PutObject on multiple-zone with versioning would not
overwrite the correct location of the object if the
object has delete marker, leading to duplicate objects
on two zones.
This PR fixes by adding affinity towards delete marker
when GetObjectInfo() returns error, use the zone index
which has the delete marker.
Bonus change to use channel to serialize triggers,
instead of using atomic variables. More efficient
mechanism for synchronization.
Co-authored-by: Nitish Tiwari <nitish@minio.io>
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
When updating all servers following the constructions of mc update,
only the endpoint server will be updated successfully.
All the other peer servers' updating failed due to the error below:
--------------------------------------------------------------------------
parsing time "2006-01-02T15:04:05Z07:00" as "<release version>": cannot parse "-01-02T15:04:05Z07:00" as "0-"
--------------------------------------------------------------------------
- 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
Historically due to lack of support for middlewares
we ended up writing wrapped handlers for all
middlewares on top of the gorilla/mux, this causes
multiple issues when we want to let's say
- Overload r.Body with some custom implementation
to track the incoming Reads()
- Add other sort of top level checks to avoid
DDOSing the server with large incoming HTTP
bodies.
Since 1.7.x release gorilla/mux provides proper
use of middlewares, which are honored by the muxer
directly. This makes sure that Go can honor its
own internal ServeHTTP(w, r) implementation where
Go net/http can wrap into its own customer readers.
This PR as a side-affect fixes rare issues of client
hangs which were reported in the wild but never really
understood or fixed in our codebase.
Fixes#9759Fixes#7266Fixes#6540Fixes#5455Fixes#5150
Refer https://github.com/boto/botocore/pull/1328 for
one variation of the same issue in #9759
PR #9801 while it is correct, the loop isEndpointConnected()
was changed to rely on endpoint.String() which has the host
information as well, which is not correct value as input to
detect if the disk is down or up, if endpoint is local use
its local path value instead.
This commit changes the data key generation such that
if a MinIO server/nodes tries to generate a new DEK
but the particular master key does not exist - then
MinIO asks KES to create a new master key and then
requests the DEK again.
From now on, a SSE-S3 master key must not be created
explicitly via: `kes key create <key-name>`.
Instead, it is sufficient to just set the env. var.
```
export MINIO_KMS_KES_KEY_NAME=<key-name>
```
However, the MinIO identity (mTLS client certificate)
must have the permission to access the `/v1/key/create/`
API. Therefore, KES policy for MinIO must look similar to:
```
[
/v1/key/create/<key-name-pattern>
/v1/key/generate/<key-name-pattern>
/v1/key/decrypt/<key-name-pattern>
]
```
However, in our guides we already suggest that.
See e.g.: https://github.com/minio/kes/wiki/MinIO-Object-Storage#kes-server-setup
***
The ability to create master keys on request may also be
necessary / useful in case of SSE-KMS.
Current code was relying on globalEndpoints as
the source of secondary truth to obtain
the missing endpoints list when the disk
is offline, this is problematic
- there is no way to know if the getDisks()
returned endpoints total is same as the
ones list of globalEndpoints and it
belongs to a particular set.
- there is no order guarantee as getDisks()
is ordered as per format.json, globalEndpoints
may not be, so potentially end up including
incorrect endpoints.
To fix this bring getEndpoints() just like getDisks()
to ensure that consistently ordered endpoints are
always available for us to ensure that returned values
are consistent with what each erasure set would observe.
Uploading files with names that could not be written to disk
would result in "reduce your request" errors returned.
Instead check explicitly for disallowed characters and reject
files with `Object name contains unsupported characters.`
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
```
For example `{1...17}/{1...52}` symmetrical
distribution of drives cannot be obtained
- Because 17 is a prime number
- Is not divisible by any pre-defined setCounts i.e
from 1 to 16
Manual healing (as background healing) creates a heal task with a
possiblity to override healing options, such as deep or normal mode.
Use a pointer type in heal opts so nil would mean use the default
healing options.
aws cli fails to set a bucket encryption configuration to MinIO server.
The reason is that aws cli does not send MD5-Content header. It seems
that MD5-Content is not required anymore.
This commit also returns Not Implemented header early to help mint tests
to ignore testing this API in gateway modes.
CopyObject was not correctly figuring out the correct
destination object location and would end up creating
duplicate objects on two different zones, reproduced
by doing encryption based key rotation.
Advantages avoids 100's of stats which are needed for each
upload operation in FS/NAS gateway mode when uploading a large
multipart object, dramatically increases performance for
multipart uploads by avoiding recursive calls.
For other gateway's simplifies the approach since
azure, gcs, hdfs gateway's don't capture any specific
metadata during upload which needs handler validation
for encryption/compression.
Erasure coding was already optimized, additionally
just avoids small allocations of large data structure.
Fixes#7206
GetDiskID() in storage rest client does not really issue a REST request
to the remote disk, but returns an in-memory value instead.
However, GetDiskID() should return an error when format.json is not
found or for other similar issues (unmounted disks, etc..)
GetDiskID() is only called when formatting disks and getting storage
informatio, hence this commit should not have a performance degradation.
Additionally also fix STS logs to filter out LDAP
password to be sent out in audit logs.
