AWS S3 trailing header support was recently enabled on the warm tier
client connection to MinIO type remote tiers. With this enabled, we are
seeing the following error message at http transport layer.
> Unsolicited response received on idle HTTP channel starting with "HTTP/1.1 400 Bad Request\r\nContent-Type: text/plain; charset=utf-8\r\nConnection: close\r\n\r\n400 Bad Request"; err=<nil>
This is an interim fix until we identify the root cause for this behaviour in the
minio-go client package.
Keep the EC in header, so it can be retrieved easily for dynamic quorum calculations.
To not force a full metadata decode on every read the value will be 0/0 for data written in previous versions.
Size is expected to increase by 2 bytes per version, since all valid values can be represented with 1 byte each.
Example:
```
λ xl-meta xl.meta
{
"Versions": [
{
"Header": {
"EcM": 4,
"EcN": 8,
"Flags": 6,
"ModTime": "2024-04-17T11:46:25.325613+02:00",
"Signature": "0a409875",
"Type": 1,
"VersionID": "8e03504e11234957b2727bc53eda0d55"
},
...
```
Not used for operations yet.
Follow up for #19528
If there are multiple existing DN mappings for the same normalized DN,
if they all have the same policy mapping value, we pick one of them of
them instead of returning an import error.
This is a change to IAM export/import functionality. For LDAP enabled
setups, it performs additional validations:
- for policy mappings on LDAP users and groups, it ensures that the
corresponding user or group DN exists and if so uses a normalized form
of these DNs for storage
- for access keys (service accounts), it updates (i.e. validates
existence and normalizes) the internally stored parent user DN and group
DNs.
This allows for a migration path for setups in which LDAP mappings have
been stored in previous versions of the server, where the name of the
mapping file stored on drives is not in a normalized form.
An administrator needs to execute:
`mc admin iam export ALIAS`
followed by
`mc admin iam import ALIAS /path/to/export/file`
The validations are more strict and returns errors when multiple
mappings are found for the same user/group DN. This is to ensure the
mappings stored by the server are unambiguous and to reduce the
potential for confusion.
Bonus **bug fix**: IAM export of access keys (service accounts) did not
export key name, description and expiration. This is fixed in this
change too.
Reading the list metacache is not protected by a lock; the code retries when it fails
to read the metacache object, however, it forgot to re-read the metacache object
from the drives, which is necessary, especially if the metacache object is inlined.
This commit will ensure that we always re-read the metacache object from the drives
when it is retrying.
When resuming a versioned listing where `version-id-marker=null`, the `null` object would
always be returned, causing duplicate entries to be returned.
Add check against empty version
unlinking() at two different locations on a disk when there
are lots to purge, this can lead to huge IOwaits, instead
rely on rename() to .trash to avoid running multiple unlinks()
in parallel.
since mid 2018 we do not have any deployments
without deployment-id, it is time to put this
code to rest, this PR removes this old code as
its no longer valuable.
on setups with 1000's of drives these are all
quite expensive operations.
The rest of the peer clients were not consistent across nodes. So, meta cache requests
would not go to the same server if a continuation happens on a different node.
When no results match or another error occurs, add an error to the stream. Keep the "inspect-input.txt" as the only thing in the zip for reference.
Example:
```
λ mc support inspect --airgap myminio/testbucket/fjghfjh/**
mc: Using public key from C:\Users\klaus\mc\support_public.pem
File data successfully downloaded as inspect-data.enc
λ inspect inspect-data.enc
Using private key from support_private.pem
output written to inspect-data.zip
2024/04/11 14:10:51 next stream: GetRawData: No files matched the given pattern
λ unzip -l inspect-data.zip
Archive: inspect-data.zip
Length Date Time Name
--------- ---------- ----- ----
222 2024-04-11 14:10 inspect-input.txt
--------- -------
222 1 file
λ
```
Modifies inspect to read until end of stream to report the error.
Bonus: Add legacy commandline params
Add following metrics:
- used_inodes
- total_inodes
- healing
- online
- reads_per_sec
- reads_kb_per_sec
- reads_await
- writes_per_sec
- writes_kb_per_sec
- writes_await
- perc_util
To be able to calculate the `per_sec` values, we capture the IOStats-related
data in the beginning (along with the time at which they were captured),
and compare them against the current values subsequently. This is because
dividing by "time since server uptime." doesn't work in k8s environments.
the disk location never changes in the lifetime of a
MinIO cluster, even if it did validate this close to the
disk instead at the higher layer.
Return appropriate errors indicating an invalid drive, so
that the drive is not recognized as part of a valid
drive.
we have had numerous reports on some config
values not having default values, causing
features misbehaving and not having default
values set properly.
This PR tries to address all these concerns
once and for all.
Each new sub-system that gets added
- must check for invalid keys
- must have default values set
- must not "return err" when being saved into
a global state() instead collate as part of
other subsystem errors allow other sub-systems
to independently initialize.
* Allow specifying the local server, with env variable _MINIO_SERVER_LOCAL, in systems where the hostname cannot be resolved to local IP
* Limit scope of the _MINIO_SERVER_LOCAL solution to only containerized implementations
Return an error when the user specifies endpoints for both source
and target. This can generate many type of errors as the code considers
a deployment remote if its endpoint is specified.
HealObject() does not return an error in some cases, for example, when
an object is successfully reconstructed in one disk but fails with other
disks, another case is when a disk does not have the object is temporarily
disconnected
Add the After heal drives result in the audit output for better
analysis.
Set object's modTime when being restored
restored here refers to making a temporary local copy in the hot tier
for a tiered object using the RestoreObject API
we have been using an LRU caching for internode
auth tokens, migrate to using a typed implementation
and also do not cache auth tokens when its an error.
