[
https://issues.apache.org/jira/browse/HDDS-14871?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Devesh Kumar Singh updated HDDS-14871:
--------------------------------------
Description:
## Problem
`StorageVolumeChecker.checkAllVolumes()` waits on a single `CountDownLatch` for
all volume health checks to complete. If the latch expires before any volume
finishes — due to any transient stall — **every pending volume is immediately
marked FAILED** with zero tolerance, producing false-positive volume failures.
The existing per-volume IO-failure sliding window in `StorageVolume.check()`
does not address this because it only applies when a check **completes**, not
when the latch times out.
## Solution
Add a per-volume consecutive latch-timeout counter (`consecutiveTimeoutCount`)
to `StorageVolume`. When `checkAllVolumes()` latch expires and a volume has not
yet reported a result, its counter is incremented. The volume is only added to
the failed set if `count > hdds.datanode.disk.check.timeout.tolerated`. A
successful check resets the counter to 0.
Volumes that explicitly return `FAILED` from `check()` (genuine IO failures,
missing directory, bad permissions) are unaffected and continue to fail
immediately.
was:
## Problem
`StorageVolumeChecker.checkAllVolumes()` waits on a single `CountDownLatch` for
all volume health checks to complete. If the latch expires before any volume
finishes — due to any transient stall — **every pending volume is immediately
marked FAILED** with zero tolerance, producing false-positive volume failures.
The existing per-volume IO-failure sliding window in `StorageVolume.check()`
does not address this because it only applies when a check **completes**, not
when the latch times out.
## Solution
Add a per-volume consecutive latch-timeout counter (`consecutiveTimeoutCount`)
to `StorageVolume`. When `checkAllVolumes()` latch expires and a volume has not
yet reported a result, its counter is incremented. The volume is only added to
the failed set if `count > hdds.datanode.disk.check.timeout.tolerated`. A
successful check resets the counter to 0.
Volumes that explicitly return `FAILED` from `check()` (genuine IO failures,
missing directory, bad permissions) are unaffected and continue to fail
immediately.
## New Configuration
| Key | Default | Description |
|-----|---------|-------------|
| `hdds.datanode.disk.check.timeout.tolerated` | `1` | Number of consecutive
latch timeouts allowed per volume before marking it FAILED. Set to `0` to
restore current behavior. |
> DataNode: tolerate per-volume health-check latch timeouts before marking
> volumes failed
> ---------------------------------------------------------------------------------------
>
> Key: HDDS-14871
> URL: https://issues.apache.org/jira/browse/HDDS-14871
> Project: Apache Ozone
> Issue Type: Task
> Components: Ozone Datanode
> Reporter: Devesh Kumar Singh
> Assignee: Devesh Kumar Singh
> Priority: Major
>
> ## Problem
> `StorageVolumeChecker.checkAllVolumes()` waits on a single `CountDownLatch`
> for all volume health checks to complete. If the latch expires before any
> volume finishes — due to any transient stall — **every pending volume is
> immediately marked FAILED** with zero tolerance, producing false-positive
> volume failures.
> The existing per-volume IO-failure sliding window in `StorageVolume.check()`
> does not address this because it only applies when a check **completes**, not
> when the latch times out.
> ## Solution
> Add a per-volume consecutive latch-timeout counter
> (`consecutiveTimeoutCount`) to `StorageVolume`. When `checkAllVolumes()`
> latch expires and a volume has not yet reported a result, its counter is
> incremented. The volume is only added to the failed set if `count >
> hdds.datanode.disk.check.timeout.tolerated`. A successful check resets the
> counter to 0.
> Volumes that explicitly return `FAILED` from `check()` (genuine IO failures,
> missing directory, bad permissions) are unaffected and continue to fail
> immediately.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