Bonus fix handle the reload of users properly by
making sure to preserve the newer users during the
reload to be not invalidated.
Fixes#9707Fixes#9644Fixes#9651
Bonus fixes in quota enforcement to use the
new datastructure and use timedValue to cache
a value/reload automatically avoids one less
global variable.
If the requested server is part of the set this will always read
from the local disk, even if the disk contains a parity shard.
In default setup there is a 50% chance that at least
one shard that otherwise would have been fetched remotely
will be read locally instead.
It basically trades RPC call overhead for reed-solomon.
On distributed localhost this seems to be fairly break-even,
with a very small gain in throughput and latency.
However on networked servers this should be a bigger
1MB objects, before:
```
Operation: GET. Concurrency: 32. Hosts: 4.
Requests considered: 76257:
* Avg: 25ms 50%: 24ms 90%: 32ms 99%: 42ms Fastest: 7ms Slowest: 67ms
* First Byte: Average: 23ms, Median: 22ms, Best: 5ms, Worst: 65ms
Throughput:
* Average: 1213.68 MiB/s, 1272.63 obj/s (59.948s, starting 14:45:44 CEST)
```
After:
```
Operation: GET. Concurrency: 32. Hosts: 4.
Requests considered: 78845:
* Avg: 24ms 50%: 24ms 90%: 31ms 99%: 39ms Fastest: 8ms Slowest: 62ms
* First Byte: Average: 22ms, Median: 21ms, Best: 6ms, Worst: 57ms
Throughput:
* Average: 1255.11 MiB/s, 1316.08 obj/s (59.938s, starting 14:43:58 CEST)
```
Bonus fix: Only ask for heal once on an object.
This value is requested on every upload when there are multiple zones.
Since this will result in an RPC call to every remote disk this scales
quite badly in a distributed setup. Load every 1second interval.
2 servers, localhost only. In large distributed setups much bigger
gains can be expected.
```
Operations: 21743 -> 22454
* Average: +3.28% (+0.0 MiB/s) throughput, +3.28% (+11.9) obj/s
* Fastest: +3.37% (+0.0 MiB/s) throughput, +3.37% (+13.0) obj/s
* 50% Median: +3.03% (+0.0 MiB/s) throughput, +3.03% (+11.2) obj/s
* Slowest: +8.03% (+0.0 MiB/s) throughput, +8.03% (+22.8) obj/s
```
For easy management of this a generic helper has been added.
some clients such as veeam expect the x-amz-meta to
be sent in lower cased form, while this does indeed
defeats the HTTP protocol contract it is harder to
change these applications, while these applications
get fixed appropriately in future.
x-amz-meta is usually sent in lowercased form
by AWS S3 and some applications like veeam
incorrectly end up relying on the case sensitivity
of the HTTP headers.
Bonus fixes
- Fix the iso8601 time format to keep it same as
AWS S3 response
- Increase maxObjectList to 50,000 and use
maxDeleteList as 10,000 whenever multi-object
deletes are needed.
No one really uses FS for large scale accounting
usage, neither we crawl in NAS gateway mode. It is
worthwhile to simply disable this feature as its
not useful for anyone.
Bonus disable bucket quota ops as well in, FS
and gateway mode
size calculation in crawler was using the real size
of the object instead of its actual size i.e either
a decrypted or uncompressed size.
this is needed to make sure all other accounting
such as bucket quota and mcs UI to display the
correct values.
This PR adds a new configuration parameter which allows readiness
check to respond within 10secs, this can be reduced to a lower value
if necessary using
```
mc admin config set api ready_deadline=5s
```
or
```
export MINIO_API_READY_DEADLINE=5s
```
net/http exposes ErrorLog but it is log.Logger
instance not an interface which can be overridden,
because of this reason the logging is interleaved
sometimes with TLS with messages like this on the
server
```
http: TLS handshake error from 139.178.70.188:63760: EOF
```
This is bit problematic for us as we need to have
consistent logging view for allow --json or --quiet
flags.
With this PR we ensure that this format is adhered to.
Groups information shall be now stored as part of the
credential data structure, this is a more idiomatic
way to support large LDAP groups.
Avoids the complication of setups where LDAP groups
can be in the range of 150+ which may lead to excess
HTTP header size > 8KiB, to reduce such an occurrence
we shall save the group information on the server as
part of the credential data structure.
Bonus change support multiple mapped policies, across
all types of users.
This PR is a continuation from #9586, now the
entire parsing logic is fully merged into
bucket metadata sub-system, simplify the
quota API further by reducing the remove
quota handler implementation.
Shuffling arguments that we pass to MinIO server are supported. However,
when that happens, Prometheus returns wrong information about disks usage
and online/offline status.
The commit fixes the issue by avoiding relying on xl.endpoints since
it is not ordered.
this is a major overhaul by migrating off all
bucket metadata related configs into a single
object '.metadata.bin' this allows us for faster
bootups across 1000's of buckets and as well
as keeps the code simple enough for future
work and additions.
Additionally also fixes#9396, #9394
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.
enable linter using golangci-lint across
codebase to run a bunch of linters together,
we shall enable new linters as we fix more
things the codebase.