This fixes a regression from #19358 which prevents policy mappings
created in the latest release from being displayed in policy entity
listing APIs.
This is due to the possibility that the base DNs in the LDAP config are
not in a normalized form and #19358 introduced normalized of mapping
keys (user DNs and group DNs). When listing, we check if the policy
mappings are on entities that parse as valid DNs that are descendants of
the base DNs in the config.
Test added that demonstrates a failure without this fix.
Create new code paths for multiple subsystems in the code. This will
make maintaing this easier later.
Also introduce bugLogIf() for errors that should not happen in the first
place.
This commit replaces the `KMS.Stat` API call with a
`KMS.GenerateKey` call. This approach is more reliable
since data key generation also works when the KMS backend
is unavailable (temp. offline), but KES has cached the
key. Ref: KES offline caching.
With this change, it is less likely that MinIO readiness
checks fail in cases where the KMS backend is offline.
Signed-off-by: Andreas Auernhammer <github@aead.dev>
Make sure to pass a nil pointer as a Transport to minio-go when the API config
is not initialized, this will make sure that we do not pass an interface
with a known type but a nil value.
This will also fix the update of the API remote_transport_deadline
configuration without requiring the cluster restart.
Use `ODirectPoolSmall` buffers for inline data in PutObject.
Add a separate call for inline data that will fetch a buffer for the inline data before unmarshal.
This fixes a bug where STS Accounts map accumulates accounts in memory
and never removes expired accounts and the STS Policy mappings were not
being refreshed.
The STS purge routine now runs with every IAM credentials load instead
of every 4th time.
The listing of IAM files is now cached on every IAM load operation to
prevent re-listing for STS accounts purging/reload.
Additionally this change makes each server pick a time for IAM loading
that is randomly distributed from a 10 minute interval - this is to
prevent server from thundering while performing the IAM load.
On average, IAM loading will happen between every 5-15min after the
previous IAM load operation completes.
If site replication enabled across sites, replicate the SSE-C
objects as well. These objects could be read from target sites
using the same client encryption keys.
Signed-off-by: Shubhendu Ram Tripathi <shubhendu@minio.io>
Instead of relying on user input values, we use the DN value returned by
the LDAP server.
This handles cases like when a mapping is set on a DN value
`uid=svc.algorithm,OU=swengg,DC=min,DC=io` with a user input value (with
unicode variation) of `uid=svc﹒algorithm,OU=swengg,DC=min,DC=io`. The
LDAP server on lookup of this DN returns the normalized value where the
unicode dot character `SMALL FULL STOP` (in the user input), gets
replaced with regular full stop.
Bonus: remove persistent md5sum calculation, turn-off
sha256 as well. Instead we always enable crc32c which
is enough for payload verification also support for
trailing headers checksum.
Fix races in IAM cache
Fixes#19344
On the top level we only grab a read lock, but we write to the cache if we manage to fetch it.
a03dac41eb/cmd/iam-store.go (L446) is also flipped to what it should be AFAICT.
Change the internal cache structure to a concurrency safe implementation.
Bonus: Also switch grid implementation.
we must attempt to convert all errors at storage-rest-client
into StorageErr() regardless of what functionality is being
called in, this PR fixes this for multiple callers including
some internally used functions.
- old version was unable to retain messages during config reload
- old version could not go from memory to disk during reload
- new version can batch disk queue entries to single for to reduce I/O load
- error logging has been improved, previous version would miss certain errors.
- logic for spawning/despawning additional workers has been adjusted to trigger when half capacity is reached, instead of when the log queue becomes full.
- old version would json marshall x2 and unmarshal 1x for every log item. Now we only do marshal x1 and then we GetRaw from the store and send it without having to re-marshal.
panic seen due to premature closing of slow channel while listing is still sending or
list has already closed on the sender's side:
```
panic: close of closed channel
goroutine 13666 [running]:
github.com/minio/minio/internal/ioutil.SafeClose[...](0x101ff51e4?)
/Users/kp/code/src/github.com/minio/minio/internal/ioutil/ioutil.go:425 +0x24
github.com/minio/minio/cmd.(*erasureServerPools).Walk.func1()
/Users/kp/code/src/github.com/minio/minio/cmd/erasure-server-pool.go:2142 +0x170
created by github.com/minio/minio/cmd.(*erasureServerPools).Walk in goroutine 1189
/Users/kp/code/src/github.com/minio/minio/cmd/erasure-server-pool.go:1985 +0x228
```
Object names of directory objects qualified for ExpiredObjectAllVersions
must be encoded appropriately before calling on deletePrefix on their
erasure set.
e.g., a directory object and regular objects with overlapping prefixes
could lead to the expiration of regular objects, which is not the
intention of ILM.
```
bucket/dir/ ---> directory object
bucket/dir/obj-1
```
When `bucket/dir/` qualifies for expiration, the current implementation would
remove regular objects under the prefix `bucket/dir/`, in this case,
`bucket/dir/obj-1`.
In handlers related to health diagnostics e.g. CPU, Network, Partitions,
etc, globalMinioHost was being passed as the addr, resulting in empty
value for the same in the health report.
Using globalLocalNodeName instead fixes the issue.
IAM loading is a lazy operation, allow these
fallbacks to be in place when we cannot find
in-memory state().
this allows us to honor the request even if pay
a small price for lookup and populating the data.
When objects have more versions than their ILM policy expects to retain
via NewerNoncurrentVersions, but they don't qualify for expiry due to
NoncurrentDays are configured in that rule.
In this case, applyNewerNoncurrentVersionsLimit method was enqueuing empty
tasks, which lead to a panic (panic: runtime error: index out of range [0] with
length 0) in newerNoncurrentTask.OpHash method, which assumes the task
to contain at least one version to expire.
