A.2.3 Error messages (sfcfrmd daemon)

This section explains the sfcfrmd daemon error messages of the GFS Shared File System.

ERROR: sfcfrmd:0001: not superuser

Explanation

The sfcfrmd daemon is not performed by super user

Response

Please start sfcfrmd daemon by sfcfrmstart(1M) command as a super user.

ERROR: sfcfrmd:0002: associated with a terminal device

Explanation

The direct module is performed.

Response

Start the sfcfrmd daemon by executing sfcfrmstart(1M) with super user access privileges.

ERROR: sfcfrmd:0003: cannot read pathname or something wrong

Explanation

Since there was injustice or file reading of pathname went wrong, sfcfrmd daemon starting went wrong.

Response

Please check whether the file of pathname is set up correctly.

ERROR: sfcfrmd:0004: cannot get node version sfile

Explanation

The sfcfrmd daemon failed to start, because a working file sfile in /var/opt/FJSVgfs was broken.

Response

Make sure that /var/opt directory has a space by df(1M) command. If there is not enough space, delete unnecessary files or move them to other file systems to increase space. Then re-start the node.

ERROR: sfcfrmd:0005: something wrong in pathname

Explanation

The sfcfrmd daemon failed to start, because /etc/services has no entry of sfcfsrm or pathname is broken.

Response

Check if sfcfsrm<Port number>/tcp is added to the "/etc/services" file on all the cluster nodes. If it is not, add it to the file. For details about the file definition, see services(4) of the Solaris documentation.

If the setting above is right, pathname seems broken. Collect the diagnostic data with FJQSS or fjsnap and contact field engineers.

ERROR: sfcfrmd:0006: cannot allocate memory space

Explanation

Since acquisition of a memory went wrong, sfcfrmd daemon starting went wrong.

Response

Please check the state of a system, and wait for the end of other processes, or increase and reboot a swap domain.

ERROR: sfcfrmd:0007: cannot start service

Explanation

The sfcfrmd daemon starting went wrong.

Response

Please confirm whether the port number of sfcfsrm in the /etc/services is the same as the node starting the sfcfrmd daemon.

When the port number is different
After changes to the same port number as the node that has started the sfcfrmd daemon, please use sfcfrmstart(1M) by the super user authority and start the sfcfrmd daemon. For details about the file definition, see services(4) of the Solaris documentation.
When the port number is the same
Collect the diagnostic data with FJQSS or fjsnap in all nodes, and contact field engineers.

ERROR: sfcfrmd:0008: I/O error or data corruption of management partition: pathname

Explanation

The sfcfrmd daemon failed in reading and writing of pathname.

Response

Please check whether a setup as a management partition domain of the GFS Shared File System is right, or a disk operates normally.

ERROR: sfcfrmd:0009: something wrong in cluster settings

Explanation

Since a cluster was not started, sfcfrmd daemon starting went wrong.

Response

Please start a cluster.

ERROR: sfcfrmd:0010: cannot get host name

Explanation

The CIP address registered in the management partition changed.

Response

By remaking and restoring the management partition, back the configuration of the partition to a previous one before a value of the CIP address changed. For details to remake and restore, see "9.6 Backup of the management partition information" and "9.7 Restoring of the management partition information".

ERROR: sfcfrmd:0011: cannot connect to sfcprmd

Explanation

Connection with sfcprmd daemon went wrong.

Response

Make sure that the sfcprmd daemon is running by ps(1) command (for example, "/usr/bin/ps -e | /usr/bin/grep sfcprmd".) If the sfcprmd daemon was stopped by the unload dependency script, restart sfcprmd daemon by the load script ("/opt/SMAW/SMAWcf/dep/start.d/S81sfcfs load".) Otherwise, re-start the node.
If this message is output while the sfcprmd daemon is running, collect a core image of the daemon, and a file output with FJQSS or fjsnap for troubleshooting information then contact field engineers. Restart the node where this problem occurred.

ERROR: sfcfrmd:0012: cannot connect to sfchnsd

Explanation

Connection with sfchnsd daemon went wrong.

Response

Make sure that the sfchnsd daemon is running by ps(1) command (for example, "/usr/bin/ps -e | /usr/bin/grep sfchnsd".) If the sfchnsd daemon was stopped by the unload dependency script, restart sfchnsd daemon by the load script ("/opt/SMAW/SMAWcf/dep/start.d/S81sfcfs load".) Otherwise, re-start the node.
If this message is output while the sfchnsd daemon is running, collect a core image of the daemon, and a file output with FJQSS or fjsnap for troubleshooting information then contact field engineers. Restart the node where this problem occurred.

ERROR: sfcfrmd:0013: cannot activate sfchnsd

Explanation

The sfchnsd daemon is not activated.

Response

Collect the diagnostic data with FJQSS or fjsnap in the node, and contact field engineers.

ERROR: sfcfrmd:0014: cannot get service port

Explanation

A service port is not able to be acquired.

Response

Check if sfcfsrm<Port number>/tcp is added to the /etc/services file on all the cluster nodes. If it is not, add it to the file. For details about the file definition, see services(4) of the Solaris documentation.

ERROR: sfcfrmd:0015: cannot get limit of newly created descriptor

Explanation

The maximum value of file descriptor generation was not able to be acquired.

Response

Please check the maximum value of a file descriptor.

ERROR: sfcfrmd:0019: no node information on special

Explanation

The node information for this node is not registered into the management partition special.

Response