This PR fixes the first stage of this
cleanup.
There is a disparency of behavior under Linux & Windows about
the returned error when trying to rename a non existant path.
err := os.Rename("/path/does/not/exist", "/tmp/copy")
Linux:
isSysErrNotDir(err) = false
os.IsNotExist(err) = true
Windows:
isSysErrNotDir(err) = true
os.IsNotExist(err) = true
ENOTDIR in Linux is returned when the destination path
of the rename call contains a file in one of the middle
segments of the path (e.g. /tmp/file/dst, where /tmp/file
is an actual file not a directory)
However, as shown above, Windows has more scenarios when
it returns ENOTDIR. For example, when the source path contains
an inexistant directory in its path.
In that case, we want errFileNotFound returned and not
errFileAccessDenied, so this commit will add a further check to close
the disparency between Windows & Linux.
The `ioutil.NopCloser(reader)` was hiding nested hash readers.
We make it an `io.Closer` so it can be attached without wrapping
and allows for nesting, by merging the requests.
The `keepHTTPResponseAlive` would cause errors to be
returned with status OK.
- Add '32' as a filler byte until a response is ready
- '0' to indicate the response is ready to be consumed
- '1' to indicate response has an error which needs
to be returned to the caller
Clear out 'file not found' errors from dir walker, since it may be
in a folder that has been deleted since it was scanned.
This PR is to ensure that we call the relevant object
layer APIs for necessary S3 API level functionalities
allowing gateway implementations to return proper
errors as NotImplemented{}
This allows for all our tests in mint to behave
appropriately and can be handled appropriately as
well.
S3 is now natively supported by B2 cloud storage provider
there is no reason to use specialized gateway for B2 anymore,
our current S3 gateway with caching would work with B2.
Resolves#8584
requests in federated setups for STS type calls which are
performed at '/' resource should be routed by the muxer,
the assumption is simply such that requests without a bucket
in a federated setup cannot be proxied, so serve them at
current server.
This commit makes the KES client use HTTP/2
when establishing a connection to the KES server.
This is necessary since the next KES server release
will require HTTP/2.
We should allow quorum errors to be send upwards
such that caller can retry while reading bucket
encryption/policy configs when server is starting
up, this allows distributed setups to load the
configuration properly.
Current code didn't facilitate this and would have
never loaded the actual configs during rolling,
server restarts.
In large setups this avoids unnecessary data transfer
across nodes and potential locks.
This PR also optimizes heal result channel, which should
be avoided for each queueHealTask as its expensive
to create/close channels for large number of objects.
This PR allows setting a "hard" or "fifo" quota
restriction at the bucket level. Buckets that
have reached the FIFO quota configured, will
automatically be cleaned up in FIFO manner until
bucket usage drops to configured quota.
If a bucket is configured with a "hard" quota
ceiling, all further writes are disallowed.
ResponseWriter & RecordAPIStats has similar role, merge them.
This commit will also fix wrong auditing for STS and Web and others
since they are using ResponseWriter instead of the RecordAPIStats.
A user can incorrectly mounts a newly fresh disk. MinIO will detect
that it is writing with a rootfs disk and will mark it down. However,
it is hard for the user to understand what's going on.
This commit will just print a notice so it will be easy to spot
such use case.
- elasticsearch client should rely on the SDK helpers
instead of pure HTTP calls.
- webhook shouldn't need to check for IsActive() for
all notifications, failure should be delayed.
- Remove DialHTTP as its never used properly
Fixes#9460
allow generating service accounts for temporary credentials
which have a designated parent, currently OpenID is not yet
supported.
added checks to ensure that service account cannot generate
further service accounts for itself, service accounts can
never be a parent to any credential.
Audit was not working properly when enabled from the environment
caused by a typo in the code.
This commit fixes that but also consider the following variables:
`MINIO_LOGGER_WEBHOOK_ENABLE_*` and
`MINIO_AUDIT_WEBHOOK_ENABLE_*` so the user can use
this latter to temporarily disable a logger or audit configuration.
data usage tracker and crawler seem to be logging
non-actionable information on console, which is not
useful and is fixed on its own in almost all deployments,
lets keep this logging to minimal.
it is possible in many screnarios that even
if the divisible value is optimal, we may
end up with uneven distribution due to number
of nodes present in the configuration.
added code allow for affinity towards various
ellipses to figure out optimal value across
ellipses such that we can always reach a
symmetric value automatically.
Fixes#9416
By monitoring PUT/DELETE and heal operations it is possible
to track changed paths and keep a bloom filter for this data.
This can help prioritize paths to scan. The bloom filter can identify
paths that have not changed, and the few collisions will only result
in a marginal extra workload. This can be implemented on either a
bucket+(1 prefix level) with reasonable performance.
The bloom filter is set to have a false positive rate at 1% at 1M
entries. A bloom table of this size is about ~2500 bytes when serialized.
