The healing code repeatedly tries to heal a root disk when it is empty
the reason is that connectEndpoint() returns errUnformattedDisk even
if the disk is a root disk. Changing that to returning another error
will avoid queueing the disk to the healing code in each connect disks
iteration.
This speed-up is intended for faster startup times
for almost all MinIO operations. Changes here are
- Drives are not re-read for 'format.json' on a regular
basis once read during init is remembered and refreshed
at 5 second intervals.
- Do not do O_DIRECT tests on drives with existing 'format.json'
only fresh setups need this check.
- Parallelize initializing erasureSets for multiple sets.
- Avoid re-reading format.json when migrating 'format.json'
from really old V1->V2->V3
- Keep a copy of local drives for any given server in memory
for a quick lookup.
```
λ mc admin decommission start alias/ http://minio{1...2}/data{1...4}
```
```
λ mc admin decommission status alias/
┌─────┬─────────────────────────────────┬──────────────────────────────────┬────────┐
│ ID │ Pools │ Capacity │ Status │
│ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Active │
│ 2nd │ http://minio{3...4}/data{1...4} │ 329 GiB (used) / 421 GiB (total) │ Active │
└─────┴─────────────────────────────────┴──────────────────────────────────┴────────┘
```
```
λ mc admin decommission status alias/ http://minio{1...2}/data{1...4}
Progress: ===================> [1GiB/sec] [15%] [4TiB/50TiB]
Time Remaining: 4 hours (started 3 hours ago)
```
```
λ mc admin decommission status alias/ http://minio{1...2}/data{1...4}
ERROR: This pool is not scheduled for decommissioning currently.
```
```
λ mc admin decommission cancel alias/
┌─────┬─────────────────────────────────┬──────────────────────────────────┬──────────┐
│ ID │ Pools │ Capacity │ Status │
│ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Draining │
└─────┴─────────────────────────────────┴──────────────────────────────────┴──────────┘
```
> NOTE: Canceled decommission will not make the pool active again, since we might have
> Potentially partial duplicate content on the other pools, to avoid this scenario be
> very sure to start decommissioning as a planned activity.
```
λ mc admin decommission cancel alias/ http://minio{1...2}/data{1...4}
┌─────┬─────────────────────────────────┬──────────────────────────────────┬────────────────────┐
│ ID │ Pools │ Capacity │ Status │
│ 1st │ http://minio{1...2}/data{1...4} │ 439 GiB (used) / 561 GiB (total) │ Draining(Canceled) │
└─────┴─────────────────────────────────┴──────────────────────────────────┴────────────────────┘
```
- add checks such that swapped disks are detected
and ignored - never used for normal operations.
- implement `unrecognizedDisk` to be ignored with
all operations returning `errDiskNotFound`.
- also add checks such that we do not load unexpected
disks while connecting automatically.
- additionally humanize the values when printing the errors.
Bonus: fixes handling of non-quorum situations in
getLatestFileInfo(), that does not work when 2 drives
are down, currently this function would return errors
incorrectly.
Logger targets were not race protected against concurrent updates from for example `HTTPConsoleLoggerSys`.
Restrict direct access to targets and make slices immutable so a returned slice can be processed safely without locks.
In (erasureServerPools).MakeBucketWithLocation deletes the created
buckets if any set returns an error.
Add `NoRecreate` option, which will not recreate the bucket
in `DeleteBucket`, if the operation fails.
Additionally use context.Background() for operations we always want to be performed.
An endpoint can be empty when a disk is offline or something
wrong with it. Avoid it by filling erasureSets.endpointStrings
with values from arguments.
DeleteObject() on existing objects before `xl.json` to
`xl.meta` change were not working, not sure when this
regression was added. This PR fixes this properly.
Also this PR ensures that we perform rename of xl.json
to xl.meta only during "write" phase of the call i.e
either during Healing or PutObject() overwrites.
Also handles few other scenarios during migration where
`backendEncryptedFile` was missing deleteConfig() will
fail with `configNotFound` this case was not ignored,
which can lead to failure during upgrades.
we will allow situations such as
```
a/b/1.txt
a/b
```
and
```
a/b
a/b/1.txt
```
we are going to document that this usecase is
not supported and we will never support it, if
any application does this users have to delete
the top level parent to make sure namespace is
accessible at lower level.
rest of the situations where the prefixes get
created across sets are supported as is.
This commit gathers MRF metrics from
all nodes in a cluster and return it to the caller. This will show information about the
number of objects in the MRF queues
waiting to be healed.
its possible that, version might exist on second pool such that
upon deleteBucket() might have deleted the bucket on pool1 successfully
since it doesn't have any objects, undo such operations properly in
all any error scenario.