When returning the status of a decommissioned pool, a pool with zero
time StartedTime will be considered an active pool, which is unexpected.
This commit will always ensure that a pool's canceled/failed/completed
status is returned.
we were prematurely not writing 4k pages while we
could have due to the fact that most buffers would
be multiples of 4k upto some number and there shall
be some remainder.
We only need to write the remainder without O_DIRECT.
at scale customers might start with failed drives,
causing skew in the overall usage ratio per EC set.
make this configurable such that customers can turn
this off as needed depending on how comfortable they
are.
Currently, the code relies on object parity to decide whether it is a
delete marker or a regular object. In the case of a delete marker, the
return quorum is half of the disks in the erasure set. However, this
calculation must be corrected with objects with EC = 0, mainly
because EC is not a one-time fixed configuration.
Though all data are correct, the manifested symptom is a 503 with an
EC=0 object. This bug was manifested after we introduced the
fast Get Object feature that does not read all data from all disks in
case of inlined objects
Metrics v3 is mainly a reorganization of metrics into smaller groups of
metrics and the removal of internal aggregation of metrics received from
peer nodes in a MinIO cluster.
This change adds the endpoint `/minio/metrics/v3` as the top-level metrics
endpoint and under this, various sub-endpoints are implemented. These
are currently documented in `docs/metrics/v3.md`
The handler will serve metrics at any path
`/minio/metrics/v3/PATH`, as follows:
when PATH is a sub-endpoint listed above => serves the group of
metrics under that path; or when PATH is a (non-empty) parent
directory of the sub-endpoints listed above => serves metrics
from each child sub-endpoint of PATH. otherwise, returns a no
resource found error
All available metrics are listed in the `docs/metrics/v3.md`. More will
be added subsequently.
Merging same-object - multiple versions from different pools would not always result in correct ordering.
When merging keep inputs separate.
```
λ mc ls --versions local/testbucket
------ before ------
[2024-03-05 20:17:19 CET] 228B STANDARD 1f163718-9bc5-4b01-bff7-5d8cf09caf10 v3 PUT hosts
[2024-03-05 20:19:56 CET] 19KiB STANDARD null v2 PUT hosts
[2024-03-05 20:17:15 CET] 228B STANDARD 73c9f651-f023-4566-b012-cc537fdb7ce2 v1 PUT hosts
------ after ------
λ mc ls --versions local/testbucket
[2024-03-05 20:19:56 CET] 19KiB STANDARD null v3 PUT hosts
[2024-03-05 20:17:19 CET] 228B STANDARD 1f163718-9bc5-4b01-bff7-5d8cf09caf10 v2 PUT hosts
[2024-03-05 20:17:15 CET] 228B STANDARD 73c9f651-f023-4566-b012-cc537fdb7ce2 v1 PUT hosts
```
Currently, the progress of the batch job is saved in inside the job
request object, which is normally not supported by MinIO. Though there
is no apparent bug, it is better to fix this now.
Batch progress is saved in .minio.sys/batch-jobs/reports/
Co-authored-by: Anis Eleuch <anis@min.io>
our PoolNumber calculation was costly,
while we already had this information per
endpoint, we needed to deduce it appropriately.
This PR addresses this by assigning PoolNumbers
field that carries all the pool numbers that
belong to a server.
properties.PoolNumber still carries a valid value
only when len(properties.PoolNumbers) == 1, otherwise
properties.PoolNumber is set to math.MaxInt (indicating
that this value is undefined) and then one must rely
on properties.PoolNumbers for server participation
in multiple pools.
addresses the issue originating from #11327
there can be a sudden spike in tiny allocations,
due to too much auditing being done, also don't hang
on the
```
h.logCh <- entry
```
after initializing workers if you do not have a way to
dequeue for some reason.
This commits adds support for using the `--endpoint` arg when creating a
tier of type `azure`. This is needed to connect to Azure's Gov Cloud
instance. For example,
```
mc ilm tier add azure TARGET TIER_NAME \
--account-name ACCOUNT \
--account-key KEY \
--bucket CONTAINER \
--endpoint https://ACCOUNT.blob.core.usgovcloudapi.net
--prefix PREFIX \
--storage-class STORAGE_CLASS
```
Prior to this, the endpoint was hardcoded to `https://ACCOUNT.blob.core.windows.net`.
The docs were even explicit about this, stating that `--endpoint` is:
"Required for `s3` or `minio` tier types. This option has no effect for any
other value of `TIER_TYPE`."
Now, if the endpoint arg is present it will be used. If not, it will
fall back to the same default behavior of `ACCOUNT.blob.core.windows.net`.
Remove api.expiration_workers config setting which was inadvertently left behind. Per review comment
https://github.com/minio/minio/pull/18926, expiration_workers can be configured via ilm.expiration_workers.
ext4, xfs support this behavior however
btrfs, nfs may not support it properly.
in-case when we see Nlink < 2 then we know
that we need to fallback on readdir()
fixes a regression from #19100fixes#19181
The middleware sets up tracing, throttling, gzipped responses and
collecting API stats.
Additionally, this change updates the names of handler functions in
metric labels to be the same as the name derived from Go lang reflection
on the handler name.
The metric api labels are now stored in memory the same as the handler
name - they will be camelcased, e.g. `GetObject` instead of `getobject`.
For compatibility, we lowercase the metric api label values when emitting the metrics.
- Use a shared worker pool for all ILM expiry tasks
- Free version cleanup executes in a separate goroutine
- Add a free version only if removing the remote object fails
- Add ILM expiry metrics to the node namespace
- Move tier journal tasks to expiryState
- Remove unused on-disk journal for tiered objects pending deletion
- Distribute expiry tasks across workers such that the expiry of versions of
the same object serialized
- Ability to resize worker pool without server restart
- Make scaling down of expiryState workers' concurrency safe; Thanks
@klauspost
- Add error logs when expiryState and transition state are not
initialized (yet)
* metrics: Add missed tier journal entry tasks
* Initialize the ILM worker pool after the object layer
With this commit, MinIO generates root credentials automatically
and deterministically if:
- No root credentials have been set.