To not force a full scan of all paths that have changed cycle bloom
filters would need to be kept, so we guarantee that dirty paths have
been scanned within cycle runs. Until cycle bloom filters have been
collected all paths are considered dirty.
this commit avoids lots of tiny allocations, repeated
channel creates which are performed when filtering
the incoming events, unescaping a key just for matching.
also remove deprecated code which is not needed
anymore, avoids unexpected data structure transformations
from the map to slice.
we have policy available for sub-admin users to set/get/delete
config, but we incorrectly decrypt the content using admin secret
key which in-fact should be the credential authenticating the
request.
global WORM mode is a complex piece for which
the time has passed, with the advent of S3 compatible
object locking and retention implementation global
WORM is sort of deprecated, this has been mentioned
in our documentation for some time, now the time
has come for this to go.
re-implement the cache purging routine to
avoid using ioutil.ReadDir which can lead
to high allocations when there are cache
directories with lots of content, or
when cache is installed in memory constrainted
environments.
Instead rely on a callback function where we
are not using memory no-more than 8KiB per
cycle.
Precursor for this change refer #9425, original
issue pointed by Caleb Case <caleb@storj.io>
OSS go sdk lacks licensing terms in their
repository, and there has been no activity
On the issue here https://github.com/aliyun/aliyun-oss-go-sdk/issues/245
This PR is to ensure we remove any dependency code which
lacks explicit license file in their repo.
- keep long running obd network tests alive
- fix error - wrong number of parents in process OBD info
- ensure that osinfo does not error out when inside containers
- remove limit on max number of connections per client transport
The generic client transport uses a default limit of 64 conns per transport.
This could end up limiting and throttling usage, and artificially slowing
down the performance of MinIO even on hardware capable of doing better.
New value defaults to 100K events by default,
but users can tune this value upto any value
they seem necessary.
* increase the limit to maxint64 while validating
This PR also fixes issues when
deletePolicy, deleteUser is idempotent so can lead to
issues when client can prematurely timeout, so a retry
call error response should be ignored when call returns
http.StatusNotFound
Fixes#9347
allow `mc admin config set mygateway/ audit_webhook --env`
to fetch the documentation as needed, this is just to
ensure that our users can still access the relevant
ENV docs while running in gateway mode.
Instead of GlobalContext use a local context for tests.
Most notably this allows stuff created to be shut down
when tests using it is done. After PR #9345 9331 CI is
often running out of memory/time.
Add two new configuration entries, api.requests-max and
api.requests-deadline which have the same role of
MINIO_API_REQUESTS_MAX and MINIO_API_REQUESTS_DEADLINE.
This PR fixes couple of behaviors with service accounts
- not need to have session token for service accounts
- service accounts can be generated by any user for themselves
implicitly, with a valid signature.
- policy input for AddNewServiceAccount API is not fully typed
allowing for validation before it is sent to the server.
- also bring in additional context for admin API errors if any
when replying back to client.
- deprecate GetServiceAccount API as we do not need to reply
back session tokens
- Introduced a function `FetchRegisteredTargets` which will return
a complete set of registered targets irrespective to their states,
if the `returnOnTargetError` flag is set to `False`
- Refactor NewTarget functions to return non-nil targets
- Refactor GetARNList() to return a complete list of configured targets
- Removes PerfInfo admin API as its not OBDInfo
- Keep the drive path without the metaBucket in OBD
global latency map.
- Remove all the unused code related to PerfInfo API
- Do not redefined global mib,gib constants use
humanize.MiByte and humanize.GiByte instead always
if needed use --no-compat to disable md5sum while
verifying any performance numbers.
bring back --compat behavior as default to avoid
additional documentation and confusing behavior,
as we are working towards improving md5sum to
be faster on AVX instructions, enabling this
should be hardly a problem in future versions
of MinIO.
fixes#8012fixes#7859fixes#7642
Continuing from previous PR #9304, comment
is a special key is not present in the
default KV list. Add it explicitly when
tokenizing fields as it may be possible that
some clients might try to set comments.
This PR adds context-based `k=v` splits based
on the sub-system which was obtained, if the
keys are not provided an error will be thrown
during parsing, if keys are provided with wrong
values an error will be thrown. Keys can now
have values which are of a much more complex
form such as `k="v=v"` or `k=" v = v"`
and other variations.
additionally, deprecate unnecessary postgres/mysql
configuration styles, support only
- connection_string for Postgres
- dsn_string for MySQL
All other parameters are removed.
This commit fixes a performance issue caused
by too many calls to the external KMS - i.e.
for single-part PUT requests.
In general, the issue is caused by a sub-optimal
code structure. In particular, when the server
encrypts an object it requests a new data encryption
key from the KMS. With this key it does some key
derivation and encrypts the object content and
ETag.
However, to behave S3-compatible the MinIO server
has to return the plaintext ETag to the client
in case SSE-S3.
Therefore, the server code used to decrypt the
(previously encrypted) ETag again by requesting
the data encryption key (KMS decrypt API) from
the KMS.
This leads to 2 KMS API calls (1 generate key and
1 decrypt key) per PUT operation - while only
one KMS call is necessary.
This commit fixes this by fetching a data key only
once from the KMS and keeping the derived object
encryption key around (for the lifetime of the request).
