Recalling File System Archiving (FSA) Placeholders may cause Windows File Servers to become unresponsive or trigger a BugCheck
|Article:TECH215527|||||Created: 2014-03-04|||||Updated: 2014-07-22|||||Article URL http://www.symantec.com/docs/TECH215527|
In Enterprise Vault (EV) 10.0 Service Pack 4 (10.0.4), while performing the FSA file recall operations, the Windows File Servers, including Failover Cluster nodes, may become unresponsive and may require a server restart to resolve the issue. Such situation may occur again for some other files being recalled. In some environments the issue has also led to a Blue Screen of Death (BSOD).
In EV 10.0.4 and corresponding Cumulative Hotfixes, a change was introduced in the EV-FSA file system mini filter driver behavior to make the recall operations on archived files visible to Microsoft Quota Management infrastructure. This was introduced to ensure the file system quota consumption accounts for recalled file data. However, in some cases this change caused deadlock conditions between the EV-FSA mini filter and cache manager on the file server.
A workaround may help prevent these issues. The suggestion is to revert the FSA Agent on the affected File Servers to 10.0.3 or earlier versions where this specific driver logic is not present. Additionally, enabling the Passthrough recall feature at the affected File Servers and corresponding Volume targets may also be necessary to avoid eventual hangs. For information on configuring the Passthrough option, please refer to the Related Articles section below.
This issue has been addressed in the following release:
Enterprise Vault 10.0.4 Cumulative Hotfix 3 Release
Recalling large ISO files results in BSOD or, with Passthrough, the cluster node hangs
Windows 2008 File Server stops responding when accessing placeholders
Article URL http://www.symantec.com/docs/TECH215527