- A KMS (KES) is configured.
- API access for the root credentials is disabled (lockdown mode).
Before, MinIO defaults to `minioadmin` for both the access and
secret keys. Now, MinIO generates unique root credentials
automatically on startup using the KMS.
Therefore, it uses the KMS HMAC function to generate pseudo-random
values. These values never change as long as the KMS key remains
the same, and the KMS key must continue to exist since all IAM data
is encrypted with it.
Backward compatibility:
This commit should not cause existing deployments to break. It only
changes the root credentials of deployments that have a KMS configured
(KES, not a static key) but have not set any admin credentials. Such
implementations should be rare or not exist at all.
Even if the worst case would be updating root credentials in mc
or other clients used to administer the cluster. Root credentials
are anyway not intended for regular S3 operations.
Signed-off-by: Andreas Auernhammer <github@aead.dev>
just like client-conn-read-deadline, added a new flag that does
client-conn-write-deadline as well.
Both are not configured by default, since we do not yet know
what is the right value. Allow this to be configurable if needed.
we should do this to ensure that we focus on
data healing as primary focus, fixing metadata
as part of healing must be done but making
data available is the main focus.
the main reason is metadata inconsistencies can
cause data availability issues, which must be
avoided at all cost.
will be bringing in an additional healing mechanism
that involves "metadata-only" heal, for now we do
not expect to have these checks.
continuation of #19154
Bonus: add a pro-active healthcheck to perform a connection
This change makes the label names consistent with the handler names.
This is in preparation to use reflection based API handler function
names for the api labels so they will be the same as tracing, auditing
and logging names for these API calls.
in k8s things really do come online very asynchronously,
we need to use implementation that allows this randomness.
To facilitate this move WriteAll() as part of the
websocket layer instead.
Bonus: avoid instances of dnscache usage on k8s
New disk healing code skips/expires objects that ILM supposed to expire.
Add more visibility to the user about this activity by calculating those
objects and print it at the end of healing activity.
This PR fixes a bug that perhaps has been long introduced,
with no visible workarounds. In any deployment, if an entire
erasure set is deleted, there is no way the cluster recovers.
This change is to decouple need for root credentials to match between
site replication deployments.
Also ensuring site replication config initialization is re-tried until
it succeeds, this deoendency is critical to STS flow in site replication
scenario.
Currently, we read from `/proc/diskstats` which is found to be
un-reliable in k8s environments. We can read from `sysfs` instead.
Also, cache the latest drive io stats to find the diff and update
the metrics.
* Remove lock for cached operations.
* Rename "Relax" to `ReturnLastGood`.
* Add `CacheError` to allow caching values even on errors.
* Add NoWait that will return current value with async fetching if within 2xTTL.
* Make benchmark somewhat representative.
```
Before: BenchmarkCache-12 16408370 63.12 ns/op 0 B/op
After: BenchmarkCache-12 428282187 2.789 ns/op 0 B/op
```
* Remove `storageRESTClient.scanning`. Nonsensical - RPC clients will not have any idea about scanning.
* Always fetch remote diskinfo metrics and cache them. Seems most calls are requesting metrics.
* Do async fetching of usage caches.
It also fixes a long-standing bug in expiring transitioned objects.
The expiration action was deleting the current version in the case'
of tiered objects instead of adding a delete marker.
only enable md5sum if explicitly asked by the client, otherwise
its not necessary to compute md5sum when SSE-KMS/SSE-C is enabled.
this is continuation of #17958
data-dir not being present is okay, however we can still
rely on the `rename()` atomic call instead of relying on
write xl.meta write which may truncate the io.EOF.
Add a new function logger.Event() to send the log to Console and
http/kafka log webhooks. This will include some internal events such as
disk healing and rebalance/decommissioning
the PR in #16541 was incorrect and hand wrong assumptions
about the overall setup, revert this since this expectation
to have offline servers is wrong and we can end up with a
bigger chicken and egg problem.
This reverts commit 5996c8c4d5.
Bonus:
- preserve disk in globalLocalDrives properly upon connectDisks()
- do not return 'nil' from newXLStorage(), getting it ready for
the next set of changes for 'format.json' loading.
The previous logic of calculating per second values for disk io stats
divides the stats by the host uptime. This doesn't work in k8s
environment as the uptime is of the pod, but the stats (from
/proc/diskstats) are from the host.
Fix this by storing the initial values of uptime and the stats at the
timme of server startup, and using the difference between current and
initial values when calculating the per second values.
globalLocalDrives seem to be not updated during the
HealFormat() leads to a requirement where the server
needs to be restarted for the healing to continue.
a/prefix
a/prefix/1.txt
where `a/prefix` is an object which does not have `/` at the end,
we do not have to aggressively recursively delete all the sub-folders
as well. Instead convert the call into self contained to deleting
'xl.meta' and then subsequently attempting to Remove the parent.
Bonus: enable audit alerts for object versions
beyond the configured value, default is '100'
versions per object beyond which scanner will
alert for each such objects.
when we expand via pools, there is no reason to stick
with the same distributionAlgo as the rest. Since the
algo only makes sense with-in a pool not across pools.
This allows for newer pools to use newer codepaths to
avoid legacy file lookups when they have a pre-existing
deployment from 2019, they can expand their new pool
to be of a newer distribution format, allowing the
pool to be more performant.