This leads to a significant performance improvement
w.r.t. to PUT workloads:
```
Operation: PUT
Operations: 161 -> 239
Duration: 28s -> 29s
* Average: +47.56% (+25.8 MiB/s) throughput, +47.56% (+2.6) obj/s
* Fastest: +55.49% (+34.5 MiB/s) throughput, +55.49% (+3.5) obj/s
* 50% Median: +58.24% (+32.8 MiB/s) throughput, +58.24% (+3.3) obj/s
* Slowest: +1.83% (+0.6 MiB/s) throughput, +1.83% (+0.1) obj/s
```
Fixes#8667
In addition to the above, if the user is mapped to a policy or
belongs in a group, the user-info API returns this information,
but otherwise, the API will now return a non-existent user error.
make rest of the Walk() function more predictable,
it was observed that in nominal deployments even
without much workload the drives are generally
slow for respond for readdir operations, for the
sleepDuration factor of 10 this can cause
unexpected slowness in the Listing calls, while
it is good for all other I/O, it may simply slow
down Listing immensely which is not useful.
fixes#9261
In FS mode under Windows, removing an object will not automatically.
remove parent empty prefixes.
The reason is that path.Dir() was used, however filepath.Dir() is
more appropriate since filepath is physical (meaning it operates
on OS filesystem paths)
This is not caught because failure for Windows CI is not caught.
fs-v1 in server mode only checks to see if the path exist, so that it
returns ready before it is indeed ready.
This change adds a check to ensure that the global object api is
available too before reporting ready.
Fixes#9283
It is some times common and convenient to use
just local IPs for testing purposes, 127.0.0.x
are special IPs regardless of being available on
an interface they can be bound to on all operating
systems.
Allow this behavior to work for minio server
fixes#9274
also, bring in an additional policy to ensure that
force delete bucket is only allowed with the right
policy for the user, just DeleteBucketAction
policy action is not enough.
This PR also tries to simplify the approach taken in
object-locking implementation by preferential treatment
given towards full validation.
This in-turn has fixed couple of bugs related to
how policy should have been honored when ByPassGovernance
is provided.
Simplifies code a bit, but also duplicates code intentionally
for clarity due to complex nature of object locking
implementation.
Too many deployments come up with an odd number
of hosts or drives, to facilitate even distribution
among those setups allow for odd and prime numbers
based packs.
- B2 does actually return an MD5 hash for newly uploaded objects
so we can use it to provide better compatibility with S3 client
libraries that assume the ETag is the MD5 hash such as boto.
- depends on change in blazer library.
- new behaviour is only enabled if MinIO's --compat mode is active.
- behaviour for multipart uploads is unchanged (works fine as is).
- Implement a graph algorithm to test network bandwidth from every
node to every other node
- Saturate any network bandwidth adaptively, accounting for slow
and fast network capacity
- Implement parallel drive OBD tests
- Implement a paging mechanism for OBD test to provide periodic updates to client
- Implement Sys, Process, Host, Mem OBD Infos
- total number of S3 API calls per server
- maximum wait duration for any S3 API call
This implementation is primarily meant for situations
where HDDs are not capable enough to handle the incoming
workload and there is no way to throttle the client.
This feature allows MinIO server to throttle itself
such that we do not overwhelm the HDDs.
- acquire since leader lock for all background operations
- healing, crawling and applying lifecycle policies.
- simplify lifecyle to avoid network calls, which was a
bug in implementation - we should hold a leader and
do everything from there, we have access to entire
name space.
- make listing, walking not interfere by slowing itself
down like the crawler.
- effectively use global context everywhere to ensure
proper shutdown, in cache, lifecycle, healing
- don't read `format.json` for prometheus metrics in
StorageInfo() call.
- Add conservative timeouts upto 3 minutes
for internode communication
- Add aggressive timeouts of 30 seconds
for gateway communication
Fixes#9105Fixes#8732Fixes#8881Fixes#8376Fixes#9028
This is to improve responsiveness for all
admin API operations and allowing callers
to cancel any on-going admin operations,
if they happen to be waiting too long.
canonicalize the ENVs such that we can bring these ENVs
as part of the config values, as a subsequent change.
- fix location of per bucket usage to `.minio.sys/buckets/<bucket_name>/usage-cache.bin`
- fix location of the overall usage in `json` at `.minio.sys/buckets/.usage.json`
(avoid conflicts with a bucket named `usage.json` )
- fix location of the overall usage in `msgp` at `.minio.sys/buckets/.usage.bin`
(avoid conflicts with a bucket named `usage.bin`
As an optimization of the healing, HealObjects() avoid sending an
object to the background healing subsystem when the object is
present in all disks.
However, HealObjects() should have checked the scan type, if this
deep, always pass the object to the healing subsystem.
Currently, a tree walking, needed to a list objects in a specific
set quits listing as long as it finds no entries in a disk, which
is wrong.
This affected background healing, because the latter is using
tree walk directly. If one object does not exist in the first
disk for example, it will be seemed like the object does not
exist at all and no healing work is needed.
This commit fixes the behavior.
The staleness of a lock should be determined by
the quorum number of entries returning stale,
this allows for situations when locks are held
when nodes are down - we don't accidentally
clear locks unintentionally when they are valid
and correct.