Also delete bucket metadata from pool layer rather than sets layer.
objectErasureMap in the audit holds information about the objects
involved in the current S3 operation such as pool index, set an index,
and disk endpoints. One user saw a crash due to a concurrent update of
objectErasureMap information. Use sync.Map to prevent a crash.
healing code was using incorrect buffers to heal older
objects with 10MiB erasure blockSize, incorrect calculation
of such buffers can lead to incorrect premature closure of
io.Pipe() during healing.
fixes#12410
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`
MRF does not detect when a node is disconnected and reconnected quickly
this change will ensure that MRF is alerted by comparing the last disk
reconnection timestamp with the last MRF check time.
Signed-off-by: Anis Elleuch <anis@min.io>
Co-authored-by: Klaus Post <klauspost@gmail.com>
With this change, MinIO's ILM supports transitioning objects to a remote tier.
This change includes support for Azure Blob Storage, AWS S3 compatible object
storage incl. MinIO and Google Cloud Storage as remote tier storage backends.
Some new additions include:
- Admin APIs remote tier configuration management
- Simple journal to track remote objects to be 'collected'
This is used by object API handlers which 'mutate' object versions by
overwriting/replacing content (Put/CopyObject) or removing the version
itself (e.g DeleteObjectVersion).
- Rework of previous ILM transition to fit the new model
In the new model, a storage class (a.k.a remote tier) is defined by the
'remote' object storage type (one of s3, azure, GCS), bucket name and a
prefix.
* Fixed bugs, review comments, and more unit-tests
- Leverage inline small object feature
- Migrate legacy objects to the latest object format before transitioning
- Fix restore to particular version if specified
- Extend SharedDataDirCount to handle transitioned and restored objects
- Restore-object should accept version-id for version-suspended bucket (#12091)
- Check if remote tier creds have sufficient permissions
- Bonus minor fixes to existing error messages
Co-authored-by: Poorna Krishnamoorthy <poorna@minio.io>
Co-authored-by: Krishna Srinivas <krishna@minio.io>
Signed-off-by: Harshavardhana <harsha@minio.io>
Multiple disks from the same set would be writing concurrently.
```
WARNING: DATA RACE
Write at 0x00c002100ce0 by goroutine 166:
github.com/minio/minio/cmd.(*erasureSets).connectDisks.func1()
d:/minio/minio/cmd/erasure-sets.go:254 +0x82f
Previous write at 0x00c002100ce0 by goroutine 129:
github.com/minio/minio/cmd.(*erasureSets).connectDisks.func1()
d:/minio/minio/cmd/erasure-sets.go:254 +0x82f
Goroutine 166 (running) created at:
github.com/minio/minio/cmd.(*erasureSets).connectDisks()
d:/minio/minio/cmd/erasure-sets.go:210 +0x324
github.com/minio/minio/cmd.(*erasureSets).monitorAndConnectEndpoints()
d:/minio/minio/cmd/erasure-sets.go:288 +0x244
Goroutine 129 (finished) created at:
github.com/minio/minio/cmd.(*erasureSets).connectDisks()
d:/minio/minio/cmd/erasure-sets.go:210 +0x324
github.com/minio/minio/cmd.(*erasureSets).monitorAndConnectEndpoints()
d:/minio/minio/cmd/erasure-sets.go:288 +0x244
```
Current implementation heavily relies on readAllFileInfo
but with the advent of xl.meta inlined with data, we cannot
easily avoid reading data when we are only interested is
updating metadata, this leads to invariably write
amplification during metadata updates, repeatedly reading
data when we are only interested in updating metadata.
This PR ensures that we implement a metadata only update
API at storage layer, that handles updates to metadata alone
for any given version - given the version is valid and
present.
This helps reduce the chattiness for following calls..
- PutObjectTags
- DeleteObjectTags
- PutObjectLegalHold
- PutObjectRetention
- ReplicateObject (updates metadata on replication status)
The local node name is heavily used in tracing, create a new global
variable to store it. Multiple goroutines can access it since it won't be
changed later.
MRF was starting to heal when it receives a disk connection event, which
is not good when a node having multiple disks reconnects to the cluster.
Besides, MRF needs Remove healing option to remove stale files.
major performance improvements in range GETs to avoid large
read amplification when ranges are tiny and random
```
-------------------
Operation: GET
Operations: 142014 -> 339421
Duration: 4m50s -> 4m56s
* Average: +139.41% (+1177.3 MiB/s) throughput, +139.11% (+658.4) obj/s
* Fastest: +125.24% (+1207.4 MiB/s) throughput, +132.32% (+612.9) obj/s
* 50% Median: +139.06% (+1175.7 MiB/s) throughput, +133.46% (+660.9) obj/s
* Slowest: +203.40% (+1267.9 MiB/s) throughput, +198.59% (+753.5) obj/s
```
TTFB from 10MiB BlockSize
```
* First Access TTFB: Avg: 81ms, Median: 61ms, Best: 20ms, Worst: 2.056s
```
TTFB from 1MiB BlockSize
```
* First Access TTFB: Avg: 22ms, Median: 21ms, Best: 8ms, Worst: 91ms
```
Full object reads however do see a slight change which won't be
noticeable in real world, so not doing any comparisons
TTFB still had improvements with full object reads with 1MiB
```
* First Access TTFB: Avg: 68ms, Median: 35ms, Best: 11ms, Worst: 1.16s
```
v/s
TTFB with 10MiB
```
* First Access TTFB: Avg: 388ms, Median: 98ms, Best: 20ms, Worst: 4.156s
```
This change should affect all new uploads, previous uploads should
continue to work with business as usual. But dramatic improvements can
be seen with these changes.
