Commit Graph

6574 Commits

Author SHA1 Message Date
Harshavardhana
fc5213258e
posix: Do not take disk offline on I/O errors (#8836)
Choosing maxAllowedIOError is arbitrary and
prone to errors, when drives might be perfectly
capable of taking I/O with only few locations
return I/O error. This is a hindrance of sort
where backend filesystems like ZFS can automatically
fix and handle these scenarios.

The added problem with current approach that we
take the drive offline, making it virtually impossible
to bring it online without restart the server which
is not desirable on a busy cluster. Remove this state
such that let the backend return error appropriately
to caller and let the caller decide what to do with
the error.
2020-01-17 13:34:43 -08:00
kannappanr
005ebbb9b2 Typo: Add missing '<' character (#8845)
Fixes #8833
2020-01-17 13:31:46 -08:00
Anis Elleuch
017067e11f data-usage: Avoid crawling duplicated call (#8843)
This fix will also picks 3 and not 4 disks from a single erasure set.
2020-01-17 09:59:37 -08:00
Harshavardhana
2bb69033e5 http: fail appropriately and return standard Go error (#8837)
return http.ErrServerClosed with proper body when
server is shutting down, allowing more context instead
of just returning '503' which doesn't mean the same
thing.
2020-01-17 05:48:39 -08:00
Harshavardhana
fca4ee84c9
gw/hdfs: listing should list directories properly (#8827)
Fixes #8822
2020-01-16 17:11:25 -08:00
poornas
60e60f68dd Add support for object locking with legal hold. (#8634) 2020-01-16 15:41:56 -08:00
Minio Trusted
ba758361b3 Update yaml files to latest version RELEASE.2020-01-16T22-40-29Z 2020-01-16 22:47:41 +00:00
Harshavardhana
c6b218e5df
fix: readiness should return 200 OK with first zone online (#8834) 2020-01-16 13:49:25 -08:00
Anis Elleuch
c18fbdb29a posix: Remove a non needed nil check in DiskInfo() (#8830)
posix.DiskInfo() returns errFaultyDisk when posix is nil,
but there is no way that this would happen any time, therefore
removing un-needed code.
2020-01-16 11:27:50 -08:00
Harshavardhana
b1ad99edbf
fix: avoid crash copy map before reading (#8825)
code of this form is always racy, when the
map itself is being written to as well

```
func (r Map) retMap() map[string]string {
     .. lock ..
     return r.internalMap
}

func (r Map) addMap(k, v string) {
     .. lock ..
     r.internalMap[k] = v
}
```

Anyone reading from `retMap()` is not protected
because of locking and we need to make sure
to avoid code in this manner. Always safe to
copy the map and return.
2020-01-16 01:35:30 -08:00
Minio Trusted
080e0c2323 Update yaml files to latest version RELEASE.2020-01-16T03-05-44Z 2020-01-16 03:11:40 +00:00
Anis Elleuch
935546d5ca xl: Implement MRF healing (#8470) 2020-01-15 18:30:32 -08:00
Harshavardhana
64fde1ab95
xl/zones: return errNoHealRequired when no heal is required (#8821)
Zone abstraction of object layer was returning `nil`
incorrectly under situations where disk healing is
not required. Returning `nil` is considered as healing
successful, which leads to unexpected ReloadFormat()
peer notification calls during startup.

This PR fixes this behavior properly for zones.
2020-01-15 17:19:13 -08:00
Anis Elleuch
169e8742fc mint: Use https for access to maven reposity (#8824)
Maven repository requires HTTPS now. This lead to issues
building mint image in aws-sdk-java & minio-java.

The PR fixes the issue and also bump aws sdk version in
aws-sdk-java to the latest.
2020-01-15 15:48:49 -08:00
Anis Elleuch
069876e262 xl: All nodes create meta volumes in its local disks (#8786)
Meta volumes directories, tmp/, background-ops/, etc..
undr .minio.sys are created when disks are formatted
but also when the cluster is started.

However using MakeVolBulk() is not appropriate in the
case of a user migrating from a version which does not
have .minio.sys/background-ops/. The reason is that
MakeVolBulk() exits early when an error is occured:
errVolumeExists in this case, which is expected since
some directories such as tmp/ already exist.

This commit will avoid use MakeVolBulk and use MakeVol
instead.