Also lock maintenance should be run by all servers,
not one server, stale locks need to be run outside
the requirement for holding distributed locks.
Thanks @klauspost for reproducing this issue
Some AWS SDKs latently rely on this value some times
to calculate the right number of parts during a parallel
GetObject request, this is feature used along with
content-range - we should support this as well.
- avoid setting last heal activity when starting self-healing
This can be confusing to users thinking that the self healing
cycle was already performed.
- add info about the next background healing round
OperationTimedout error occurs when locking
timesout, trying to acquire a lock. This
error should be returned appropriately to
the client with http status "408" (request timedout)
This translation was broken, fix it.
Bulk delete API was using cleanupObjectsBulk() which calls posix
listing and delete API to remove objects internal files in the
backend (xl.json and parts) one by one.
Add DeletePrefixes in the storage API to remove the content
of a directory in a single call.
Also use a remove goroutine for each disk to accelerate removal.
Currently the code assumed some orthogonal requirements
which led situations where when we have a setup where
we have let's say for example 168 drives, the final
set_drive_count chosen was 14. Indeed 168 drives are
divisible by 12 but this wasn't allowed due to an
unexpected requirement to have 12 to be a perfect modulo
of 14 which is not possible. This assumption was incorrect.
This PR fixes this old assumption properly, also adds
few tests and some negative tests as well. Improvements
are seen in error messages as well.
- Remove the requirement to honor storage class for deletes
- Improve `posix.DeleteFileBulk` code to Stat the volumeDir
only once per call, rather than for all object paths.
Recent modification in the code led to incorrect calculation
of offline disks.
This commit saves the endpoint list in a xlObjects then we know
the name of each disk.
lock ownership is limited to endpoints on first zone,
as we do not hold locks on other zones in an expanded
setup. current code unintentionally expired active locks
when it couldn't see ownership from the secondary zone
which leads to unexpected bugs as locking fails to work
as expected.
this PR enforces md5sum verification for following
API's to be compatible with AWS S3 spec
- PutObjectRetention
- PutObjectLegalHold
Co-authored-by: Harshavardhana <harsha@minio.io>
Allow downloading goroutine dump to help detect leaks
or overuse of goroutines.
Extensions are now type dependent.
Change `profiling` -> `profile` prefix, since that is what they are
not the abstract concept.
This is a precursor change before versioning,
removes/deprecates the requirement of remembering
partName and partETag which are not useful after
a multipart transaction has finished.
This PR reduces the overall size of the backend
JSON for large file uploads.
For a non-existent user server would return STS not initialized
```
aws --profile harsha --endpoint-url http://localhost:9000 \
sts assume-role \
--role-arn arn:xxx:xxx:xxx:xxxx \
--role-session-name anything
```
instead return an appropriate error as expected by STS API
Additionally also format the `trace` output for STS APIs
Upgrades between releases are failing due to strict
rule to avoid rolling upgrades, it is enough to
bump up APIs between versions to allow for quorum
failure and wait times. Authentication failures are
catastrophic in nature which leads to server not
be able to upgrade properly.
Fixes#9021Fixes#8968
To allow better control the cache eviction process.
Introduce MINIO_CACHE_WATERMARK_LOW and
MINIO_CACHE_WATERMARK_HIGH env. variables to specify
when to stop/start cache eviction process.
Deprecate MINIO_CACHE_EXPIRY environment variable. Cache
gc sweeps at 30 minute intervals whenever high watermark is
reached to clear least recently accessed entries in the cache
until sufficient space is cleared to reach the low watermark.
Garbage collection uses an adaptive file scoring approach based
on last access time, with greater weights assigned to larger
objects and those with more hits to find the candidates for eviction.
Thanks to @klauspost for this file scoring algorithm
Co-authored-by: Klaus Post <klauspost@minio.io>
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.
Metrics used to have its own code to calculate offline disks.
StorageInfo() was avoided because it is an expensive operation
by sending calls to all nodes.
To make metrics & server info share the same code, a new
argument `local` is added to StorageInfo() so it will only
query local disks when needed.
Metrics now calls StorageInfo() as server info handler does
but with the local flag set to false.
Co-authored-by: Praveen raj Mani <praveen@minio.io>
Co-authored-by: Harshavardhana <harsha@minio.io>
Add dummy calls which respond success when ACL's
are set to be private and fails, if user tries
to change them from their default 'private'
Some applications such as nuxeo may have an
unnecessary requirement for this operation,
we support this anyways such that don't have
to fully implement the functionality just that
we can respond with success for default ACLs
Avoid GetObjectNInfo call from cache in CopyObjectHandler
- in the case of server side copy with metadata replacement,
the reader returned from cache is never consumed, but the net
effect of GetObjectNInfo from cache layer, is cache holding a
write lock to fill the cache. Subsequent stat operation on cache in
CopyObject is not able to acquire a read lock, thus causing the hang.
Fixes#8991
we don't need to validateFormats again once we have obtained
reference format, because it is possible that at this stage
another server is doing a disk heal during startup, once
in a while due to delays we get false positives and our
server doesn't start.
Format in quorum as reference format can be assumed as valid
and we proceed further, until and unless HealFormat re-inits
the disks after a successful heal.
