Chapter 116 NonStop TCP/IPv6 Messages

The messages in this chapter are generated by the NonStop TCP/IPv6 subsystem. The subsystem ID displayed by these messages includes ZTC6 as the subsystem name.




	NOTE: Negative-numbered messages are common to most subsystems. If you receive a negative-numbered message that is not described in this chapter, see Chapter 15.

-3

subnet changed from old-state to new-state reason.

`subnet`	is the name of the subnet.
`old-state`	indicates the former summary state of the subnet.
`new-state`	indicates the present summary state of the subnet.
`reason`	is the reason for the change in summary state

Cause The summary state of a NonStop TCP/IPv6 subnet has changed because of a command or an error.

Effect None

Recovery If an operator command caused the summary-state change, no action is necessary. If the state changed for any other reason, note down the error description in the accompanying event message that is returned and contact your service provider.

TCPIPv6 no longer operational on CPU cpu-num, cause

`cpu-num`	is the number of the processor from which the event was reported.
`cause`	contains the text information about the halt.

Cause The NonStop TCP/IPv6 monitor process detected an internal error in the subsystem and has halted the processor. The NonStop TCP/IPv6 subsystem could not recover.

Effect All the established NonStop TCP/IPv6 connections through that processor are lost.

Recovery Restart the NonStop TCP/IPv6 subsystem on the processor where the event occurred and send the online processor dump file to your service provider for analysis.

ARP: Duplicate IP address ip-addr from Ethernet address enet-addr.

`ip-addr`	is the IP address in dotted decimal notation.
`enet-add`r	is the Ethernet address.

Cause The Address Resolution Protocol (ARP) module of the NonStop TCP/IPv6 subsystem has detected that another machine on the network is broadcasting the same IP address.

Effect The IP-address conflict might cause packets to go to the wrong machine.

Recovery Determine which machine is broadcasting incorrectly and reassign a unique IP address to that machine.

ARP: Ethernet broadcast address (FF.FF.FF.FF.FF.FF) used by IP address ip-addr.

ip-addr

is the IP address in dotted decimal notation.

Cause The Address Resolution Protocol (ARP) module of the NonStop TCP/IPv6 subsystem has detected that another machine on the Ethernet LAN is using the broadcast address (FF.FF.FF.FF.FF.FF) as its Ethernet address.

Effect All packets sent to the broadcast address are received by all machines on the Ethernet LAN. Performance on all machines is adversely affected. In addition, some implementations of TCP/IP might try to forward the packet(s) back over the Ethernet LAN, flooding the LAN and crashing most machines on the LAN.

Recovery Determine which machine is using the broadcast address as its Ethernet address and fix the machine. Note that you cannot reassign an Ethernet address.

TCPIPv6 monitor failed to open $Receive, CPU cpu-num, err err-code.

`cpu-num`	is the number of the processor in which the open operation failed.
`err-code`	is defined by the file system. To obtain information about the error code, type the code `err-code` at the TACL prompt.

Cause The NonStop TCP/IPv6 monitor process in the NonStop TCP/IPv6 subsystem failed to open the $RECEIVE during the initialization

Effect The NonStop TCP/IPv6 subsystem no longer exists in that processor.

Recovery Restart the NonStop TCP/IPv6 subsystem in the processor in which the event occurred and send the online processor dump file to your service provider for analysis. This event is an indication that a system internal problem occurred.

en if-unit: can't handle af family.

`if-unit`	is the interface unit
`family`	is the protocol family (INET or INET6).

Cause The loopback module of the NonStop TCP/IPv6 subsystem detected an unsupported protocol family.

Effect All the NonStop TCP/IPv6 subsystem activities related to the loopback interface no longer function.

Recovery Restart the NonStop TCP/IPv6 subsystem in the processor in which the event occurred and send the online processor dump file to your service provider for analysis.

in_cksum: out of mbufs

Cause Checksum failed in the checksum routine for Internet Protocol family headers.

Effect The connection may have been lost.

Recovery .Report the incident to your service provider.

