Commit Graph

85 Commits

Author SHA1 Message Date
Harshavardhana ae4ee95d25
change default lock retry interval to 50ms (#15560)
competing calls on the same object on versioned bucket
mutating calls on the same object may unexpected have
higher delays.

This can be reproduced with a replicated bucket
overwriting the same object writes, deletes repeatedly.

For longer locks like scanner keep the 1sec interval
2022-08-19 16:21:05 -07:00
Harshavardhana 7bd9f821dd
return correct context errors for locking operations (#14569)
if a context is canceled do not need to return a timeout error
instead, return the appropriate error for context canceled.
2022-03-18 15:32:45 -07:00
Klaus Post b030ef1aca
tests: Clean up dsync package (#14415)
Add non-constant timeouts to dsync package.

Reduce test runtime by minutes. Hopefully not too aggressive.
2022-03-01 11:14:28 -08:00
Harshavardhana f527c708f2
run gofumpt cleanup across code-base (#14015) 2022-01-02 09:15:06 -08:00
Harshavardhana 1f262daf6f
rename all remaining packages to internal/ (#12418)
This is to ensure that there are no projects
that try to import `minio/minio/pkg` into
their own repo. Any such common packages should
go to `https://github.com/minio/pkg`
2021-06-01 14:59:40 -07:00
Harshavardhana 64f6020854
fix: cleanup locking, cancel context upon lock timeout (#12183)
upon errors to acquire lock context would still leak,
since the cancel would never be called. since the lock
is never acquired - proactively clear it before returning.
2021-04-29 20:55:21 -07:00
Anis Elleuch 9e797532dc
lock: Always cancel the returned Get(R)Lock context (#12162)
* lock: Always cancel the returned Get(R)Lock context

There is a leak with cancel created inside the locking mechanism. The
cancel purpose was to cancel operations such erasure get/put that are
holding non-refreshable locks.

This PR will ensure the created context.Cancel is passed to the unlock
API so it will cleanup and avoid leaks.

* locks: Avoid returning nil cancel in local lockers

Since there is no Refresh mechanism in the local locking mechanism, we
do not generate a new context or cancel. Currently, a nil cancel
function is returned but this can cause a crash. Return a dummy function
instead.
2021-04-27 16:12:50 -07:00
Harshavardhana 069432566f update license change for MinIO
Signed-off-by: Harshavardhana <harsha@minio.io>
2021-04-23 11:58:53 -07:00
Harshavardhana e85b28398b
fix: pre-allocate certain slices with expected capacity (#12044)
Avoids append() based tiny allocations on known
allocated slices repeated access.
2021-04-12 13:45:06 -07:00
Anis Elleuch 7be7109471
locking: Add Refresh for better locking cleanup (#11535)
Co-authored-by: Anis Elleuch <anis@min.io>
Co-authored-by: Harshavardhana <harsha@minio.io>
2021-03-03 18:36:43 -08:00
Harshavardhana 4550ac6fff
fix: refactor locks to apply them uniquely per node (#11052)
This refactor is done for few reasons below

- to avoid deadlocks in scenarios when number
  of nodes are smaller < actual erasure stripe
  count where in N participating local lockers
  can lead to deadlocks across systems.

- avoids expiry routines to run 1000 of separate
  network operations and routes per disk where
  as each of them are still accessing one single
  local entity.

- it is ideal to have since globalLockServer
  per instance.

- In a 32node deployment however, each server
  group is still concentrated towards the
  same set of lockers that partipicate during
  the write/read phase, unlike previous minio/dsync
  implementation - this potentially avoids send
  32 requests instead we will still send at max
  requests of unique nodes participating in a
  write/read phase.

- reduces overall chattiness on smaller setups.
2020-12-10 07:28:37 -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
Harshavardhana eafa775952
fix: add lock ownership to expire locks (#10571)
- Add owner information for expiry, locking, unlocking a resource
- TopLocks returns now locks in quorum by default, provides
  a way to capture stale locks as well with `?stale=true`
- Simplify the quorum handling for locks to avoid from storage
  class, because there were challenges to make it consistent
  across all situations.
- And other tiny simplifications to reset locks.
2020-09-25 19:21:52 -07:00
Harshavardhana 83a82d818e
allow lock tolerance to match storage-class drive tolerance (#10270) 2020-08-14 18:17:14 -07:00
Harshavardhana d55f4336ae
preserve context per request for local locks (#9828)
In the Current bug we were re-using the context
from previously granted lockers, this would
lead to lock timeouts for existing valid
read or write locks, leading to premature
timeout of locks.