Also use separate port for healing tests to avoid any
conflicts with regular build testing.
Fixes#8884
RegisterNotificationTargets() cleans up all connections
that it makes to notification targets when an error occurs
during its execution.
However there is a typo in the code that makes the function to always
try to access to a nil pointer in the defer code since the function
in question will always return nil in the case of any error.
This commit fixes the typo in the code.
http.Request.ContentLength can be negative, which affects
the gateway_s3_bytes_received value in Prometheus output.
The commit only increases the value of the total received bytes
in gateway mode when r.ContentLength is greater than zero.
First step is to ensure that Path component is not decoded
by gorilla/mux to avoid routing issues while handling
certain characters while uploading through PutObject()
Delay the decoding and use PathUnescape() to escape
the `object` path component.
Thanks to @buengese and @ncw for neat test cases for us
to test with.
Fixes#8950Fixes#8647
We added support for caching and S3 related metrics in #8591. As
a continuation, it would be helpful to add support for Azure & GCS
gateway related metrics as well.
The logging subsystem was initialized under init() method in
both gateway-main.go and server-main.go which are part of
same package. This created two logging targets and hence
errors were logged twice. This PR moves the init() method
to common-main.go
Streams are returning a readcloser and returning would
decrement io count instantly, fix it.
change maxActiveIOCount to 3, meaning it will pause
crawling if 3 operations are running.
Remove the random sleep. This is running in 4 goroutines,
so mostly doing nothing.
We use the getSize latency to estimate system load,
meaning when there is little load on the system and
we get the result fast we sleep a little.
If it took a long time we have high load and release
ourselves longer.
We are sleeping inside the mutex so this affects all
goroutines doing IO.
Due to a typo in the code, a cluster was not correctly creating
`background-ops` in all disks and nodes print the following error:
minio3_1 | API: SYSTEM()
minio3_1 | Time: 19:32:45 UTC 02/06/2020
minio3_1 | DeploymentID: d67c20fa-4a1e-41f5-b319-7e3e90f425d8
minio3_1 | Error: Bucket not found: .minio.sys/background-ops
minio3_1 | 2: cmd/data-usage.go:109:cmd.runDataUsageInfo()
minio3_1 | 1: cmd/data-usage.go:56:cmd.runDataUsageInfoUpdateRoutine()
This commit fixes the typo.
Adding mutex slows down the crawler to avoid large
spikes in CPU, also add millisecond interval jitter
in calculation of disk usage to slow down the spikes
further.
This is to fix a situation where an object name incorrectly
is sent with '//' in its path heirarchy, we should reject
such object names because they may be hashed to a set where
the object might not originally belong because, this can
cause situations where once object is uploaded we cannot
delete it anymore.
Fixes#8873
This commit fixes typos in the displayed server info
w.r.t. the KMS and removes the update status.
For more information about why the update status
is removed see: PR #8943
This commit removes the `Update` functionality
from the admin API. While this is technically
a breaking change I think this will not cause
any harm because:
- The KMS admin API is not complete, yet.
At the moment only the status can be fetched.
- The `mc` integration hasn't been merged yet.
So no `mc` client could have used this API
in the past.
The `Update`/`Rewrap` status is not useful anymore.
It provided a way to migrate from one master key version
to another. However, KES does not support the concept of
key versions. Instead, key migration should be implemented
as migration from one master key to another.
Basically, the `Update` functionality has been implemented just
for Vault.
- pkg/bucket/encryption provides support for handling bucket
encryption configuration
- changes under cmd/ provide support for AES256 algorithm only
Co-Authored-By: Poorna <poornas@users.noreply.github.com>
Co-authored-by: Harshavardhana <harsha@minio.io>
XL crawling wrongly returns a zero buckets count when
there are no objects uploaded in the server yet. The reason is
data of the crawler of posix returns invalid result when all
disks has zero objects.
A simple fix is to always pick the crawling result of the first
disk but choose over the result of the disk which has the most
objects in it.
looks like 1024 buffer size is not enough in
all situations, use 8192 instead which
can satisfy all the rare situations that
may arise in base64 decoding.
multi-delete API failed with write quorum errors
under following situations
- list of files requested for delete doesn't exist
anymore can lead to quorum errors and failure
- due to usage of query param for paths, for really
long paths MinIO server rejects these requests as
malformed as unexpected.
This was reproduced with warp
The server info handler makes a http connection to other
nodes to check if they are up but does not load the custom
CAs in ~/.minio/certs/CAs.
This commit fix it.
Co-authored-by: Harshavardhana <harsha@minio.io>
JWT parsing is simplified by using a custom claim
data structure such as MapClaims{}, also writes
a custom Unmarshaller for faster unmarshalling.
- Avoid as much reflections as possible
- Provide the right types for functions as much
as possible
- Avoid strings.Join, strings.Split to reduce
allocations, rely on indexes directly.
For 'snapshot' type profiles, record a 'before' profile that can be used
as `go tool pprof -base=before ...` to compare before and after.
"Before" profiles are included in the zipped package.
[`runtime.MemProfileRate`](https://golang.org/pkg/runtime/#pkg-variables)
should not be updated while the application is running, so we set it at startup.