Error err-code on LIF lif-name.

err-code

is the error code returned from the QIO subsystem. The values include:

5 ZSM_ERR_SEGMENT_NOT_INITIALIZED
35 ZSM_ERR_Q_FULL
73 ZSM_ERR_BAD_Q_ID
75 ZSM_ERR_BAD_USER_DATA_PTR

lif=name

is the name of the LIF involved.

Cause The NonStop TCP/IPv6 subsystem failed to communicate with the underlying LIF.

Effect The NonStop TCP/IPv6 subsystem either tries to register/de-register the filter or to repeat the last I/O operation. If this problem continues, the LIF becomes inaccessible.

Recovery An operator may need to check that the processor has access to the LIF, that the LIF is in the proper working state, and that the QIO subsystem is in the proper working state. If the error recurs after problems are found and corrected, report the incident to your service provider.

COULD NOT CREATE QIO QUEUE For Subnet subnet, Error err-code.

`subnet`	is the name of the subnet.
`err-code`	is the error code returned from the QIO subsystem. The possible values include:

5 ZSM_ERR_SEGMENT_NOT_INITIALIZED
10 ZSM_ERR_GETPOOL_FAILED
30 ZSM_ERR_Q_DUPLICATE
75 ZSM_ERR_BAD_MODULE_ID

Cause The QIO interface module of the NonStop TCP/IPv6 subsystem failed to create an inbound queue for filters to use

Effect The subnet configured in the NonStop TCP/IPv6 subsystem remains nonfunctional until the problem is resolved.

Recovery Check the state of the QIO subsystem to make sure it is in a proper working state. If the error recurs after problems are found and corrected, report the incident to your service provider.

LIF lif-name Not in STARTED State in CPU cpu-num.

`lif-name`	is the name of the LIF.
`cpu-num`	is the number of the processor in which the error occurred.

Cause The LIF is not in the STARTED state.

Effect The LIF configured in the NonStop TCP/IPv6 subsystem is not functional until the problem is resolved.

Recovery Make sure that the LIF is in the STARTED state.

lif-name Access Down in CPU cpu-num.

`lif-name`	is the name of the LIF.
`cpu-num`	is the number of the processor in which the event occurred.

Cause The LIF is not accessible

Effect The LIF configured in the NonStop TCP/IPv6 subsystem is not functional until the problem is resolved.

Recovery Make sure that the access state of the LIF is UP.

info-text.

info-text

is the informational text explaining this information event

Cause This event is informational only.

Effect None

Recovery None

TCPIPv6 Monitor process stopped by operator.

Cause The operator or system administrator stopped the monitor process in the NonStop TCP/IPv6 subsystem.

Effect All the network activity through the monitor in that processor ceases.

Recovery Because the NonStop TCP/IPv6 subsystem was brought down by the operator or system administrator, no further action is recommended.

Subnet subnet LIF Filter register timeout.

subnet

is the name of the NonStop TCP/IPv6 subsystem subnet attempting to register the filter.

Cause The NonStop TCP/IPv6 subsystem is waiting for a filter registration to complete.

Effect The NonStop TCP/IPv6 subsystem will attempt to register the filter again. If this problem continues, the LIF becomes inaccessible

Recovery If this event continues after another attempt has been made to register the filter, abort and restart the LIF.

LIF lif-name access unavailable from CPU cpu-num.

`lif-name`	is the name of the LIF.
`cpu-num`	is the number of the processor in which the event occurred.

Cause The NonStop TCP/IPv6 subsystem cannot access the LIF.

Effect The LIF configured in the NonStop TCP/IPv6 subsystem continues to be inaccessible until the problem is resolved.

Recovery Make sure that the LIF is accessible from the processor where the event occurred.

Unknown Device lif-name.

lif-name

is the name of the inaccessible LIF.

Cause The device name specified for configuring the subnet does not exist.

Effect The LIF configured for the NonStop TCP/IPv6 subsystem is not accessible until the problem is resolved.

Recovery Make sure that the LIF is configured in the SLSA subsystem.

Device lif-name Returned a NULL MAC Address.

lif-name

is the name of the LIF associated with this event.