- bucket metadata does not need to look for legacy things
anymore if b.Created is non-zero
- stagger bucket metadata loads across lots of nodes to
avoid the current thundering herd problem.
- Remove deadlines for RenameData, RenameFile - these
calls should not ever be timed out and should wait
until completion or wait for client timeout. Do not
choose timeouts for applications during the WRITE phase.
- increase R/W buffer size, increase maxMergeMessages to 30
Depending on when the context cancelation is picked up the handler may return and close the channel before `SubscribeJSON` returns, causing:
```
Feb 05 17:12:00 s3-us-node11 minio[3973657]: panic: send on closed channel
Feb 05 17:12:00 s3-us-node11 minio[3973657]: goroutine 378007076 [running]:
Feb 05 17:12:00 s3-us-node11 minio[3973657]: github.com/minio/minio/internal/pubsub.(*PubSub[...]).SubscribeJSON.func1()
Feb 05 17:12:00 s3-us-node11 minio[3973657]: github.com/minio/minio/internal/pubsub/pubsub.go:139 +0x12d
Feb 05 17:12:00 s3-us-node11 minio[3973657]: created by github.com/minio/minio/internal/pubsub.(*PubSub[...]).SubscribeJSON in goroutine 378010884
Feb 05 17:12:00 s3-us-node11 minio[3973657]: github.com/minio/minio/internal/pubsub/pubsub.go:124 +0x352
```
Wait explicitly for the goroutine to exit.
Bonus: Listen for doneCh when sending to not risk getting blocked there is channel isn't being emptied.
this fixes rare bugs we have seen but never really found a
reproducer for
- PutObjectRetention() returning 503s
- PutObjectTags() returning 503s
- PutObjectMetadata() updates during replication returning 503s
These calls return errors, and this perpetuates with
no apparent fix.
This PR fixes with correct quorum requirement.
To force limit the duration of STS accounts, the user can create a new
policy, like the following:
{
"Version": "2012-10-17",
"Statement": [{
"Effect": "Allow",
"Action": ["sts:AssumeRoleWithWebIdentity"],
"Condition": {"NumericLessThanEquals": {"sts:DurationSeconds": "300"}}
}]
}
And force binding the policy to all OpenID users, whether using a claim name or role
ARN.
disk tokens usage is not necessary anymore with the implementation
of deadlines for storage calls and active monitoring of the drive
for I/O timeouts.
Functionality kicking off a bad drive is still supported, it's just that
we do not have to serialize I/O in the manner tokens would do.
Interpret `null` inline policy for access keys as inheriting parent
policy. Since MinIO Console currently sends this value, we need to honor it
for now. A larger fix in Console and in the server are required.
Fixes#18939.
Allow internal types to support a `Recycler` interface, which will allow for sharing of common types across handlers.
This means that all `grid.MSS` (and similar) objects are shared across in a common pool instead of a per-handler pool.
Add internal request reuse of internal types. Add for safe (pointerless) types explicitly.
Only log params for internal types. Doing Sprint(obj) is just a bit too messy.
With this change, only a user with `UpdateServiceAccountAdminAction`
permission is able to edit access keys.
We would like to let a user edit their own access keys, however the
feature needs to be re-designed for better security and integration with
external systems like AD/LDAP and OpenID.
This change prevents privilege escalation via service accounts.
for actionable, inspections we have `mc support inspect`
we do not need double logging, healing will report relevant
errors if any, in terms of quorum lost etc.
Each Put, List, Multipart operations heavily rely on making
GetBucketInfo() call to verify if bucket exists or not on
a regular basis. This has a large performance cost when there
are tons of servers involved.
We did optimize this part by vectorizing the bucket calls,
however its not enough, beyond 100 nodes and this becomes
fairly visible in terms of performance.
- healing must not set the write xattr
because that is the job of active healing
to update. what we need to preserve is
permanent deletes.
- remove older env for drive monitoring and
enable it accordingly, as a global value.
local disk metrics were polluting cluster metrics
Please remove them instead of adding relevant ones.
- batch job metrics were incorrectly kept at bucket
metrics endpoint, move it to cluster metrics.
- add tier metrics to cluster peer metrics from the node.
- fix missing set level cluster health metrics
it is entirely possible that a rebalance process which was running
when it was asked to "stop" it failed to write its last statistics
to the disk.
After this a pool expansion can cause disruption and all S3 API
calls would fail at IsPoolRebalancing() function.
This PRs makes sure that we update rebalance.bin under such
conditions to avoid any runtime crashes.
add new update v2 that updates per node, allows idempotent behavior
new API ensures that
- binary is correct and can be downloaded checksummed verified
- committed to actual path
- restart returns back the relevant waiting drives
do not need to be defensive in our approach,
we should simply override anything everything
in import process, do not care about what
currently exists on the disk - backup is the
source of truth.
Right now the format.json is excluded if anything within `.minio.sys` is requested.
I assume the check was meant to exclude only if it was actually requesting it.
- Move RenameFile to websockets
- Move ReadAll that is primarily is used
for reading 'format.json' to to websockets
- Optimize DiskInfo calls, and provide a way
to make a NoOp DiskInfo call.
AlmosAll uses of NewDeadlineWorker, which relied on secondary values, were used in a racy fashion,
which could lead to inconsistent errors/data being returned. It also propagates the deadline downstream.
Rewrite all these to use a generic WithDeadline caller that can return an error alongside a value.
Remove the stateful aspect of DeadlineWorker - it was racy if used - but it wasn't AFAICT.
