?})
        runtime/panic.go:992 +0x71 fp=0xc004d29b20 sp=0xc004d29af0 pc=0x438671
runtime.mapiternext(0xc0d6e87f18?)
        runtime/map.go:871 +0x4eb fp=0xc004d29b90 sp=0xc004d29b20 pc=0x41002b
```
2022-07-07 17:04:25 -07:00
Harshavardhana
ae92521310
remove unnecessary nAgreed value in partial() func (#15242) 2022-07-07 13:45:34 -07:00
Harshavardhana
5802df4365
retry and resume decom operation upon retriable failures (#15244)
it is possible in a k8s-like system reading pool.bin
might not have quorum during startup, however, add
a way to retry after this failure.
2022-07-07 12:31:44 -07:00
Anis Elleuch
8d98282afd
Better reporting of total/free usable capacity of the cluster (#15230)
The current code uses approximation using a ratio. The approximation 
can skew if we have multiple pools with different disk capacities.

Replace the algorithm with a simpler one which counts data 
disks and ignore parity disks.
2022-07-06 13:29:49 -07:00
Harshavardhana
3af6073576
no 'replicate status' without replication config (#15233)
'replicate status' shouldn't be displaying historic
values unless replication config is present on the
relevant bucket.
2022-07-06 09:53:33 -07:00
Harshavardhana
2518af5f9e
fix: allow certain mutations on objects during decommissioning (#15231)
fix: allow certain mutation on objects during decommission

currently by mistake deletion of objects was skipped,
if the object resided on the pool being decommissioned.

delete's are okay to be allowed since decommission is
designed to run on a cluster with active I/O.
2022-07-06 09:53:16 -07:00
Harshavardhana
7b793d84c8
fix: calculate scanner metric paths for single drive (#15232)
Additionally use pathJoin() to avoid double `//`
in path names.
2022-07-06 07:48:38 -07:00
Aditya Manthramurthy
af9bc7ea7d
Add external IDP management Admin API for OpenID (#15152) 2022-07-05 18:18:04 -07:00
Klaus Post
ac055b09e9
Add detailed scanner metrics (#15161) 2022-07-05 14:45:49 -07:00
haslersn
df42914da6
Fix missing whitespace in error message for IncompleteBody (#15227) 2022-07-05 12:19:57 -07:00
Klaus Post
2471bdda00
fix: for DiskInfo call cache disk metrics (#15229)
Small uploads spend a significant amount of time (~5%) fetching disk info metrics. Also maps are allocated for each call.

Add a 100ms cache to disk metrics.
2022-07-05 11:02:30 -07:00
Harshavardhana
9d80ff5a05
fix: decommission delete markers for non-current objects (#15225)
versioned buckets were not creating the delete markers
present in the versioned stack of an object, this essentially
would stop decommission to succeed.

This PR fixes creating such delete markers properly during
a decommissioning process, adds tests as well.
2022-07-05 07:37:24 -07:00
Harshavardhana
b311abed31
decom IAM, Bucket metadata properly (#15220)
Current code incorrectly passed the
config asset object name while decommissioning,
make sure that we pass the right object name
to be hashed on the newer set of pools.

This PR fixes situations after a successful
decommission, the users and policies might go
missing due to wrong hashed set.
2022-07-04 14:02:54 -07:00
Harshavardhana
ce667ddae0
do not print errFileNotFound in entries.resolve() (#15216) 2022-07-04 06:40:46 -07:00
Harshavardhana
0fee993a4b
return appropriate error under 'decom status' (#15213)
fixes #15208
2022-07-01 16:21:23 -07:00
Poorna
0ea5c9d8e8
site healing: Skip stale iam asset updates from peer. (#15203)
Allow healing to apply IAM change only when peer
gave the most recent update.
2022-07-01 13:19:13 -07:00
Harshavardhana
63ac260bd5
Simplify Prometheus metrics gather (#15210) 2022-07-01 13:18:39 -07:00
Harshavardhana
f9a4ad7904
update banner with version+runtime (#15206) 2022-06-30 13:58:09 -07:00
Minio Trusted
e60b67d246 Revert "Tighten enforcement of object retention (#14993)"
This reverts commit 5e3010d455.

This commit causes regression on object locked buckets causine
delete-markers to be not created.
2022-06-30 13:06:32 -07:00
Klaus Post
9004d69c6f
Make ReqInfo concurrency safe (#15204)
Some read/writes of ReqInfo did not get appropriate locks, leading to races.

Make sure reading and writing holds appropriate locks.