This bug affects only local lockers in FS
or standalone erasure coded mode. This issue
is rather historical as well and was present
in lsync for some time but we were lucky to
not see it.

Similar changes are done in dsync as well
to keep the code more familiar

Fixes #9827
2020-06-14 07:43:10 -07:00
Harshavardhana 4915433bd2
Support bucket versioning (#9377)
- Implement a new xl.json 2.0.0 format to support,
  this moves the entire marshaling logic to POSIX
  layer, top layer always consumes a common FileInfo
  construct which simplifies the metadata reads.
- Implement list object versions
- Migrate to siphash from crchash for new deployments
  for object placements.

Fixes #2111
2020-06-12 20:04:01 -07:00
Harshavardhana febe9cc26a
fix: avoid timer leaks in dsync/lsync (#9781)
At a customer setup with lots of concurrent calls
it can be observed that in newRetryTimer there
were lots of tiny alloations which are not
relinquished upon retries, in this codepath
we were only interested in re-using the timer
and use it wisely for each locker.

```
(pprof) top
Showing nodes accounting for 8.68TB, 97.02% of 8.95TB total
Dropped 1198 nodes (cum <= 0.04TB)
Showing top 10 nodes out of 79
      flat  flat%   sum%        cum   cum%
    5.95TB 66.50% 66.50%     5.95TB 66.50%  time.NewTimer
    1.16TB 13.02% 79.51%     1.16TB 13.02%  github.com/ncw/directio.AlignedBlock
    0.67TB  7.53% 87.04%     0.70TB  7.78%  github.com/minio/minio/cmd.xlObjects.putObject
    0.21TB  2.36% 89.40%     0.21TB  2.36%  github.com/minio/minio/cmd.(*posix).Walk
    0.19TB  2.08% 91.49%     0.27TB  2.99%  os.statNolog
    0.14TB  1.59% 93.08%     0.14TB  1.60%  os.(*File).readdirnames
    0.10TB  1.09% 94.17%     0.11TB  1.25%  github.com/minio/minio/cmd.readDirN
    0.10TB  1.07% 95.23%     0.10TB  1.07%  syscall.ByteSliceFromString
    0.09TB  1.03% 96.27%     0.09TB  1.03%  strings.(*Builder).grow
    0.07TB  0.75% 97.02%     0.07TB  0.75%  path.(*lazybuf).append
```
2020-06-08 11:28:40 -07:00
Harshavardhana 6de410a0aa
fix: possiblity of double write lockers on same resource (#9616)
To avoid this issue with refCounter refactor the code
such that

- locker() always increases refCount upon success
- unlocker() always decrements refCount upon success
  (as a special case removes the resource if the
  refCount is zero)

By these two assumptions we are able to see that we
are never granted two write lockers in any situation.

Thanks to @vcabbage for writing a nice reproducer.
2020-05-18 17:33:35 -07:00
Harshavardhana b730bd1396
fix: possible race in FS local lockMap (#9598) 2020-05-14 23:59:07 -07:00
Harshavardhana f44cfb2863
use GlobalContext whenever possible (#9280)
This change is throughout the codebase to
ensure that all codepaths honor GlobalContext
2020-04-09 09:30:02 -07:00
Harshavardhana ab7d3cd508
fix: Speed up multi-object delete by taking bulk locks (#8974)
Change distributed locking to allow taking bulk locks
across objects, reduces usually 1000 calls to 1.

Also allows for situations where multiple clients sends
delete requests to objects with following names

```
{1,2,3,4,5}
```

```
{5,4,3,2,1}
```

will block and ensure that we do not fail the request
on each other.
2020-02-21 11:29:57 +05:30
Harshavardhana 347b29d059 Implement bucket expansion (#8509) 2019-11-19 17:42:27 -08:00
Harshavardhana e9b2bf00ad Support MinIO to be deployed on more than 32 nodes (#8492)
This PR implements locking from a global entity into
a more localized set level entity, allowing for locks
to be held only on the resources which are writing
to a collection of disks rather than a global level.

In this process this PR also removes the top-level
limit of 32 nodes to an unlimited number of nodes. This
is a precursor change before bring in bucket expansion.
2019-11-13 12:17:45 -08:00
Harshavardhana 4e63e0e372 Return appropriate errors API versions changes across REST APIs (#8480)
This PR adds code to appropriately handle versioning issues
that come up quite constantly across our API changes. Currently
we were also routing our requests wrong which sort of made it
harder to write a consistent error handling code to appropriately
reject or honor requests.