Fixes races like:
```
WARNING: DATA RACE
Read at 0x00c130b29d10 by goroutine 470237:
github.com/minio/minio/cmd.(*xlStorageDiskIDCheck).ReadVersion()
github.com/minio/minio/cmd/xl-storage-disk-id-check.go:702 +0x611
github.com/minio/minio/cmd.readFileInfo()
github.com/minio/minio/cmd/erasure-metadata-utils.go:160 +0x122
github.com/minio/minio/cmd.erasureObjects.getObjectFileInfo.func1.1()
github.com/minio/minio/cmd/erasure-object.go:809 +0x27a
github.com/minio/minio/cmd.erasureObjects.getObjectFileInfo.func1.2()
github.com/minio/minio/cmd/erasure-object.go:828 +0x61
Previous write at 0x00c130b29d10 by goroutine 470298:
github.com/minio/minio/cmd.(*xlStorageDiskIDCheck).ReadVersion.func1()
github.com/minio/minio/cmd/xl-storage-disk-id-check.go:698 +0x244
github.com/minio/minio/internal/ioutil.(*DeadlineWorker).Run.func1()
github.com/minio/minio/internal/ioutil/ioutil.go:141 +0x33
WARNING: DATA RACE
Write at 0x00c0ba6e6c00 by goroutine 94507:
github.com/minio/minio/cmd.(*xlStorageDiskIDCheck).StatVol.func1()
github.com/minio/minio/cmd/xl-storage-disk-id-check.go:419 +0x104
github.com/minio/minio/internal/ioutil.(*DeadlineWorker).Run.func1()
github.com/minio/minio/internal/ioutil/ioutil.go:141 +0x33
Previous read at 0x00c0ba6e6c00 by goroutine 94463:
github.com/minio/minio/cmd.(*xlStorageDiskIDCheck).StatVol()
github.com/minio/minio/cmd/xl-storage-disk-id-check.go:422 +0x47e
github.com/minio/minio/cmd.getBucketInfoLocal.func1()
github.com/minio/minio/cmd/peer-s3-server.go:275 +0x122
github.com/minio/pkg/v2/sync/errgroup.(*Group).Go.func1()
```
Probably back from #17701
protection was in place. However, it covered only some
areas, so we re-arranged the code to ensure we could hold
locks properly.
Along with this, remove the DataShardFix code altogether,
in deployments with many drive replacements, this can affect
and lead to quorum loss.
Also limit the amount of concurrency when sending
binary updates to peers, avoid high network over
TX that can cause disconnection events for the
node sending updates.
If site replication is enabled, we should still show the size and
version distribution histogram metrics at bucket level.
Signed-off-by: Shubhendu Ram Tripathi <shubhendu@minio.io>
New API now verifies any hung disks before restart/stop,
provides a 'per node' break down of the restart/stop results.
Provides also how many blocked syscalls are present on the
drives and what users must do about them.
Adds options to do pre-flight checks to provide information
to the user regarding any hung disks. Provides 'force' option
to forcibly attempt a restart() even with waiting syscalls
on the drives.
On a policy detach operation, if there are no policies remaining
attached to the user/group, remove the policy mapping file, instead of
leaving a file containing an empty list of policies.
Healing dangling buckets is conservative, and it is a typical use case to
fail to remove a dangling bucket because it contains some data because
healing danging bucket code is not allowed to remove data: only healing
the dangling object is allowed to do so.
reference format is constant for any lifetime of
a minio cluster, we do not have to ever replace
it during HealFormat() as it will never change.
additionally we should simply reject reference
formats that we do not understand early on.
GetActualSize() was heavily relying on o.Parts()
to be non-empty to figure out if the object is multipart or not,
However, we have many indicators of whether an object is multipart
or not.
Blindly assuming that o.Parts == nil is not a multipart, is an
incorrect expectation instead, multipart must be obtained via
- Stored metadata value indicating this is a multipart encrypted object.
- Rely on <meta>-actual-size metadata to get the object's actual size.
This value is preserved for additional reasons such as these.
- ETag != 32 length
support proxying of tagging requests in active-active replication
Note: even if proxying is successful, PutObjectTagging/DeleteObjectTagging
will continue to report a 404 since the object is not present locally.
New intervals:
[1024B, 64KiB)
[64KiB, 256KiB)
[256KiB, 512KiB)
[512KiB, 1MiB)
The new intervals helps us see object size distribution with higher
resolution for the interval [1024B, 1MiB).
- HealFormat() was leaking healthcheck goroutines for
disks, we are only interested in enabling healthcheck
for the newly formatted disk, not for existing disks.
- When disk is a root-disk a random disk monitor was
leaking while we ignored the drive.
- When loading the disk for each erasure set, we were
leaking goroutines for the prepare-storage.go disks
which were replaced via the globalLocalDrives slice
- avoid disk monitoring utilizing health tokens that
would cause exhaustion in the tokens, prematurely
which were meant for incoming I/O. This is ensured
by avoiding writing O_DIRECT aligned buffer instead
write 2048 worth of content only as O_DSYNC, which is
sufficient.
Add a hidden configuration under the scanner sub section to configure if
the scanner should sleep between two objects scan. The configuration has
only effect when there is no drive activity related to s3 requests or
healing.
By default, the code will keep the current behavior which is doing
sleep between objects.
To forcefully enable the full scan speed in idle mode, you can do this:
`mc admin config set myminio scanner idle_speed=full`
fixes#18724
A regression was introduced in #18547, that attempted
to file adding a missing `null` marker however we
should not skip returning based on versionID instead
it must be based on if we are being asked to create
a DEL marker or not.
The PR also has a side-affect for replicating `null`
marker permanent delete, as it may end up adding a
`null` marker while removing one.
This PR should address both scenarios.
NOTE: This feature is not retro-active; it will not cater to previous transactions
on existing setups.
To enable this feature, please set ` _MINIO_DRIVE_QUORUM=on` environment
variable as part of systemd service or k8s configmap.
Once this has been enabled, you need to also set `list_quorum`.
