Commit Graph

7503 Commits

Author SHA1 Message Date
Harshavardhana
267d7bf0a9 Revert "Add delete marker replication support (#10396)"
This reverts commit 50c10a5087.

PR is moved to origin/dev branch
2020-11-12 11:43:14 -08:00
cksac
be83dfc52a
fix: HDFS list bucket when subpath is provided (#10884) 2020-11-12 11:26:51 -08:00
Harshavardhana
ca88ca753c
ignore typed errors correctly in list cache layer (#10879)
bonus write bucket metadata cache with enough quorum

Possible fix for #10868
2020-11-12 09:28:56 -08:00
Klaus Post
f86d3538f6
Allow deeper sleep (#10883)
Allow each crawler operation to sleep up to 10 seconds on very heavily loaded systems.

This will of course make minimum crawler speed less, but should be more effective at stopping.
2020-11-12 09:17:56 -08:00
Klaus Post
1c3590078d
Skip 0 byte stream writes (#10875)
Don't send a packet when receiving 0 bytes or there is an error recorded
2020-11-11 18:07:40 -08:00
Harshavardhana
aa158228f9
fix: simplify healing metadata objects per set (#10867) 2020-11-11 10:58:16 -08:00
Klaus Post
8747834c69
DeletedObjects: Return objects on lock failure (#10874)
Return objects when locking fails.

