Error Conditions Column Set

This column set shows the most common situations that require attention. This is the default column set that Sonar displays, and issues that show on this column set should be addressed first.

DataCollectionState

This is not an FRS-reported error. Instead, it indicates that Sonar was unable to collect one or more pieces of status information about the FRS system. The details of the failure can be determined by either hovering the mouse pointer over the cell, or by looking at the DataCollectionError column in the All Columns view:

• PerfCtr. Failed to read Performance counter

  Known to fail on Windows 2000 SP2 systems and Windows Server 2003 RC1 systems. On other platforms, issues are normally resolved with a refresh.

• Registry. Failed to read registry
• SCM. Failed to query Service Control Manager
• DS. Failed to query Active Directory
• TimeZone. Failed to get TimeZone information
• Proc. Failed to collect Process information
• WMI. Failed to query information via WMI
• EventLog. Failed to query Event log
• SYSVOL. Failed to connect to SYSVOL file share
• FRSVer. Failure of the FRS Version RPC interface
• FRSInlog. Failure of the FRS Inlog RPC interface
• FRSSets. Failure of the FRS Sets RPC interface

FRSState

Servers are normally in the ACTIVE state, look for servers in other states:

• ALLOCATED. Replica set is in an initializing state
• INITIALIZING. Replica set is in an initializing state
• STARTING. Starting the replica set
• STOPPED*. Replica set is now stopped, typically because the set no longer has connections to any partners
• ERROR*. Replica set is stopped due to an error
• JRNL_WRAP_ERROR*. Replica set is stopped due to data loss in the NTFS journal
• REPLICA_DELETED. Replica set is marked as deleted

If a member is in one of the states marked with an asterisk (*), it requires attention before it can replicate again

SCM State

Shows if the FRS Service is running according to the Windows Service Control Manager. Look for servers that are not in the "Running" state, and investigate. FRS servers should normally be left running.

SYSVOLShared

If the replica set is a SYSVOL, look for "Not shared" or "Not a junction." For troubleshooting information, see Verifying SYSVOL Junction Points.

LongJoinCycle

High numbers indicate servers that have had problems joining with a replication partner. For troubleshooting details, see Troubleshooting FRS Events 13508 without FRS Event 13509.

ExcessiveReplicationCycle

Shows servers that have detected files that have been touched by some process, but not updated. If this is non-zero, see Excessive Replication and High CPU and Disk Usage by Ntfrs.exe.

Huge File Cycle

This shows the number of 13523 events logged on this member in the last 24 hours. This indicates that a file could not be replicated since it was larger than the staging directory size, and so cannot be replicated until the staging area has been reconfigured.

Increase the staging directory size on computers in the replica set so that these files can be replicated.

To increase the size of the staging directory

1. In the Run dialog box, type regedit.exe, and click OK.
2. Navigate to HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\NtFrs\Parameters.
3. Right-click the Staging Space Limit in KB entry and click Modify.
4. Click Decimal, type the desired size of the staging directory in kilobytes, and click OK.
5. Close the registry editor.
6. At the command prompt, type net stop ntfrs & net start ntfrs.

Staging Full Cycle

This shows the number of 13522 (Staging Area Full) events logged on this member in the last 24 hours. This indicates that a file could not be replicated because the staging directory was full. To resolve this, increase the size of the staging directory.

ChangeOrdersMorphedCycle and ChangeOrdersMorphedTotal

Shows the number of directories morphed to foldername_NTFRS_xxxxxxxx names due to conflicting updates. See Morphed Folders for details on the causes and cleanup of this situation.

CommTimeoutDelta, CommTimeoutCycle, and CommTimeoutTotal

Look for computers with high numbers of communication timeouts. This is an indication of network connectivity problems, and if it persists, should be investigated.