Commit Graph

6 Commits

Author SHA1 Message Date
Harshavardhana
9693c382a8
make renameData() more defensive during overwrites (#19548)
instead upon any error in renameData(), we still
preserve the existing dataDir in some form for
recoverability in strange situations such as out
of disk space type errors.

Bonus: avoid running list and heal() instead allow
versions disparity to return the actual versions,
uuid to heal. Currently limit this to 100 versions
and lesser disparate objects.

an undo now reverts back the xl.meta from xl.meta.bkp
during overwrites on such flaky setups.

Bonus: Save N depth syscalls via skipping the parents
upon overwrites and versioned updates.

Flaky setup examples are stretch clusters with regular
packet drops etc, we need to add some defensive code
around to avoid dangling objects.
2024-04-23 10:15:52 -07:00
Harshavardhana
b6e98aed01
fix: found races in accessing globalLocalDrives (#19069)
make a copy before accessing globalLocalDrives

Bonus: update console v0.46.0

Signed-off-by: Harshavardhana <harsha@minio.io>
2024-02-16 17:15:57 -08:00
Harshavardhana
630963fa6b
protect tracker copy properly to avoid race (#18984)
```
WARNING: DATA RACE
Write at 0x00c000aac1e0 by goroutine 1133:
  github.com/minio/minio/cmd.(*healingTracker).updateProgress()
      github.com/minio/minio/cmd/background-newdisks-heal-ops.go:183 +0x117
  github.com/minio/minio/cmd.(*erasureObjects).healErasureSet.func5()
      github.com/minio/minio/cmd/global-heal.go:292 +0x1d3

Previous read at 0x00c000aac1e0 by goroutine 1003:
  github.com/minio/minio/cmd.(*allHealState).updateHealStatus()
      github.com/minio/minio/cmd/admin-heal-ops.go:136 +0xcb
  github.com/minio/minio/cmd.(*healingTracker).save()
      github.com/minio/minio/cmd/background-newdisks-heal-ops.go:223 +0x424
```
2024-02-06 08:56:59 -08:00
Harshavardhana
e1e33077e8
fix: tests and resync replication status (#18244) 2023-10-13 17:03:34 -07:00
Harshavardhana
4a425cbac1
cleanup scripts and apply shfmt (#17284) 2023-05-25 22:07:25 -07:00
Harshavardhana
a6e0ec4e6f
Add support converting non-inlined to inlined (#15444)
This is a feature to allow for inode compaction on
large clusters that use a lot of small files spread
across a large heirarchy.
2022-08-02 23:10:22 -07:00