* Provide information on *actively* healing, buckets healed/queued, objects healed/failed.
* Add concurrent healing of multiple sets (typically on startup).
* Add bucket level resume, so restarts will only heal non-healed buckets.
* Print summary after healing a disk is done.
also re-use storage disks for all `mc admin server info`
calls as well, implement a new LocalStorageInfo() API
call at ObjectLayer to lookup local disks storageInfo
also fixes bugs where there were double calls to StorageInfo()
most of the delete calls today spend time in
a blocking operation where multiple calls need
to be recursively sent to delete the objects,
instead we can use rename operation to atomically
move the objects from the namespace to `tmp/.trash`
we can schedule deletion of objects at this
location once in 15, 30mins and we can also add
wait times between each delete operation.
this allows us to make delete's faster as well
less chattier on the drives, each server runs locally
a groutine which would clean this up regularly.
This commit removes the `GetObject` method
from the `ObjectLayer` interface.
The `GetObject` method is not longer used by
the HTTP handlers implementing the high-level
S3 semantics. Instead, they use the `GetObjectNInfo`
method which returns both, an object handle as well
as the object metadata.
Therefore, it is no longer necessary that a concrete
`ObjectLayer` implements `GetObject`.
store the cache in-memory instead of disks to avoid large
write amplifications for list heavy workloads, store in
memory instead and let it auto expire.
server startup code expects the object layer to properly
convert error into a proper type, so that in situations when
servers are coming up and quorum is not available servers
wait on each other.
If the periodic `case <-t.C:` save gets held up for a long time it will end up
synchronize all disk writes for saving the caches.
We add jitter to per set writes so they don't sync up and don't hold a
lock for the write, since it isn't needed anyway.
If an outage prevents writes for a long while we also add individual
waits for each disk in case there was a queue.
Furthermore limit the number of buffers kept to 2GiB, since this could get
huge in large clusters. This will not act as a hard limit but should be enough
for normal operation.
This change moves away from a unified constructor for plaintext and encrypted
usage. NewPutObjReader is simplified for the plain-text reader use. For
encrypted reader use, WithEncryption should be called on an initialized PutObjReader.
Plaintext:
func NewPutObjReader(rawReader *hash.Reader) *PutObjReader
The hash.Reader is used to provide payload size and md5sum to the downstream
consumers. This is different from the previous version in that there is no need
to pass nil values for unused parameters.
Encrypted:
func WithEncryption(encReader *hash.Reader,
key *crypto.ObjectKey) (*PutObjReader, error)
This method sets up encrypted reader along with the key to seal the md5sum
produced by the plain-text reader (already setup when NewPutObjReader was
called).
Usage:
```
pReader := NewPutObjReader(rawReader)
// ... other object handler code goes here
// Prepare the encrypted hashed reader
pReader, err = pReader.WithEncryption(encReader, objEncKey)
```
- lock maintenance loop was incorrectly sleeping
as well as using ticker badly, leading to
extra expiration routines getting triggered
that could flood the network.
- multipart upload cleanup should be based on
timer instead of ticker, to ensure that long
running jobs don't get triggered twice.
- make sure to get right lockers for object name
currently we had a restriction where older setups would
need to follow previous style of "stripe" count being same
expansion, we can relax that instead newer pools can be
expanded for older setups with newer constraints of
common parity ratio.
For objects with `N` prefix depth, this PR reduces `N` such network
operations by converting `CheckFile` into a single bulk operation.
Reduction in chattiness here would allow disks to be utilized more
cleanly, while maintaining the same functionality along with one
extra volume check stat() call is removed.
Update tests to test multiple sets scenario
parentDirIsObject is not using set level understanding
to check for parent objects, without this it can lead to
objects that can actually reside on a separate set as
objects and would conflict.
Current implementation requires server pools to have
same erasure stripe sizes, to facilitate same SLA
and expectations.
This PR allows server pools to be variadic, i.e they
do not have to be same erasure stripe sizes - instead
they should have SLA for parity ratio.