This PR potentially fixes issues

 - old mc is used against new minio release which is incompatible
   returns an appropriate for client action.
 - any older servers talking to each other, report appropriate error
 - incompatible peer servers should report error and reject the calls
   with appropriate error
2019-11-04 09:30:59 -08:00
Harshavardhana d48fd6fde9
Remove unusued params and functions (#8399) 2019-10-15 18:35:41 -07:00
Harshavardhana 36e12a6038 Assume local endpoints appropriately in k8s deployments (#8375)
On Kubernetes/Docker setups DNS resolves inappropriately
sometimes where there are situations same endpoints with
multiple disks come online indicating either one of them
is local and some of them are not local. This situation
can never happen and its only a possibility in orchestrated
deployments with dynamic DNS. Following code ensures that we
treat if one of the endpoint says its local for a given host
it is true for all endpoints for the same host. Following code
ensures that this assumption is true and it works in all
scenarios and it is safe to assume for a given host.

This PR also adds validation such that we do not crash the
server if there are bugs in the endpoints list in dsync
initialization.

Thanks to Daniel Valdivia <hola@danielvaldivia.com> for
reproducing this, this fix is needed as part of the
https://github.com/minio/m3 project.
2019-10-10 10:14:17 +05:30
Krishna Srinivas 2ab0681c0c Do not ignore Lock()'s return value (#8142) 2019-08-28 16:12:57 -07:00
Krishna Srinivas 338e9a9be9 Put object client disconnect (#7824)
Fail putObject  and postpolicy in case client prematurely disconnects
Use request's context to cancel lock requests on client disconnects
2019-06-28 22:09:17 -07:00
Harshavardhana 2c0b3cadfc Update go mod with sem versions of our libraries (#7687) 2019-05-29 16:35:12 -07:00
kannappanr d2f42d830f
Lock: Use REST API instead of RPC (#7469)
In distributed mode, use REST API to acquire and manage locks instead
of RPC.

RPC has been completely removed from MinIO source.

Since we are moving from RPC to REST, we cannot use rolling upgrades as the
nodes that have not yet been upgraded cannot talk to the ones that have
been upgraded.

We expect all minio processes on all nodes to be stopped and then the
upgrade process to be completed.

Also force http1.1 for inter-node communication
2019-04-17 23:16:27 -07:00
kannappanr 5ecac91a55
Replace Minio refs in docs with MinIO and links (#7494) 2019-04-09 11:39:42 -07:00
Krishna Srinivas ef791764e0 Do no access nsLockMap.lockMap when using dsync (#7464)
There is no need to access nsLockMap.lockMap when using dsync
2019-04-02 12:27:20 -07:00
Harshavardhana df35d7db9d Introduce staticcheck for stricter builds (#7035) 2019-02-13 18:29:36 +05:30
kannappanr ce870466ff
Top Locks command implementation (#7052)
API to list locks used in distributed XL mode
2019-01-24 07:22:14 -08:00
Pontus Leitzler df60b3c733 Remove unnecessary contexts passed as data to FatalIf. No need to log an empty context. (#6487) 2018-09-21 16:04:11 -07:00
Harshavardhana 13fbb96736 Hold locks granularly in nslockMap (#6242)
With benchmarks increases the performance for small files
by almost 4x times the previous releases.
2018-08-06 08:55:25 +05:30
Harshavardhana 556a51120c Deprecate ListLocks and ClearLocks (#6233)
No locks are ever left in memory, we also
have a periodic interval of clearing stale locks
anyways. The lock instrumentation was not complete
and was seldom used.

Deprecate this for now and bring it back later if
it is really needed. This also in-turn seems to improve
performance slightly.
2018-08-02 23:09:42 +05:30
Harshavardhana 28d526bc68 Change CriticalIf to FatalIf for proper error message (#6040)
During startup until the object layer is initialized
logger is disabled to provide for a cleaner UI error
message. CriticalIf is disabled, use FatalIf instead.

Also never call os.Exit(1) on running servers where
you can return error to client in handlers.
2018-06-14 10:17:07 -07:00
Bala FA 6a53dd1701 Implement HTTP POST based RPC (#5840)
Added support for new RPC support using HTTP POST.  RPC's 
arguments and reply are Gob encoded and sent as HTTP 
request/response body.