Co-authored-by: Harshavardhana <harsha@minio.io>
On every restart of the server, usage was being
calculated which is not useful instead wait for
sufficient time to start the crawling routine.
This PR also avoids lots of double allocations
through strings, optimizes usage of string builders
and also avoids crawling through symbolic links.
Fixes#8844
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.
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.
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.
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.
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.
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
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.
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
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.
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)
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
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
- 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>
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.
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.
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
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>
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.
In existing functionality we simply return a generic
error such as "MalformedPolicy" which indicates just
a generic string "invalid resource" which is not very
meaningful when there might be multiple types of errors
during policy parsing. This PR ensures that we send
these errors back to client to indicate the actual
error, brings in two concrete types such as
- iampolicy.Error
- policy.Error
Refer #8202
minio server /data{1..4} shows an error about inability to bind a port, though
the real problem is /data{1..4} cannot be created because of the lack of
permissions.
This commit fix the behavior.
- We should declare a cluster ready even if read quorum is achieved (atleast n/2 disks are online).
- Such that, all the zones should have enough read quorum. Thus making the cluster ready for reads.
We had messy cyclical dependency problem with `mc`
due to dependencies in pkg/console, moved the pkg/console
to minio for more control and also to avoid any further
cyclical dependencies of `mc` clobbering up the
dependencies on server.
Fixes#8659
If two distinct clusters are started with different domains
along with single common domain, this situation was leading
to conflicting buckets getting created on different clusters
To avoid this do not prematurely error out if the key has no
entries, let the caller decide on which entry matches and
which entry is valid. This allows support for MINIO_DOMAIN
with one common domain, but each cluster may have their own
domains.
Fixes#8705
This is to ensure that when we have multiple tenants
deployed all sharing the same etcd for global bucket
should avoid listing each others buckets, this leads
to information leak which should be avoided unless
etcd is not namespaced for IAM assets in which case
it can be assumed that its a federated setup.
Federated setup and namespaced IAM assets on etcd
is not supported since namespacing is only useful
when you wish to separate the tenants as isolated
instances of MinIO.
This PR allows a new type of behavior, primarily
driven by the usecase of m3(mkube) multi-tenant
deployments with global bucket support.
Currently all bucket events are sent to all watchers
with matching prefix and event names, this becomes
problematic and prone to performance issues, fix this
situation by filtering based on buckets as well.
This PR fixes the issue where we might allow policy changes
for temporary credentials out of band, this situation allows
privilege escalation for those temporary credentials. We
should disallow any external actions on temporary creds
as a practice and we should clearly differentiate which
are static and which are temporary credentials.
Refer #8667
Changes in IP underneath are dynamic in replica sets
with multiple tenants, so deploying in that fashion
will not work until we wait for atleast one participatory
server to be local.
This PR also ensures that multi-tenant zone expansion also
works in replica set k8s deployments.
Introduces a new ENV `KUBERNETES_REPLICA_SET` check to call
appropriate code paths.
Continuation from #8629 which basically broke
zone deployments on k8s statefulset environment
due to incorrect assumptions which made it work
on replicated set.
Fix this properly such that this container works
for both replicated set and stateful set deployment
This commit removes github.com/minio/kes as
a dependency and implements the necessary
client-side functionality without relying
on the KES project.
This resolves the licensing issue since
KES is licensed under AGPL while MinIO
is licensed under Apache.
With this PR,cache eviction will continue until
no LRU entries older than an hour can be cache
evicted or sufficient percentage of disk space
has been reclaimed.
This ensures that we can update the
- .minio.sys is updated for accounting/data usage purposes
- .minio.sys is updated to indicate if backend is encrypted
or not.
The approach is that now safe mode is only invoked when
we cannot read the config or under some catastrophic
situations, but not under situations when config entries
are invalid or unreachable. This allows for maximum
availability for MinIO and not fail on our users unlike
most of our historical releases.
This commit adds support for the minio/kes KMS.
See: https://github.com/minio/kes
In particular you can configure it as KMS by:
- `export MINIO_KMS_KES_ENDPOINT=` // Server URL
- `export MINIO_KMS_KES_KEY_FILE=` // TLS client private key
- `export MINIO_KMS_KES_CERT_FILE=` // TLS client certificate
- `export MINIO_KMS_KES_CA_PATH=` // Root CAs issuing server cert
- `export MINIO_KMS_KES_KEY_NAME=` // The name of the (default)
master key
Harshavardhana
Klaus Post
ebozduman
Anis Elleuch
ethan ho
kannappanr
Andreas Auernhammer
poornas
P R
Praveen raj Mani
Nathan Brown
Anton Huck
Erkki Eilonen
Krishna Srinivas
Sidhartha Mani
remche
Bala FA
Egor Rudinsky
Dmitry Gadeev
BigUstad
Boaz
Egon Elbre
Nitish Tiwari
Taras Parkhomenko
Aditya Manthramurthy
Pontus Leitzler
César Nieto
Ingmar Runge
Stephen N
Kody A Kantor
astorath
Krishnan Parthasarathi
Ashish Kumar Sinha
Forest Lovewood
George Xie
Amol Umbarkar