```
~ mc admin config set alias/ api list_quorum=auto`
```
A new debugging tool is available to check for any missing counters.
Following policies if present
```
"Condition": {
"IpAddress": {
"aws:SourceIp": [
"54.240.143.0/24",
"2001:DB8:1234:5678::/64"
]
}
}
```
And client is making a request to MinIO via IPv6 can
potentially crash the server.
Workarounds are turn-off IPv6 and use only IPv4
This PR also increases per node bpool memory from 1024 entries
to 2048 entries; along with that, it also moves the byte pool
centrally instead of being per pool.
minio_node_tier_ttlb_seconds - Distribution of time to last byte for streaming objects from warm tier
minio_node_tier_requests_success - Number of requests to download object from warm tier that were successful
minio_node_tier_requests_failure - Number of requests to download object from warm tier that failed
SUBNET now has a v2 of license that is returned in the new key
`license_v2`. mc will start reading and storing the same. (The old key
`license` is deprecated but is still available in SUBNET response to
ensure that the current released version of minio doesn't break)
`(*xlStorageDiskIDCheck).CreateFile` wraps the incoming reader in `xioutil.NewDeadlineReader`.
The wrapped reader is handed to `(*xlStorage).CreateFile`. This performs a Read call via `writeAllDirect`,
which reads into an `ODirectPool` buffer.
`(*DeadlineReader).Read` spawns an async read into the buffer. If a timeout is hit while reading,
the read operation returns to `writeAllDirect`. The operation returns an error and the buffer is reused.
However, if the async `Read` call unblocks, it will write to the now recycled buffer.
Fix: Remove the `DeadlineReader` - it is inherently unsafe. Instead, rely on the network timeouts.
This is not a disk timeout, anyway.
Regression in https://github.com/minio/minio/pull/17745
This patch adds the targetID to the existing notification target metrics
and deprecates the current target metrics which points to the overall
event notification subsystem
historically, we have always kept storage-rest-server
and a local storage API separate without much trouble,
since they both can independently operate due to no
special state() between them.
however, over some time, we have added state()
such as
- drive monitoring threads now there will be "2" of
them per drive instead of just 1.
- concurrent tokens available per drive are now twice
instead of just single shared, allowing unexpectedly
high amount of I/O to go through.
- applying serialization by using walkMutexes can now
be adequately honored for both remote callers and local
callers.
Regression from #18285. CopyObject options were inheriting source MTime
for metadata timestamps if unspecified, removing this prevented metadata
updates from being applied on target.
By default the cpu load is the cumulative of all cores. Capture the
percentage load (load * 100 / cpu-count)
Also capture the percentage memory used (used * 100 / total)
use memory for async events when necessary and dequeue them as
needed, for all synchronous events customers must enable
```
MINIO_API_SYNC_EVENTS=on
```
Async events can be lost but is upto to the admin to
decide what they want, we will not create run-away number
of goroutines per event instead we will queue them properly.
Currently the max async workers is set to runtime.GOMAXPROCS(0)
which is more than sufficient in general, but it can be made
configurable in future but may not be needed.
there is potential for danglingWrites when quorum failed, where
only some drives took a successful write, generally this is left
to the healing routine to pick it up. However it is better that
we delete it right away to avoid potential for quorum issues on
version signature when there are many versions of an object.
it is okay if the warm-tier cannot keep up, we should continue
to take I/O at hot-tier, only fail hot-tier or block it when
we are disk full.
Bonus: add metrics counter for these missed tasks, we will
know for sure if one of the node is lagging behind or is
losing too many tasks during transitioning.
A disk that is not able to initialize when an instance is started
will never have a handler registered, which means a user will
need to restart the node after fixing the disk;
This will also prevent showing the wrong 'upgrade is needed.'
error message in that case.
When the disk is still failing, print an error every 30 minutes;
Disk reconnection will be retried every 30 seconds.
Co-authored-by: Anis Elleuch <anis@min.io>
`OpMuxConnectError` was not handled correctly.
Remove local checks for single request handlers so they can
run before being registered locally.
Bonus: Only log IAM bootstrap on startup.
While healing the latest changes of expiry rules across sites
if target had pre existing transition rules, they were getting
overwritten as cloned latest expiry rules from remote site were
getting written as is. Fixed the same and added test cases as
well.
Signed-off-by: Shubhendu Ram Tripathi <shubhendu@minio.io>
moveToTrash() function moves a folder to .trash, for example, when
doing some object deletions: a data dir that has many parts will be
renamed to the trash folder; However, ENOSPC is a valid error from
rename(), and it can cripple a user trying to free some space in an
entire disk situation.
Therefore, this commit will try to do a recursive delete in that case.
This allows batch replication to basically do not
attempt to copy objects that do not have read quorum.
This PR also allows walk() to provide custom
values for quorum under batch replication, and
key rotation.
this PR allows following policy
```
{
"Version": "2012-10-17",
"Statement": [
{
"Sid": "Deny a presigned URL request if the signature is more than 10 min old",
"Effect": "Deny",
"Action": "s3:*",
"Resource": "arn:aws:s3:::DOC-EXAMPLE-BUCKET1/*",
"Condition": {
"NumericGreaterThan": {
"s3:signatureAge": 600000
}
}
}
]
}
```
This is to basically disable all pre-signed URLs that are older than 10 minutes.
AWS S3 closes keep-alive connections frequently
leading to frivolous logs filling up the MinIO
logs when the transition tier is an AWS S3 bucket.
Ignore such transient errors, let MinIO retry
it when it can.
When minio runs with MINIO_CI_CD=on, it is expected to communicate
with the locally running SUBNET. This is happening in the case of MinIO
via call home functionality. However, the subnet-related functionality inside the
console continues to talk to the SUBNET production URL. Because of this,
the console cannot be tested with a locally running SUBNET.
