Sammi Chen created HDDS-14888:
---------------------------------
Summary: Improve DiskCheckUtil.checkReadWrite to tolerate disk
full and no handle case
Key: HDDS-14888
URL: https://issues.apache.org/jira/browse/HDDS-14888
Project: Apache Ozone
Issue Type: Improvement
Reporter: Sammi Chen
Assignee: Sammi Chen
DiskCheckUtil.checkReadWrite does file write and read check to make sure disk
is writable and readable.
If disk is full or process's file handler is exhausted, it will catch the
exception and return false.
The outer StorageVolume#check will decide whether mark this time volume check
as fail or not, depending on whether volume has more than 100B *2 free space.
As the disk space will dynamically change, so there is change that
checkReadWrite() is failed due to disk full, while when below logic is
evaluated, it has enough available space again.
{code:java}
if (!diskChecksPassed && getCurrentUsage().getAvailable() < minimumDiskSpace) {
{code}
Further more, if the checkReadWrite() failure is because of file handler
exhausted, which also could be a temporal failure.
The goal of this target is to analysis the exception in checkReadWrite()
directly, so that disk full and no file handler case will not return false.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
