[
https://issues.apache.org/jira/browse/IGNITE-18313?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Vladimir Steshin updated IGNITE-18313:
--------------------------------------
Description:
We delete snapshot in `_IgniteSnapshotManager.initLocalSnapshotEndStage()`_ if
a snapshot creation error occurred:
{code:java}
if (req.error() != null) {
snpReq.error(req.error());
deleteSnapshot(snapshotLocalDir(req.snapshotName(), req.snapshotPath()),
pdsSettings.folderName());
}
{code}
This call is performed in _GridDiscoveryManager_'s thread,
"_disco-notifier-worker-..._".
Probably, it is better to use the snapshot pool for snapshot-related I/O
operations `_IgniteSnapshotManager.snapshotExecutorService()_`. This would help
not to stop other processings of various discovery events.
was:
We delete snapshot in `_IgniteSnapshotManager.initLocalSnapshotEndStage()`_ if
a snapshot creation error occurred:
{code:java}
if (req.error() != null) {
snpReq.error(req.error());
deleteSnapshot(snapshotLocalDir(req.snapshotName(), req.snapshotPath()),
pdsSettings.folderName());
}
{code}
This call is performed in _GridDiscoveryManager_'s thread,
"_disco-notifier-worker-..._".
Probably, it is better to use the snapshot pool for snapshot-related I/O
operations `_IgniteSnapshotManager.snapshotExecutorService()_`. This would help
not to stop other processing of various discovery events.
> Delete snapshot on error in the snapshot thread pool.
> -----------------------------------------------------
>
> Key: IGNITE-18313
> URL: https://issues.apache.org/jira/browse/IGNITE-18313
> Project: Ignite
> Issue Type: Improvement
> Reporter: Vladimir Steshin
> Priority: Minor
>
> We delete snapshot in `_IgniteSnapshotManager.initLocalSnapshotEndStage()`_
> if a snapshot creation error occurred:
> {code:java}
> if (req.error() != null) {
> snpReq.error(req.error());
> deleteSnapshot(snapshotLocalDir(req.snapshotName(), req.snapshotPath()),
> pdsSettings.folderName());
> }
> {code}
> This call is performed in _GridDiscoveryManager_'s thread,
> "_disco-notifier-worker-..._".
> Probably, it is better to use the snapshot pool for snapshot-related I/O
> operations `_IgniteSnapshotManager.snapshotExecutorService()_`. This would
> help not to stop other processings of various discovery events.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