Set the env variable CONSOLE_SUBNET_URL correctly in such cases.
(The console already has code to use the value of this variable
as the subnet URL)
Optionally allows customers to enable
- Enable an external cache to catch GET/HEAD responses
- Enable skipping disks that are slow to respond in GET/HEAD
when we have already achieved a quorum
Bonus: allow replication to attempt Deletes/Puts when
the remote returns quorum errors of some kind, this is
to ensure that MinIO can rewrite the namespace with the
latest version that exists on the source.
This PR adds a WebSocket grid feature that allows servers to communicate via
a single two-way connection.
There are two request types:
* Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small
roundtrips with small payloads.
* Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`,
which allows for different combinations of full two-way streams with an initial payload.
Only a single stream is created between two machines - and there is, as such, no
server/client relation since both sides can initiate and handle requests. Which server
initiates the request is decided deterministically on the server names.
Requests are made through a mux client and server, which handles message
passing, congestion, cancelation, timeouts, etc.
If a connection is lost, all requests are canceled, and the calling server will try
to reconnect. Registered handlers can operate directly on byte
slices or use a higher-level generics abstraction.
There is no versioning of handlers/clients, and incompatible changes should
be handled by adding new handlers.
The request path can be changed to a new one for any protocol changes.
First, all servers create a "Manager." The manager must know its address
as well as all remote addresses. This will manage all connections.
To get a connection to any remote, ask the manager to provide it given
the remote address using.
```
func (m *Manager) Connection(host string) *Connection
```
All serverside handlers must also be registered on the manager. This will
make sure that all incoming requests are served. The number of in-flight
requests and responses must also be given for streaming requests.
The "Connection" returned manages the mux-clients. Requests issued
to the connection will be sent to the remote.
* `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)`
performs a single request and returns the result. Any deadline provided on the request is
forwarded to the server, and canceling the context will make the function return at once.
* `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)`
will initiate a remote call and send the initial payload.
```Go
// A Stream is a two-way stream.
// All responses *must* be read by the caller.
// If the call is canceled through the context,
//The appropriate error will be returned.
type Stream struct {
// Responses from the remote server.
// Channel will be closed after an error or when the remote closes.
// All responses *must* be read by the caller until either an error is returned or the channel is closed.
// Canceling the context will cause the context cancellation error to be returned.
Responses <-chan Response
// Requests sent to the server.
// If the handler is defined with 0 incoming capacity this will be nil.
// Channel *must* be closed to signal the end of the stream.
// If the request context is canceled, the stream will no longer process requests.
Requests chan<- []byte
}
type Response struct {
Msg []byte
Err error
}
```
There are generic versions of the server/client handlers that allow the use of type
safe implementations for data types that support msgpack marshal/unmarshal.
With an odd number of drives per erasure set setup, the write/quorum is
the half + 1; however the decommissioning listing will still list those
objects and does not consider those as stale.
Fix it by using (N+1)/2 formula.
Co-authored-by: Anis Elleuch <anis@min.io>
Immediate transition use case and is mostly used to fill warm
backend with a lot of data when a new deployment is created
Currently, if the transition queue is complete, the transition will be
deferred to the scanner; change this behavior by blocking the PUT request
until the transition queue has a new place for a transition task.
Currently if the object does not exist in quorum disks of an erasure
set, the dangling code is never called because the returned error will
be errFileNotFound or errFileVersionNotFound;
With this commit, when errFileNotFound or errFileVersionNotFound is
returning when trying to calculate the quorum of a given object, the
code checks if a disk returned nil, which means a stale object exists in
that disk, that will trigger deleteIfDangling() function
This commit splits the liveness and readiness
handler into two separate handlers. In K8S, a
liveness probe is used to determine whether the
pod is in "live" state and functioning at all.
In contrast, the readiness probe is used to
determine whether the pod is ready to serve
requests.
A failing liveness probe causes pod restarts while
a failing readiness probe causes k8s to stop routing
traffic to the pod. Hence, a liveness probe should
be as robust as possible while a readiness probe
should be used to load balancing.
Ref: https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/
Signed-off-by: Andreas Auernhammer <github@aead.dev>
This patch takes care of loading the bucket configs of failed buckets
during the periodic refresh. This makes sure the event notifiers and
remote bucket targets are properly initialized.
users might use MinIO on NFS, GPFS that provide dynamic
inodes and may not even have a concept of free inodes.
to allow users to use MinIO on top of GPFS relax the
free inode check.
* creating a byte buffer for SFTP file segments
* Adding an error condition for when there are
remaining segments in the queue
* Simplification of the queue using a map
it is possible that ILM or Deletes got triggered on batch
of objects that we are attempting to batch replicate, ignore
this scenario as valid behavior.
sendfile implementation to perform DMA on all platforms
Go stdlib already supports sendfile/splice implementations
for
- Linux
- Windows
- *BSD
- Solaris
Along with this change however O_DIRECT for reads() must be
removed as well since we need to use sendfile() implementation
The main reason to add O_DIRECT for reads was to reduce the
chances of page-cache causing OOMs for MinIO, however it would
seem that avoiding buffer copies from user-space to kernel space
this issue is not a problem anymore.
There is no Go based memory allocation required, and neither
the page-cache is referenced back to MinIO. This page-
cache reference is fully owned by kernel at this point, this
essentially should solve the problem of page-cache build up.
With this now we also support SG - when NIC supports Scatter/Gather
https://en.wikipedia.org/wiki/Gather/scatter_(vector_addressing)
`monitorAndConnectEndpoints` will continue to attempt to reconnect offline disks.
Since disks were never closed, a `MarkOffline` would continue to try to check these disks forever.
Close previous disks.
