Commit Graph

182 Commits

Author SHA1 Message Date
Harshavardhana 1d3bd02089
avoid close 'nil' panics if any (#18890)
brings a generic implementation that
prints a stack trace for 'nil' channel
closes(), if not safely closes it.
2024-01-28 10:04:17 -08:00
Harshavardhana 74851834c0
further bootstrap/startup optimization for reading 'format.json' (#18868)
- 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.
2024-01-25 12:45:46 -08:00
Harshavardhana e377bb949a
migrate bootstrap logic directly to websockets (#18855)
improve performance for startup sequences by 2x for 300+ nodes.
2024-01-24 13:36:44 -08:00
Harshavardhana 52229a21cb
avoid reload of 'format.json' over the network under normal conditions (#18842) 2024-01-23 14:11:46 -08:00
Harshavardhana a0e1163fb6
reject reference format from a different deployment (#18800)
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.
2024-01-16 15:13:14 -08:00
Harshavardhana e5c8794b8b
avoid disk monitoring leaks under various conditions (#18777)
- 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.
2024-01-12 01:48:36 -08:00
Shubhendu e31081d79d
Heal buckets at node level (#18612)
Signed-off-by: Shubhendu Ram Tripathi <shubhendu@minio.io>
2024-01-09 20:34:04 -08:00
Harshavardhana a50ea92c64
feat: introduce list_quorum="auto" to prefer quorum drives (#18084)
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.
2023-12-29 15:52:41 -08:00
Harshavardhana 5b2ced0119
re-use globalLocalDrives properly (#18721) 2023-12-29 09:30:10 -08:00
Anis Eleuch 8432fd5ac2
prom: Add online and healing drives metrics per erasure set (#18700) 2023-12-21 16:56:43 -08:00
Harshavardhana 7c948adf88
allow pre-allocating buffers to reduce frequent GCs during growth (#18686)
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.
2023-12-21 08:59:38 -08:00
Harshavardhana b3314e97a6
re-use the same local drive used by remote-peer (#18645)
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.
2023-12-13 19:27:55 -08:00
Harshavardhana e30c0e7ca3 Revert "Heal buckets at node level (#18504)"
This reverts commit 708296ae1b.
2023-12-05 22:34:46 -08:00
Shubhendu 708296ae1b
Heal buckets at node level (#18504) 2023-12-05 02:17:35 -08:00
Klaus Post 2229509362
fix: leaking offline disks in MarkOffline() thread (#18414)
`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.
2023-11-09 09:33:32 -08:00
Aditya Manthramurthy 1c99fb106c
Update to minio/pkg/v2 (#17967) 2023-09-04 12:57:37 -07:00
Harshavardhana eb55034dfe
optimize deletePrefix, use direct set location via object name (#17827)
* optimize deletePrefix, use direct set location via object name

instead of fanning out the calls for an object force delete
we can assume the set location and not do fan-out calls

* Apply suggestions from code review

Co-authored-by: Krishnan Parthasarathi <krisis@users.noreply.github.com>

---------

Co-authored-by: Krishnan Parthasarathi <krisis@users.noreply.github.com>
2023-08-09 16:30:22 -07:00
Harshavardhana b0f0e53bba
fix: make sure to correctly initialize health checks (#17765)
health checks were missing for drives replaced since