2022-06-30 10:48:50 -07:00
Harshavardhana
8856a2d77b
finalize startup-banner and remove unnecessary logs (#15202) 2022-06-29 16:32:04 -07:00
Anis Elleuch
54a061bdda
Save minio version information centrally (#15181) 2022-06-29 14:45:49 -07:00
Poorna
7cc9286e0f
site healing: Skip stale bucket metadata updates from peer (#15186)
Allow healing to apply bucket metadata change only when peer
gave the most recent update.
2022-06-28 18:09:20 -07:00
Harshavardhana
2f25639ea0
update banner to reflect the final agreed UI (#15192) 2022-06-28 16:37:40 -07:00
Harshavardhana
2070c215a2
handle missing funcNames for handlers (#15188)
also use designated names for internal
calls

- storageREST calls are storageR
- lockREST calls are lockR
- peerREST calls are just peer

Named in this fashion to facilitate wildcard matches
by having prefixes of the same name.

Additionally, also enable funcNames for generic handlers
that return errors, currently we disable '<unknown>'
2022-06-28 05:04:10 -07:00
Harshavardhana
9c605ad153
allow support for parity '0', '1' enabling support for 2,3 drive setups (#15171)
allows for further granular setups

- 2 drives (1 parity, 1 data)
- 3 drives (1 parity, 2 data)

Bonus: allows '0' parity as well.
2022-06-27 20:22:18 -07:00
Anis Elleuch
b7c7e59dac
Revert proxying requests with precondition errors (#15180)
In a replicated setup, when an object is updated in one cluster but
still waiting to be replicated to the other cluster, GET requests with
if-match, and range headers will likely fail. It is better to proxy
requests instead.

Also, this commit avoids printing verbose logs about precondition &
range errors.
2022-06-27 14:03:44 -07:00
Harshavardhana
699cf6ff45
perform object sweep after equeue the latest CopyObject() (#15183)
keep it similar to PutObject/CompleteMultipart
2022-06-27 12:11:33 -07:00
Anis Elleuch
9201870f6c
Remove unnecessary code in WalkDir() (#15168)
Recalculating forward is useless. It is never used and it will be
computed again when calling scanDir() again.
2022-06-27 10:26:56 -07:00
Harshavardhana
6722f58668
save MinIO version with each version (8-bytes extra) (#15170)
store MinIO version along with each version in 'xl.meta'
for future purposes, can be used as ways to add specific
code for bug fixes if any.
2022-06-27 03:59:41 -07:00
Harshavardhana
7b9b7cef11
add license banner for GNU AGPLv3 (#15178)
Bonus: rewrite subnet re-use of Transport
2022-06-27 03:58:25 -07:00
Harshavardhana
bd099f5e71
fix: change timedValue to return the previously cached value (#15169)
fix: change timedvalue to return previous cached value

caller can interpret the underlying error and decide
accordingly, places where we do not interpret the
errors upon timedValue.Get() - we should simply use
the previously cached value instead of returning "empty".

Bonus: remove some unused code
2022-06-25 08:50:16 -07:00
Klaus Post
baf257adcb
fix: health client leak when calling UpdateAllTargets (#15167)
When `LoadBucketMetadataHandler` is called and `UpdateAllTargets` gets called.

Since targets are rebuilt we cancel all.
2022-06-24 11:12:52 -07:00
Anis Elleuch
4fd1986885
Trace all http requests (#15064)
Add a generic handler that adds a new tracing context to the request if
tracing is enabled. Other handlers are free to modify the tracing
context to update information on the fly, such as, func name, enable
body logging etc..

With this commit, requests like this 

```
curl -H "Host: ::1:3000" http://localhost:9000/
```

will be traced as well.
2022-06-23 23:19:24 -07:00
Harshavardhana
e1afac9439
reduce sha256 CPU usage by turning it off for speedtests (#15154)
continuation of the PR #15151, keeping signature v4 for
the headers however avoiding sha256 for the body.
2022-06-23 11:26:53 -07:00
Poorna
580d9db85e
Add APIs to import/export IAM data (#15014) 2022-06-23 09:25:15 -07:00
Anis Elleuch
42e2fd35d8
heal: Include dir markers when healing a fresh disk (#15158)
Directories markers are not healed when healing a new fresh disk. A
a proper fix would be moving object names encoding/decoding to erasure
object level but it is too late now since the object to set distribution is
calculated at a higher level.