Also the PR will make each node creates meta volumes
in its local disks and stop relying on the first disk
since the first node could be offline.
2020-01-15 12:36:52 -08:00
Harshavardhana
442e1698cb
heal: Avoid spinning up object healing during startup (#8819)
auto-heal disks, metadata and buckets in background but
not objects, let the auto heal kick in for objects after
the cluster has been up for a while.
2020-01-15 01:08:39 -08:00
poornas
d76518eeb9 Remove TestPutObjectPartDiskNotFound unit test (#8815) 2020-01-14 18:46:33 -08:00
Harshavardhana
0879a4f743 rest/storage: Remove racy LastError usage (#8817)
instead perform a liveness check call to
verify if server is online and print relevant
errors.

Also introduce a StorageErr string error type
instead of errors.New() deprecate usage of
VerifyFileError, DeleteFileError for gob,
change in datastructure also requires bump in
storage REST version to v13.

Fixes #8811
2020-01-14 18:45:17 -08:00
Harshavardhana
9be7066715
fix: Hold locks before closing all drives (#8818)
Fixes #8813
2020-01-14 17:13:58 -08:00
Klaus Post
d8660b30cc Reduce MemProfileRate (#8814)
Enabling the memory profiling has a significant impact on performance.

Reduce the profiling rate by 2 orders of magnitude. It is still 128x smaller than default so it should be plenty.
2020-01-14 16:18:45 -08:00
poornas
30922148fb Fix bug preventing overwrite of object if (#8796)
object lock config is enabled for a bucket.

Creating a bucket with object lock configuration
enabled does not automatically cause WORM protection
to be applied. PUT operation needs to specifically
request object locking or bucket has to have default
retention settings configured.

Fixes regression introduced in #8657
2020-01-13 17:29:31 -08:00
Klaus Post
37b32199e3 Validate XL sets on format (#8779)
When formatting a set validate if a host failure will likely lead to data loss.

While we don't know what config will be set in the future 
evaluate to our best knowledge, assuming default settings.
2020-01-13 13:09:10 -08:00
ebozduman
d74818b227 Typo in MySQL notification doc (#8807) 2020-01-13 13:06:42 -08:00
Klaus Post
627fdfeab7 Fix Windows console printing (#8805)
Print to console which does translation and not directly to stdout.

Fixes #8804
2020-01-13 13:05:51 -08:00
Harshavardhana
3320878dfb Add missing SSE_MASTER_KEY_FILE entry (#8800)
Fixes #8757
2020-01-13 12:31:18 +05:30
Harshavardhana
3f20011862 docs: update how to query the keys and args (#8797)
remove link to missing config.json.sample
2020-01-11 15:13:13 -08:00
Etienne
5f34b5e6a5 Update README.md (#8788) 2020-01-10 21:08:09 -08:00
poornas
9199033db7 Set X-Cache and X-Cache-Lookup headers for cache (#8794)
X-Cache sets cache status of HIT if object is
served from the disk cache, or MISS otherwise.
X-Cache-Lookup is set to HIT if object was found
in the cache even if not served (for e.g. if cache
 entry was invalidated by ETag verification)
2020-01-10 20:21:13 -08:00
Klaus Post
2bf6cf0e15 Enable multiple concurrent profile types (#8792) 2020-01-10 17:19:58 -08:00
Harshavardhana
686d4656de
fix: set appropriate defaults when new keys added (#8795)
A new key was added in identity_openid recently
required explicitly for client to set the optional
value without that it would be empty, handle this
appropriately.

Fixes #8787
2020-01-10 16:57:18 -08:00
Harshavardhana
5aa5dcdc6d
lock: improve locker initialization at init (#8776)
Use reference format to initialize lockers
during startup, also handle `nil` for NetLocker
in dsync and remove *errorLocker* implementation

Add further tuning parameters such as

 - DialTimeout is now 15 seconds from 30 seconds
 - KeepAliveTimeout is not 20 seconds, 5 seconds
   more than default 15 seconds
 - ResponseHeaderTimeout to 10 seconds
 - ExpectContinueTimeout is reduced to 3 seconds
 - DualStack is enabled by default remove setting
   it to `true`
 - Reduce IdleConnTimeout to 30 seconds from
   1 minute to avoid idleConn build up