<details>
<summary>Panic</summary>

```
: 2020/11/10 04:15:55 http: panic serving 10.10.62.153:44858: runtime error: index out of range [0] with length 0
: goroutine 363537270 [running]:
: net/http.(*conn).serve.func1(0xc019232780)
:         net/http/server.go:1801 +0x147
: panic(0x1cadd60, 0xc001719260)
:         runtime/panic.go:975 +0x47a
: github.com/minio/minio/cmd.criticalErrorHandler.ServeHTTP.func1(0xc0121d1200, 0x210cda0, 0xc0141940e0)
:         github.com/minio/minio/cmd/generic-handlers.go:781 +0x1a8
: panic(0x1cadd60, 0xc001719260)
:         runtime/panic.go:969 +0x1b9
: github.com/minio/minio/cmd.objectAPIHandlers.DeleteMultipleObjectsHandler(0x1e71ce8, 0x1e71cc8, 0x2108420, 0xc0192328c0, 0xc0121d1400)
:         github.com/minio/minio/cmd/bucket-handlers.go:465 +0x2490
: net/http.HandlerFunc.ServeHTTP(...)
:         net/http/server.go:2042
: github.com/minio/minio/cmd.httpTraceAll.func1(0x2108420, 0xc0192328c0, 0xc0121d1400)
:         github.com/minio/minio/cmd/handler-utils.go:353 +0x158
: net/http.HandlerFunc.ServeHTTP(...)
:         net/http/server.go:2042
: github.com/minio/minio/cmd.collectAPIStats.func1(0x2108420, 0xc019232820, 0xc0121d1400)
:         github.com/minio/minio/cmd/handler-utils.go:380 +0xed
: net/http.HandlerFunc.ServeHTTP(...)
:         net/http/server.go:2042
: github.com/minio/minio/cmd.maxClients.func1(0x2108420, 0xc019232820, 0xc0121d1400)
:         github.com/minio/minio/cmd/handler-api.go:132 +0x33b
: net/http.HandlerFunc.ServeHTTP(0xc00271d590, 0x2108420, 0xc019232820, 0xc0121d1400)
:         net/http/server.go:2042 +0x44
: github.com/minio/minio/cmd.redirectHandler.ServeHTTP(0x20e2180, 0xc00271d590, 0x2108420, 0xc019232820, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:192 +0x156
: github.com/minio/minio/cmd.customHeaderHandler.ServeHTTP(0x20e1060, 0xc0141a22b0, 0x21083e0, 0xc01814d2e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:751 +0x162
: github.com/minio/minio/cmd.securityHeaderHandler.ServeHTTP(0x20e0fc0, 0xc0141a22c0, 0x21083e0, 0xc01814d2e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:766 +0x1d6
: github.com/minio/minio/cmd.bucketForwardingHandler.ServeHTTP(0xc0121c7a40, 0x20e1120, 0xc0141a22d0, 0x21083e0, 0xc01814d2e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:624 +0xbf
: github.com/minio/minio/cmd.requestValidityHandler.ServeHTTP(0x20e0f20, 0xc01814d280, 0x21083e0, 0xc01814d2e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:608 +0x42a
: github.com/minio/minio/cmd.httpStatsHandler.ServeHTTP(0x20e10c0, 0xc0141a2300, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:536 +0xe4
: github.com/minio/minio/cmd.requestSizeLimitHandler.ServeHTTP(0x20e0fe0, 0xc0141a2310, 0x50004000000, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:68 +0xd4
: github.com/minio/minio/cmd.requestHeaderSizeLimitHandler.ServeHTTP(0x20e10a0, 0xc01814d2a0, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:93 +0x1b7
: github.com/minio/minio/cmd.crossDomainPolicy.ServeHTTP(0x20e1080, 0xc0141a2320, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/crossdomain-xml-handler.go:51 +0x82
: github.com/minio/minio/cmd.browserRedirectHandler.ServeHTTP(0x20e0fa0, 0xc0141a2330, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:276 +0x68
: github.com/minio/minio/cmd.minioReservedBucketHandler.ServeHTTP(0x20e0f00, 0xc0141a2340, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:344 +0xb8
: github.com/minio/minio/cmd.cacheControlHandler.ServeHTTP(0x20e1020, 0xc0141a2350, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:303 +0x1ce
: github.com/minio/minio/cmd.timeValidityHandler.ServeHTTP(0x20e0f40, 0xc0141a2360, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:414 +0x3ca
: github.com/minio/minio/cmd.resourceHandler.ServeHTTP(0x20e1160, 0xc0141a2370, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:516 +0xab
: github.com/minio/minio/cmd.authHandler.ServeHTTP(0x20e1100, 0xc0141a2380, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/auth-handler.go:502 +0x2e7
: github.com/minio/minio/cmd.sseTLSHandler.ServeHTTP(0x20e0ee0, 0xc0141a2390, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:802 +0x79
: github.com/minio/minio/cmd.reservedMetadataHandler.ServeHTTP(0x20e1140, 0xc0141a23a0, 0x210cda0, 0xc0141940e0, 0xc0121d1400)
:         github.com/minio/minio/cmd/generic-handlers.go:139 +0x1b7
: github.com/gorilla/mux.(*Router).ServeHTTP(0xc00073fb00, 0x210cda0, 0xc0141940e0, 0xc0121d1200)
:         github.com/gorilla/mux@v1.8.0/mux.go:210 +0xd3
: github.com/rs/cors.(*Cors).Handler.func1(0x210cda0, 0xc0141940e0, 0xc0121d1200)
:         github.com/rs/cors@v1.7.0/cors.go:219 +0x1b9
: net/http.HandlerFunc.ServeHTTP(0xc0009aece0, 0x210cda0, 0xc0141940e0, 0xc0121d1200)
:         net/http/server.go:2042 +0x44
: github.com/minio/minio/cmd.criticalErrorHandler.ServeHTTP(0x20e2180, 0xc0009aece0, 0x210cda0, 0xc0141940e0, 0xc0121d1200)
:         github.com/minio/minio/cmd/generic-handlers.go:784 +0x85
: github.com/minio/minio/cmd/http.(*Server).Start.func1(0x210cda0, 0xc0141940e0, 0xc0121d1200)
:         github.com/minio/minio/cmd/http/server.go:101 +0x258
: net/http.HandlerFunc.ServeHTTP(0xc000dc4080, 0x210cda0, 0xc0141940e0, 0xc0121d1200)
:         net/http/server.go:2042 +0x44
: net/http.serverHandler.ServeHTTP(0xc000764c60, 0x210cda0, 0xc0141940e0, 0xc0121d1200)
:         net/http/server.go:2843 +0xa3
: net/http.(*conn).serve(0xc019232780, 0x2114720, 0xc03381f6c0)
:         net/http/server.go:1925 +0x8ad
: created by net/http.(*Server).Serve
:         net/http/server.go:2969 +0x36c
```
</details>
2020-11-11 09:14:32 -08:00
Anton Melser
2c1e37197b
fix: bad example json for policy in replication docs (#10869) 2020-11-10 17:49:49 -08:00
Poorna Krishnamoorthy
50c10a5087
Add delete marker replication support (#10396)
Delete marker replication is implemented for V2
configuration specified in AWS spec (though AWS
allows it only in the V1 configuration).