This patch also removes Go RPC based implementation.
2018-06-06 14:21:56 +05:30
kannappanr f8a3fd0c2a
Create logger package and rename errorIf to LogIf (#5678)
Removing message from error logging
Replace errors.Trace with LogIf
2018-04-05 15:04:40 -07:00
Harshavardhana fb96779a8a Add large bucket support for erasure coded backend (#5160)
This PR implements an object layer which
combines input erasure sets of XL layers
into a unified namespace.

This object layer extends the existing
erasure coded implementation, it is assumed
in this design that providing > 16 disks is
a static configuration as well i.e if you started
the setup with 32 disks with 4 sets 8 disks per
pack then you would need to provide 4 sets always.

Some design details and restrictions:

- Objects are distributed using consistent ordering
  to a unique erasure coded layer.
- Each pack has its own dsync so locks are synchronized
  properly at pack (erasure layer).
- Each pack still has a maximum of 16 disks
  requirement, you can start with multiple
  such sets statically.
- Static sets set of disks and cannot be
  changed, there is no elastic expansion allowed.
- Static sets set of disks and cannot be
  changed, there is no elastic removal allowed.
- ListObjects() across sets can be noticeably
  slower since List happens on all servers,
  and is merged at this sets layer.

Fixes #5465
Fixes #5464
Fixes #5461
Fixes #5460
Fixes #5459
Fixes #5458
Fixes #5460
Fixes #5488
Fixes #5489
Fixes #5497
Fixes #5496
2018-02-15 17:45:57 -08:00
Harshavardhana f3f09ed14e Fix a bug in dsync initialization and communication (#5428)
In current implementation we used as many dsync clients
as per number of endpoints(along with path) which is not
the expected implementation. The implementation of Dsync
was expected to be just for the endpoint Host alone such
that if you have 4 servers and each with 4 disks we need
to only have 4 dsync clients and 4 dsync servers. But
we currently had 8 clients, servers which in-fact is
unexpected and should be avoided.

This PR brings the implementation back to its original
intention. This issue was found #5160
2018-01-22 10:25:10 -08:00
ebozduman 24d9d7e5fa Removes logrus package and refactors logging messages (#5293)
This fix removes logrus package dependency and refactors the console
logging as the only logging mechanism by removing file logging support.
It rearranges the log message format and adds stack trace information
whenever trace information is not available in the error structure.
It also adds `--json` flag support for server logging.
When minio server is started with `--json` flag, all log messages are
displayed in json format, with no start-up and informational log
messages.
Fixes #5265 #5220 #5197
2018-01-17 07:24:46 -08:00
poornas 0bb6247056 Move nslocking from s3 layer to object layer (#5382)
Fixes #5350
2018-01-13 10:04:52 +05:30
Krishna Srinivas 14e6c5ec08 Simplify the steps to make changes to config.json (#5186)
This change introduces following simplified steps to follow 
during config migration.

```
 // Steps to move from version N to version N+1
 // 1. Add new struct serverConfigVN+1 in config-versions.go
 // 2. Set configCurrentVersion to "N+1"
 // 3. Set serverConfigCurrent to serverConfigVN+1
 // 4. Add new migration function (ex. func migrateVNToVN+1()) in config-migrate.go
 // 5. Call migrateVNToVN+1() from migrateConfig() in config-migrate.go
 // 6. Make changes in config-current_test.go for any test change
```
2017-11-29 13:12:47 -08:00
Harshavardhana 3d0dced23c Remove go1.9 specific code for windows (#5033)
Following fix https://go-review.googlesource.com/#/c/41834/ has
been merged upstream and released with go1.9.
2017-10-13 15:31:15 +05:30
Frank Wessels 61e0b1454a Add support for timeouts for locks (#4377) 2017-08-31 14:43:59 -07:00
Aditya Manthramurthy 986aa8fabf Bypass network in lock requests to local server (#4465)
This makes lock RPCs similar to other RPCs where requests to the local
server bypass the network. Requests to the local lock-subsystem may
bypass the network layer and directly access the locking
data-structures.

This incidentally fixes #4451.
2017-06-05 12:25:04 -07:00
Frank cae4683971 Make clearing of stale debug lock info independent of deleting map entry of lock itself. (#4353)
This is believed to address issue #4337 where stale information for debug locks in shown.
2017-05-16 07:19:17 -07:00
Bala FA de204a0a52 Add extensive endpoints validation (#4019) 2017-04-11 15:44:27 -07:00