If the parity ratio cannot be guaranteed by the new
server pool, the deployment is rejected i.e server
pool expansion is not allowed.
issue was introduced in #11106 the following
pattern
<-t.C // timer fired
if !t.Stop() {
<-t.C // timer hangs
}
Seems to hang at the last `t.C` line, this
issue happens because a fired timer cannot be
Stopped() anymore and t.Stop() returns `false`
leading to confusing state of usage.
Refactor the code such that use timers appropriately
with exact requirements in place.
crawler should only ListBuckets once not for each serverPool,
buckets are same across all pools, across sets and ListBuckets
always returns an unified view, once list buckets returns
sort it by create time to scan the latest buckets earlier
with the assumption that latest buckets would have lesser
content than older buckets allowing them to be scanned faster
and also to be able to provide more closer to latest view.
With new refactor of bucket healing, healing bucket happens
automatically including its metadata, there is no need to
redundant heal buckets also in ListBucketsHeal remove
it.
optimization mainly to avoid listing the entire
`.minio.sys/buckets/.minio.sys` directory, this
can get really huge and comes in the way of startup
routines, contents inside `.minio.sys/buckets/.minio.sys`
are rather transient and not necessary to be healed.
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.
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
Bonus fixes, we do not need reload format anymore
as the replaced drive is healed locally we only need
to ensure that drive heal reloads the drive properly.
We preserve the UUID of the original order, this means
that the replacement in `format.json` doesn't mean that
the drive needs to be reloaded into memory anymore.
fixes#10791
only newly replaced drives get the new `format.json`,
this avoids disks reloading their in-memory reference
format, ensures that drives are online without
reloading the in-memory reference format.
keeping reference format in-tact means UUIDs
never change once they are formatted.
reference format should be source of truth
for inconsistent drives which reconnect,
add them back to their original position
remove automatic fix for existing offline
disk uuids
Bonus fixes
- logging improvements to ensure that we don't use
`go logger.LogIf` to avoid runtime.Caller missing
the function name. log where necessary.
- remove unused code at erasure sets
connect disks pre-emptively upon startup, to ensure we have
enough disks are connected at startup rather than wait
for them.
we need to do this to avoid long wait times for server to
be online when we have servers come up in rolling upgrade
fashion
This PR fixes a hang which occurs quite commonly at higher concurrency
by allowing following changes
- allowing lower connections in time_wait allows faster socket open's
- lower idle connection timeout to ensure that we let kernel
reclaim the time_wait connections quickly
- increase somaxconn to 4096 instead of 2048 to allow larger tcp
syn backlogs.
fixes#10413
- select lockers which are non-local and online to have
affinity towards remote servers for lock contention
- optimize lock retry interval to avoid sending too many
messages during lock contention, reduces average CPU
usage as well
- if bucket is not set, when deleteObject fails make sure
setPutObjHeaders() honors lifecycle only if bucket name
is set.
- fix top locks to list out always the oldest lockers always,
avoid getting bogged down into map's unordered nature.
add a hint on the disk to allow for tracking fresh disk
being healed, to allow for restartable heals, and also
use this as a way to track and remove disks.
There are more pending changes where we should move
all the disk formatting logic to backend drives, this
PR doesn't deal with this refactor instead makes it
easier to track healing in the future.
- 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.
Also, revamp the way ListBuckets work make few portions
of the healing logic parallel
- walk objects for healing disks in parallel
- collect the list of buckets in parallel across drives
- provide consistent view for listBuckets()
It was observed in VMware vsphere environment during a
pod replacement, `mc admin info` might report incorrect
offline nodes for the replaced drive. This issue eventually
goes away but requires quite a lot of time for all servers
to be in sync.
This PR fixes this behavior properly.
This is to ensure that Go contexts work properly, after some
interesting experiments I found that Go net/http doesn't
cancel the context when Body is non-zero and hasn't been
read till EOF.
The following gist explains this, this can lead to pile up
of go-routines on the server which will never be canceled
and will die at a really later point in time, which can
simply overwhelm the server.
https://gist.github.com/harshavardhana/c51dcfd055780eaeb71db54f9c589150
To avoid this refactor the locking such that we take locks after we
have started reading from the body and only take locks when needed.
Also, remove contextReader as it's not useful, doesn't work as expected
context is not canceled until the body reaches EOF so there is no point
in wrapping it with context and putting a `select {` on it which
can unnecessarily increase the CPU overhead.
We will still use the context to cancel the lockers etc.
Additional simplification in the locker code to avoid timers
as re-using them is a complicated ordeal avoid them in
the hot path, since locking is very common this may avoid
lots of allocations.
inconsistent drive healing when one of the drive is offline
while a new drive was replaced, this change is to ensure
that we can add the offline drive back into the mix by
healing it again.
Anis Elleuch
Harshavardhana
yinhen
Klaus Post
Krishnan Parthasarathi
Andreas Auernhammer
Poorna Krishnamoorthy