This PR also brings in a MinIO only extension of
replicating permanent deletes, i.e. deletes specifying
version id are replicated to target cluster.
2020-11-10 15:24:14 -08:00
Minio Trusted
9f4ad873bc Update yaml files to latest version RELEASE.2020-11-10T21-02-24Z 2020-11-10 21:18:35 +00:00
Steven Reitsma
4683a623dc
fix: negative STS IAM token TTL value (#10866) 2020-11-10 12:24:01 -08:00
Klaus Post
06899210a7
Reduce health check output (#10859)
This will make the health check clients 'silent'.
Use `IsNetworkOrHostDown` determine if network is ok so it mimics the functionality in the actual client.
2020-11-10 09:28:23 -08:00
Harshavardhana
cbdab62c1e
fix: heal user/metadata right away upon server startup (#10863)
this is needed such that we make sure to heal the
users, policies and bucket metadata right away as
we do listing based on list cache which only lists
'3' sufficiently good drives, to avoid possibly
losing access to these users upon upgrade make
sure to heal them.
2020-11-10 09:02:06 -08:00
Harshavardhana
8df6112204
fix: avoid divide by zero error single node distributed setup (#10862) 2020-11-09 20:40:39 -08:00
Anis Elleuch
8e8ddf7233
doc: Add definition of 1KB and 1MB in prometheus (#10857) 2020-11-09 10:05:01 -08:00
Klaus Post
311ab43d4c
Upgrade msgp to official release (#10858)
Code identical, but official release.
2020-11-09 07:52:24 -08:00
Harshavardhana
97692bc772
re-route requests if IAM is not initialized (#10850) 2020-11-07 21:03:06 -08:00
Minio Trusted
21016265e5 Update yaml files to latest version RELEASE.2020-11-06T23-17-07Z 2020-11-06 23:34:19 +00:00
Steven Reitsma
54120107ce
fix: infinite loop in cleanupStaleUploads of encrypted MPUs (#10845)
fixes #10588
2020-11-06 11:53:42 -08:00
Klaus Post
9bf5990ea9
metadata: Invalidate cache if unreadable and not updating (#10844)
If a scanning server shuts down unexpectedly we may have "successful" caches that are incomplete on a set.

In this case mark the cache with an error so it will no longer be handed out.
2020-11-06 08:54:09 -08:00
Steven Reitsma
74f7cf24ae
fix: s3 gateway SSE pagination (#10840)
Fixes #10838
2020-11-05 15:04:03 -08:00
Harshavardhana
fb28aa847b
fix: add missing deleted key element in multiObjectDelete (#10839)
fixes #10832
2020-11-05 12:47:46 -08:00
Klaus Post
0724205f35
metacache: Add option for life extension (#10837)
Add `MINIO_API_EXTEND_LIST_CACHE_LIFE` that will extend 
the life of generated caches for a while.

This changes caches to remain valid until no updates have been 
received for the specified time plus a fixed margin.

This also changes the caches from being invalidated when the *first* 
set finishes until the *last* set has finished plus the specified time 
has passed.
2020-11-05 11:49:56 -08:00
Harshavardhana
b72cac4cf3
fix: dangling objects on actual namespace (#10822) 2020-11-05 11:48:55 -08:00
Klaus Post
47d715f642
Update msgp to main branch (#10835)
The main branch has error unwrapping to clean up logs on canceled contexts, etc.

Asked for a new release: https://github.com/tinylib/msgp/issues/282
2020-11-05 08:13:03 -08:00
Klaus Post
bd77f29fc4
Don't replace caches that are receiving updates (#10834)
Keep caches while they are receiving updates.
Move update code to separate function.
2020-11-05 07:34:08 -08:00
Klaus Post
d1e1205036
metacache: Always close the s2 writer (#10836)
The s2 writer could be leaked if there was an error.

Make sure it is always closed.
2020-11-05 07:30:14 -08:00
Harshavardhana
71753e21e0
add missing TTL for STS credentials on etcd (#10828) 2020-11-04 13:06:05 -08:00
Harshavardhana
fde3299bf3
re-use optimized readdir for isDirEmpty() (#10829)
reduces effective memory usage by an order
of magnitude, also increases performance for
small objects
2020-11-04 13:05:21 -08:00
Harshavardhana
1a1f00fa15
fix: use internode data for DisksInfo, VolsInfo in message pack (#10821)
Similar to #10775 for fewer memory allocations, since we use
getOnlineDisks() extensively for listing we should optimize it
further.

Additionally, remove all unused walkers from the storage layer
2020-11-04 10:10:54 -08:00
Bill Thorp
4a1efabda4
Context based AccessKey passing (#10615)
A new field called AccessKey is added to the ReqInfo struct and populated.
Because ReqInfo is added to the context, this allows the AccessKey to be
accessed from 3rd-party code, such as a custom ObjectLayer.

Co-authored-by: Harshavardhana <harsha@minio.io>
Co-authored-by: Kaloyan Raev <kaloyan@storj.io>
2020-11-04 09:13:34 -08:00
Klaus Post
3b88a646ec
Add remote online/offline information (#10825)
Log information about remote clients being marked offline.

This will help to identify root causes of failures.
2020-11-04 08:27:32 -08:00
Klaus Post
2294e53a0b
Don't retain context in locker (#10515)
Use the context for internal timeouts, but disconnect it from outgoing 
calls so we always receive the results and cancel it remotely.
2020-11-04 08:25:42 -08:00
Klaus Post
f0819cce75
Keep transient lists while they are updating (#10826)
On extremely long running listings keep the transient list 15 minutes after last update instead of using start time.