2022-06-23 06:47:33 -07:00
Harshavardhana
1a40c7c27c
use signature-v2 for 'object perf' tests to avoid CPU using sha256 (#15151)
It is observed in a local 8 drive system the CPU seems to be
bottlenecked at

```
(pprof) top
Showing nodes accounting for 1385.31s, 88.47% of 1565.88s total
Dropped 1304 nodes (cum <= 7.83s)
Showing top 10 nodes out of 159
      flat  flat%   sum%        cum   cum%
      724s 46.24% 46.24%       724s 46.24%  crypto/sha256.block
   219.04s 13.99% 60.22%    226.63s 14.47%  syscall.Syscall
   158.04s 10.09% 70.32%    158.04s 10.09%  runtime.memmove
   127.58s  8.15% 78.46%    127.58s  8.15%  crypto/md5.block
    58.67s  3.75% 82.21%     58.67s  3.75%  github.com/minio/highwayhash.updateAVX2
    40.07s  2.56% 84.77%     40.07s  2.56%  runtime.epollwait
    33.76s  2.16% 86.93%     33.76s  2.16%  github.com/klauspost/reedsolomon._galMulAVX512Parallel84
     8.88s  0.57% 87.49%     11.56s  0.74%  runtime.step
     7.84s   0.5% 87.99%      7.84s   0.5%  runtime.memclrNoHeapPointers
     7.43s  0.47% 88.47%     22.18s  1.42%  runtime.pcvalue
```

Bonus changes:

- re-use transport for bucket replication clients, also site replication clients.
- use 32KiB buffer for all read and writes at transport layer seems to help
  TLS read connections.
- Do not have 'MaxConnsPerHost' this is problematic to be used with net/http
  connection pooling 'MaxIdleConnsPerHost' is enough.
2022-06-22 16:28:25 -07:00
Poorna
cb097e6b0a
CopyObject: fix read/write err on closed pipe (#15135)
Fixes: #15128
Regression from PR#14971
2022-06-21 19:20:11 -07:00
Poorna
1cfb03fb74
replication: Avoid proxying when precondition failed (#15134)
Proxying is not required when content is on this cluster and
does not meet pre-conditions specified in the request.

Fixes #15124
2022-06-21 14:11:35 -07:00
Harshavardhana
f293df647c
s3/zip: extract metadata properly for Zipped objects (#15123)
s3/zip: extra metadata properly for Zipped objects

fixes #15121
2022-06-21 14:11:12 -07:00
sota
e2e5bd6f19
fix: cant parse comment without '=' in environment file (#15130) 2022-06-21 10:37:15 -07:00
Andreas Auernhammer
cd7a0a9757
fips: simplify TLS configuration (#15127)
This commit simplifies the TLS configuration.
It inlines the FIPS / non-FIPS code.

Signed-off-by: Andreas Auernhammer <hi@aead.dev>
2022-06-21 07:54:48 -07:00
Anis Elleuch
b3eda248a3
Parallelize new disks healing of different erasure sets (#15112)
- Always reformat all disks when a new disk is detected, this will
  ensure new uploads to be written in new fresh disks
- Always heal all buckets first when an erasure set started to be healed
- Use a lock to prevent two disks belonging to different nodes but in
  the same erasure set to be healed in parallel
- Heal different sets in parallel

Bonus:
- Avoid logging errUnformattedDisk when a new fresh disk is inserted but
  not detected by healing mechanism yet (10 seconds lag)
2022-06-21 07:53:55 -07:00
Harshavardhana
486888f595
remove gateway banner and some other TODO loggers (#15125) 2022-06-21 05:25:40 -07:00
Poorna
b3ebc69034
improve error message for bucket metadata export/import API (#15120) 2022-06-20 16:13:45 -07:00
Harshavardhana
761dde2f1b
fix: add 'mc support inspect' support for single drive deployment (#15122) 2022-06-20 16:11:19 -07:00
Harshavardhana
2bb6a3f4d0
cleanup site replication error handling (#15113)
site replication errors were printed at
various random locations, repeatedly - this
PR attempts to remove double logging and
capture all of them at a common place.

This PR also enhances the code to show
partial success and errors as well.
2022-06-20 10:48:11 -07:00
Anis Elleuch
73733a8fb9
heal: Report correctly in multip-pools setup (#15117)
`mc admin heal -r <alias>` in a multi setup pools returns incorrectly
grey objects. The reason is that erasure-server-pools.HealObject() runs
HealObject in all pools and returns the result of the first nil
error. However, in the lower erasureObject level, HealObject() returns
nil if an object does not exist + missing error in each disk of the object
in that pool, therefore confusing mc.

Make erasureObject.HealObject() to return not found error in the lower
level, so at least erasureServerPools will know what pools to ignore.
2022-06-20 08:07:45 -07:00
Poorna
2fa1d8ac48
Add import/export APIs to migrate bucket metadata (#14929) 2022-06-18 06:55:39 -07:00