- HealFormat() would replace the drives without a health check
- disconnected drives when they reconnect via connectEndpoint()
  the loop also loses health checks for local disks and merges
  these into a single code.
- other than this separate cleanUp, health check variables to avoid
  overloading them with similar requirements.
- also ensure that we compete via context selector for disk monitoring
  such that the canceled disks don't linger around longer waiting for
  the ticker to trigger.
- allow disabling active monitoring.
2023-08-01 10:54:26 -07:00
Harshavardhana 81be718674
fix: optimize DiskInfo() call avoid metrics when not needed (#17763) 2023-07-31 15:20:48 -07:00
Aditya Manthramurthy 5a1612fe32
Bump up madmin-go and pkg deps (#17469) 2023-06-19 17:53:08 -07:00
Klaus Post c839b64f6a
fix: compressed+encrypted block overhead (#17289) 2023-05-26 10:57:07 -07:00
jiuker f037c9b286
Protecting the read index is not out of bounds (#17226) 2023-05-17 12:09:41 -07:00
Praveen raj Mani 72802a5972
Use 'minio/pkg/sync/errgroup' and 'minio/pkg/workers' (#17069) 2023-04-25 22:57:40 -07:00
Harshavardhana 84f31ed45d
simplify MRF, converge it to regular healing (#17026) 2023-04-19 07:47:42 -07:00
Harshavardhana 6825bd7e75
fix: inlined objects don't need to honor long locks (#17039) 2023-04-17 12:16:37 -07:00
Poorna d1e775313d
support decommissioning of tiered objects (#16751) 2023-03-16 07:48:05 -07:00
Harshavardhana b4ef5ff294
remove unnecessary code checking for supported features (#16423) 2023-01-17 19:37:47 +05:30
Anis Elleuch 2146ed4033
xl: Quit early when EC config is incorrect (#16390)
Co-authored-by: Anis Elleuch <anis@min.io>
2023-01-09 23:07:45 -08:00
Anis Elleuch 0333412148
fix: heal only once per disk per set among multiple disks (#16358) 2023-01-05 20:41:19 -08:00
Harshavardhana a15a2556c3
converge listBuckets() as a peer call (#16346) 2023-01-03 23:39:40 -08:00
Harshavardhana f1bbb7fef5
vectorize cluster-wide calls such as bucket operations (#16313) 2023-01-03 08:16:39 -08:00
Harshavardhana f93183f66e
fix: a deadlock by refactoring listBuckets() under site replication (#16323) 2022-12-29 00:08:31 -08:00
Harshavardhana b882310e2b
avoid locks for internal and invalid buckets in MakeBucket() (#16302) 2022-12-23 07:46:00 -08:00
Aditya Manthramurthy a30cfdd88f
Bump up madmin-go to v2 (#16162) 2022-12-06 13:46:50 -08:00
Harshavardhana 5a8df7efb3
re-implement StorageInfo to be a peer call (#16155) 2022-12-01 14:31:35 -08:00
Klaus Post cc1d8f0057
Check for abandoned data when healing (#16122) 2022-11-28 10:20:55 -08:00
Klaus Post 98ba622679
Reduce temporary file clean-up waits (#16110) 2022-11-22 07:23:36 -08:00
Klaus Post ecc932d5dd
Clean entire tmp-old on restart (#15979) 2022-10-31 07:27:50 -07:00
Harshavardhana 23b329b9df
remove gateway completely (#15929) 2022-10-24 17:44:15 -07:00
Klaus Post 3c605c93fe
warn when 0 parity has been set as default parity (#15790) 2022-10-04 22:41:42 -07:00
Klaus Post a9f1ad7924
Add extended checksum support (#15433) 2022-08-29 16:57:16 -07:00
ebozduman b57e7321e7
Replaces 'disk'=>'drive' visible to end user (#15464) 2022-08-04 16:10:08 -07:00
Poorna 426c902b87
site replication: fix healing of bucket deletes. (#15377)
This PR changes the handling of bucket deletes for site 
replicated setups to hold on to deleted bucket state until 
it syncs to all the clusters participating in site replication.
2022-07-25 17:51:32 -07:00
Anis Elleuch ed0cbfb31e
fix: rootdisk detection by not using cached value when GetDiskInfo() errors out (#15249)
GetDiskInfo() uses timedValue to cache the disk info for one second.

timedValue behavior was recently changed to return an old cached value
when calculating a new value returns an error.

When a mount point is empty, GetDiskInfo() will return errUnformattedDisk,
timedValue will return cached disk info with unexpected IsRootDisk value,
e.g. false if the mount point belongs to a root disk. Therefore, the mount
point will be considered a valid disk and will be formatted as well.

This commit will also add more defensive code when marking root disks:
always mark a disk offline for any GetDiskInfo() error except
errUnformattedDisk. The server will try anyway to reconnect to those
disks every 10 seconds.
2022-07-07 17:05:23 -07:00
Harshavardhana 32b2f6117e
fix: do not pass around sync.Map (#15250)
it is not safe to pass around sync.Map
through pointers, as it may be concurrently
updated by different callers.

this PR simplifies by avoiding sync.Map
altogether, we do not need sync.Map
to keep object->erasureMap association.

This PR fixes a crash when concurrently
using this value when audit logs are
configured.

```
fatal error: concurrent map iteration and map write

goroutine 247651580 [running]:
runtime.throw({0x277a6c1?, 0xc002381400?})
        runtime/panic.go:992 +0x71 fp=0xc004d29b20 sp=0xc004d29af0 pc=0x438671
runtime.mapiternext(0xc0d6e87f18?)
        runtime/map.go:871 +0x4eb fp=0xc004d29b90 sp=0xc004d29b20 pc=0x41002b
```
2022-07-07 17:04:25 -07:00
Klaus Post ac055b09e9
Add detailed scanner metrics (#15161) 2022-07-05 14:45:49 -07:00
Harshavardhana bd099f5e71
fix: change timedValue to return the previously cached value (#15169)
fix: change timedvalue to return previous cached value

caller can interpret the underlying error and decide
accordingly, places where we do not interpret the
errors upon timedValue.Get() - we should simply use
the previously cached value instead of returning "empty".

Bonus: remove some unused code
2022-06-25 08:50:16 -07:00
Poorna cb097e6b0a
CopyObject: fix read/write err on closed pipe (#15135)
Fixes: #15128
Regression from PR#14971
2022-06-21 19:20:11 -07:00
Harshavardhana f1abb92f0c
feat: Single drive XL implementation (#14970)
Main motivation is move towards a common backend format
for all different types of modes in MinIO, allowing for
a simpler code and predictable behavior across all features.

This PR also brings features such as versioning, replication,
transitioning to single drive setups.
2022-05-30 10:58:37 -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