Also don't do overlap checks on transient lists.
2020-11-04 08:01:33 -08:00
Klaus Post
1e11b4629f
Add remote Diskinfo caching (#10824)
Add 1 second remote disk info cache.

Should decrease need for remote calls a great deal due to how actively it is used now.
2020-11-04 08:00:18 -08:00
Harshavardhana
5c72a34fa8
fix: honor delimiter as per AWS S3 spec (#10823) 2020-11-04 07:56:58 -08:00
Klaus Post
b9277c8030
metacache: Add trashcan (#10820)
Add trashcan that keeps recently updated lists after bucket deletion.
All caches were deleted once a bucket was deleted, so caches still running would report errors. Now they are canceled.
Fix `.minio.sys` not being transient.
2020-11-03 12:47:52 -08:00
Harshavardhana
8c76e1353e
initialize IAM after etcd has initialized (#10819) 2020-11-03 12:12:30 -08:00
Harshavardhana
ad382799b1
use list cache for Walk() with webUI and quota (#10814)
bring list cache optimizations for web UI
object listing, also FIFO quota enforcement
through list cache as well.
2020-11-03 08:53:48 -08:00
Harshavardhana
68de5a6f6a
fix: IAM store fallback to list users and policies from disk (#10787)
Bonus fixes, remove package retry it is harder to get it
right, also manage context remove it such that we don't have
to rely on it anymore instead use a simple Jitter retry.
2020-11-02 17:52:13 -08:00
Harshavardhana
4ea31da889
fix: move list quorum ENV to config (#10804) 2020-11-02 17:21:56 -08:00
Klaus Post
0a796505c1
metacache: Check only one disk for updates (#10809)
Check only one disk for updates.

This will reduce IO while waiting for lists to finish.
2020-11-02 17:20:27 -08:00
Klaus Post
37749f4623
Optimize FileInfo(Version) transfer (#10775)
File Info decoding, in particular, is showing up as a major 
allocator and time consumer for internode data transfers

Switch to message pack for cross-server transfers:

```
MSGP:

Size: 945 bytes

BenchmarkEncodeFileInfoMsgp-32    	 1558444	       866 ns/op	   1.16 MB/s	       0 B/op	       0 allocs/op
BenchmarkDecodeFileInfoMsgp-32    	  479968	      2487 ns/op	   0.40 MB/s	     848 B/op	      18 allocs/op

GOB:

Size: 1409 bytes

BenchmarkEncodeFileInfoGOB-32    	  333339	      3237 ns/op	   0.31 MB/s	     576 B/op	      19 allocs/op
BenchmarkDecodeFileInfoGOB-32    	   20869	     57837 ns/op	   0.02 MB/s	   16439 B/op	     428 allocs/op
```
2020-11-02 17:07:52 -08:00
Klaus Post
86e0d272f3
Reduce WriteAll allocs (#10810)
WriteAll saw 127GB allocs in a 5 minute timeframe for 4MiB buffers 
used by `io.CopyBuffer` even if they are pooled.

Since all writers appear to write byte buffers, just send those 
instead and write directly. The files are opened through the `os` 
package so they have no special properties anyway.

This removes the alloc and copy for each operation.

REST sends content length so a precise alloc can be made.
2020-11-02 16:14:31 -08:00
Harshavardhana
8527f22df1
optimize request URL encoding for internode (#10811)
this reduces allocations in order of magnitude

Also, revert "erasure: delete dangling objects automatically (#10765)" 
affects list caching should be investigated.
2020-11-02 15:15:12 -08:00
Anis Elleuch
b456292295
erasure: delete dangling objects automatically (#10765) 2020-11-02 10:49:30 -08:00
Poorna Krishnamoorthy
03fdbc3ec2
Add async caching commit option in diskcache (#10742)
Add store and a forward option for a single part
uploads when an async mode is enabled with env
MINIO_CACHE_COMMIT=writeback 

It defaults to `writethrough` if unspecified.
2020-11-02 10:00:45 -08:00
Harshavardhana
4c773f7068
re-use remote transports in Peer,Storage,Locker clients (#10788)
use one transport for internode communication
2020-11-02 07:43:11 -08:00
Venkata K Sadineni
d8e07f2c41
go mod update golang.org/x/text to address a CVE (#10805) 2020-11-01 19:42:48 -08:00
Harshavardhana
5412d730c1
simplify monitoring doesn't need to be canceled (#10803)
connect disks monitoring doesn't need to be canceled
upon drive replacement, since we only need to replace
the newly replaced drive.
2020-10-31 14:10:12 -07:00