Commit Graph

22 Commits

Author SHA1 Message Date
Poorna ca2a1c3f60
replication: clone metrics while loading metrics cache (#16462) 2023-01-24 02:10:32 -08:00
Harshavardhana a5f8af4efb
serialize replication stats() only when needed (#16280) 2022-12-20 00:07:53 -08:00
Klaus Post 2894dd4d1a
fix: hold lock while serializing replication stats (#16007) 2022-11-04 09:59:14 -07:00
Harshavardhana 94dbb4a427
fix: generalize SC config and also skip healing sub-sys under SD (#15757) 2022-09-26 09:04:54 -07:00
Minio Trusted d89f6af6c4 avoid replication stats crash in Prometheus 2022-09-16 17:09:45 -07:00
Poorna 6b9fd256e1
Persist in-memory replication stats to disk (#15594)
to avoid relying on scanner-calculated replication metrics.
This will improve the accuracy of the replication stats reported.

This PR also adds on to #15556 by handing replication
traffic that could not be queued by available workers to the 
MRF queue so that entries in `PENDING` status are healed faster.
2022-09-12 12:40:02 -07:00
Harshavardhana f8650a3493
fetch bucket replication stats across peers in single call (#14956)
current implementation relied on recursively calling one bucket
at a time across all peers, this would be very slow and chatty
when there are 100's of buckets which would mean 100*peerCount
amount of network operations.

This PR attempts to reduce this entire call into `peerCount`
amount of network calls only. This functionality addresses also a
concern where the Prometheus metrics would significantly slow
down when one of the peers is offline.
2022-05-23 09:15:30 -07:00
Aditya Manthramurthy 9aadd725d2
Avoid calling .Reset() on active timer (#14941)
.Reset() documentation states:

    For a Timer created with NewTimer, Reset should be invoked only on stopped
    or expired timers with drained channels.

This change is just to comply with this requirement as there might be some
runtime dependent situation that might lead to unexpected behavior.
2022-05-18 15:37:58 -07:00
Harshavardhana 6cfb1cb6fd
fix: timer usage across codebase (#14935)
it seems in some places we have been wrongly using the
timer.Reset() function, nicely exposed by an example
shared by @donatello https://go.dev/play/p/qoF71_D1oXD

this PR fixes all the usage comprehensively
2022-05-17 22:42:59 -07:00
Harshavardhana f527c708f2
run gofumpt cleanup across code-base (#14015) 2022-01-02 09:15:06 -08:00
Aditya Manthramurthy 997e808088
fix; race in bucket replication stats (#13942)
- r.ulock was not locked when r.UsageCache was being modified

Bonus:

- simplify code by removing some unnecessary clone methods - we can 
do this because go arrays are values (not pointers/references) that are 
automatically copied on assignment.

- remove some unnecessary map allocation calls
2021-12-17 15:33:13 -08:00
Anis Elleuch 4caed7cc0d
metrics: Add replication latency metrics (#13515)
Add a new Prometheus metric for bucket replication latency

e.g.:
minio_bucket_replication_latency_ns{
    bucket="testbucket",
    operation="upload",
    range="LESS_THAN_1_MiB",
    server="127.0.0.1:9001",
    targetArn="arn:minio:replication::45da043c-14f5-4da4-9316-aba5f77bf730:testbucket"} 2.2015663e+07

Co-authored-by: Klaus Post <klauspost@gmail.com>
2021-11-17 12:10:57 -08:00
Poorna K e7f559c582
Fixes to replication metrics (#13493)
For reporting ReplicaSize and loading initial
replication metrics correctly.
2021-10-21 18:52:55 -07:00
Poorna Krishnamoorthy c4373ef290
Add support for multi site replication (#12880) 2021-09-18 13:31:35 -07:00
Harshavardhana 1f262daf6f
rename all remaining packages to internal/ (#12418)
This is to ensure that there are no projects
that try to import `minio/minio/pkg` into
their own repo. Any such common packages should
go to `https://github.com/minio/pkg`
2021-06-01 14:59:40 -07:00
Poorna Krishnamoorthy 3690de0c6b
Drop Pending size and count from replication metrics (#12378)
Real-time metrics calculated in-memory rely on the initial
replication metrics saved with data usage. However, this can
lag behind the actual state of the cluster at the time of server 
restart leading to inaccurate Pending size/counts reported to
Prometheus. Dropping the Pending metrics as this can be more 
reliably monitored by applications with replication notifications.

Signed-off-by: Poorna Krishnamoorthy <poorna@minio.io>
2021-05-31 20:26:52 -07:00
Harshavardhana 82dc6aff1c
add support for configurable replication MRF workers (#12125)
just like replication workers, allow failed replication
workers to be configurable in situations like DR failures
etc to catch up on replication sooner when DR is back
online.

Signed-off-by: Harshavardhana <harsha@minio.io>
2021-04-23 21:58:45 -07:00
Harshavardhana 069432566f update license change for MinIO
Signed-off-by: Harshavardhana <harsha@minio.io>
2021-04-23 11:58:53 -07:00
Poorna Krishnamoorthy 40409437cd
Add initial usage in GetBucketReplicationMetrics API (#11985) 2021-04-06 11:32:52 -07:00
Poorna Krishnamoorthy 075bccda42
Fix cluster bucket stats API for prometheus (#11970)
Metrics calculation was accumulating inital usage across all nodes
rather than using initial usage only once.

Also fixing:
- bug where all  peer traffic was going to the same node.
- reset counters when replication status changes from
PENDING -> FAILED
2021-04-06 08:36:54 -07:00
Harshavardhana 09ee303244
add cluster support for realtime bucket stats (#11963)
implementation in #11949 only catered from single
node, but we need cluster metrics by capturing
from all peers. introduce bucket stats API that
will be used for capturing in-line bucket usage
as well eventually
2021-04-04 15:34:33 -07:00
Poorna Krishnamoorthy 47c09a1e6f
Various improvements in replication (#11949)
- collect real time replication metrics for prometheus.
- add pending_count, failed_count metric for total pending/failed replication operations.

- add API to get replication metrics

- add MRF worker to handle spill-over replication operations

- multiple issues found with replication
- fixes an issue when client sends a bucket
 name with `/` at the end from SetRemoteTarget
 API call make sure to trim the bucket name to 
 avoid any extra `/`.

- hold write locks in GetObjectNInfo during replication
  to ensure that object version stack is not overwritten
  while reading the content.

- add additional protection during WriteMetadata() to
  ensure that we always write a valid FileInfo{} and avoid
  ever writing empty FileInfo{} to the lowest layers.

Co-authored-by: Poorna Krishnamoorthy <poorna@minio.io>
Co-authored-by: Harshavardhana <harsha@minio.io>
2021-04-03 09:03:42 -07:00