Fixes #8773
2020-01-10 02:35:06 -08:00
Harshavardhana
0a70bc24ac
Disallow only policy statements which are exactly same (#8785) 2020-01-09 19:29:57 -08:00
Kevin Humphreys
656146b699 doc: Prometheus metrics name fix (#8774)
changed docs to reflect proper Prometheus metrics
2020-01-09 18:36:58 -08:00
Harshavardhana
5e40b9a563
fix: docs for live/ready check implementation details 2020-01-09 18:29:24 -08:00
Joe Adams
89d1221217 Fix typo in prometheus monitoring docs (#8780) 2020-01-09 09:08:41 -08:00
Praveen raj Mani
4cd1bbb50a This PR fixes two things (#8772)
- Stop spawning store replay routines when testing the notification targets
- Properly honor the target.Close() to clean the resources used

Fixes #8707

Co-authored-by: Harshavardhana <harsha@minio.io>
2020-01-09 19:45:44 +05:30
Harshavardhana
c2cde6beb5
policy: Allow duplicate statements with different effects (#8775)
This allows "Allow" and "Deny" conflicting statements,
where we evaluate to implicit "Deny".
2020-01-08 23:00:54 -08:00
Harshavardhana
abc1c1070a Add custom policy claim name (#8764)
In certain organizations policy claim names
can be not just 'policy' but also things like
'roles', the value of this field might also
be *string* or *[]string* support this as well

In this PR we are still not supporting multiple
policies per STS account which will require a
more comprehensive change.
2020-01-08 17:21:58 -08:00
poornas
fd56aa42a6 Fix error message wording for PutObjectLockConfig (#8759)
Co-Authored-By: kannappanr <30541348+kannappanr@users.noreply.github.com>
Co-authored-by: Harshavardhana <harsha@minio.io>
2020-01-08 15:36:23 -08:00
Klaus Post
3d318bae76 init: Use constant time retries (#8769)
Exponential backoff does not seem like a good fit for
this function since we can expect a few roundtrips on
initial startup.

This retry loop get slow pretty quickly with initial
wait being 1 second and each try being double the
wait until 30 seconds is reached.

Instead simply try 2 times per second.
2020-01-08 13:37:34 -08:00
Harshavardhana
aa2e89bfe3 Use jsoniter whenever applicable instead of encoding/json (#8766)
This PR adds jsoniter package to replace encoding/json
in places where faster json unmarshal is necessary
whenever input JSON is large enough.

Some benchmarking comparison between jsoniter and enconding/json

benchmark                            old MB/s     new MB/s     speedup
BenchmarkParseUnmarshal/N10-4        110.02       331.17       3.01x
BenchmarkParseUnmarshal/N100-4       125.74       524.09       4.17x
BenchmarkParseUnmarshal/N500-4       131.68       542.60       4.12x
BenchmarkParseUnmarshal/N1000-4      133.93       514.88       3.84x
BenchmarkParseUnmarshal/N5000-4      122.10       415.36       3.40x
BenchmarkParseUnmarshal/N10000-4     132.13       403.90       3.06x
2020-01-08 17:01:42 +05:30
Harshavardhana
60813bef29
Allow proper setCount SLAs across zones (#8752)
Fixes scenario where zones are appropriately
handled, along with supporting overriding set
count. The new fix also ensures that we handle
the various setup types properly.

Update documentation to properly indicate the
behavior.

Fixes #8750

Co-authored-by: Nitish Tiwari <nitish@minio.io>
2020-01-07 09:13:44 -08:00
Harshavardhana
b123be5612 fix: browser should listBuckets from etcd in global federation (#8760) 2020-01-07 09:03:00 +05:30
Harshavardhana
933c60bc3a Add crypto context errors (#8740)
Currently when connections to vault fail, client
perpetually retries this leads to assumptions that
the server has issues and masks the problem.

Re-purpose *crypto.Error* type to send appropriate
errors back to the client.
2020-01-06 16:15:22 -08:00
ebozduman
796cca4166 Creates zipped files with correct mod times for objects (#8693) 2020-01-06 12:43:00 -08:00
Klaus Post
fe379f9428 Copy metadata on update (#8755)
Fixes #8706

Co-authored-by: Harshavardhana <harsha@minio.io>
2020-01-06 10:15:44 -08:00
Harshavardhana
ae0b165431 fix: --anonymous flag shouldn't print any keys (#8753)
Fixes #8744
2020-01-06 22:12:47 +05:30
Harshavardhana
7a06e158f1
Add docs for sysctl tuning parameters (#8749) 2020-01-06 00:30:04 -08:00
Harshavardhana
5eab3db344 Fix doc reference for prometheus (#8748) 2020-01-05 13:44:39 -08:00
Brian Candler
9f44fcd540 Clarify behaviour of erasure coding sets (#8745) 2020-01-05 13:00:11 -08:00