Cause The SLSA subsystem returned a NULL MAC address to the NonStop TCP/IPv6 subsystem when NonStop TCP/IPv6 tried to obtain the MAC address for the interface

Effect The LIF configured for the NonStop TCP/IPv6 subsystem is inaccessible until the problem is resolved

Recovery Make sure that the LIF is properly configured in the SLSA subsystem.

name: Filter Register For IP Address ip-addr Failed, Error= err-code, Filter Type = type, Local Port = port

name

is the name of the interface (subnet) or the name of a filter.

ip-addr

is the IP address.

err-code

is the filter registration error code. The possible values of err-code are:

200	Filter exists
201	Invalid owner
202	No filter tree
203	No filter
204	Invalid parent filter
205	Bad LIF header
206	Filter other
207	Device inaccessible
208	Invalid parameter

type

is the type of filter registration. The possible values of type are listed in Table 116-1.

port

is the local port number.

Cause A filter registration failed on subnet name for the ip-addr. This event could be due to an improperly configured NonStop TCP/IPv6 subsystem or an internal NonStop TCP/IPv6 subsystem problem.

Effect The NonStop TCP/IPv6 subsystem will attempt to re-register the filter. If the filter registration is for a Subnet filter and the problem continues, the Subnet becomes inaccessible. If the registration is for a Port filter and the problem continues, filter registration will fail.

Recovery Determine the type and the cause of the failed registration from the err‑code value and take appropriate action.

if-name: Filter Register For IP Address ip-addr Failed, Timed Out.

`if-name`	is the subnet name.
`ip-addr`	is the IP address in dotted notation.

Cause A filter Register timeout occurred on subnet if-name for ip-addr.

Effect None

Recovery The NonStop TCP/IPv6 subsystem will try the filter registration at a later time. No operator intervention is required.

name: Filter Register For IP Address ip-addr Failed, Already In Use, Filter Type = type, Local port = l-port, Foreign port = f-port.

`name`	is the name of the interface (subnet) or the name of a filter.
`ip-addr`	is the IP address.
`type`	is the type of filter registration. The values of `type` are listed in Table 116-1.
`l-port`	is the local port number.
`f-port`	is the foreign port number.

Cause A filter register is already in use on subnet name for ip-addr. This event could be due to an improperly configured NonStop TCP/IPv6 subsystem or an internal subsystem problem.

Effect No retry will be done by the NonStop TCP/IPv6 subsystem. The application will receive a proper error code.

Recovery Determine the type and the cause of the failed registration from the error code value and take the appropriate action.

name: Filter De-register For IP Address ip-addr Failed, Error = err-code, Filter Type = type Local Port = port.

name

is the name of the interface (subnet) or the name of a filter.

ip-addr

is the IP address.

err-code

is the filter de-registration error code. The possible values of err-code are:

200	Filter exists
201	Invalid owner
202	No filter tree
203	No filter
204	Invalid parent filter
205	Bad LIF header
206	Filter other
207	Device inaccessible
208	Invalid parameter

type

is the type of filter de-registration. The possible values of type are listed in Table 116-1.

port

is the local port number.

Cause A filter de-registration failed on subnet x for the IP address. This event could be due to an improperly configured NonStop TCP/IPv6 subsystem or an internal NonStop TCP/IPv6 subsystem problem.

Effect None

Recovery Determine the type and cause of the failed registration from the err-code value and take the appropriate action.

module: No ARP cache for Route route-name: IP Address ip-addr.

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`route-name`	is the name of the route.
`ip-addr`	is the IP address in dotted notation.

Cause The NonStop TCP/IPv6 subsystem failed to resolve an IP address into an Ethernet address.

Effect All the communication attempts will use ip-addr to resolve a link-level route, and the ARP cache will fail.

Recovery Examine the routing tables in the processor in which this event is reported to ensure that:

The routing tables are set up properly,
The routers attached to the network are functioning
The LIF associated with the subnet is in a working state.

If the problem persists, report this problem to your service provider and include an online processor dump.

module: No Route Entry for IP Address ip-addr.

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`ip-addr`	is the IP address in dotted notation.

Cause The Address Resolution Protocol (ARP) module of the NonStop TCP/IPv6 subsystem failed to locate the route entry for the given IP address.

Effect All communication attempts to use ip-addr to resolve a link-level route and the ARP cache will fail.

Recovery Examine the routing tables in the processor in which this event is reported to ensure that:

The routing tables are setup properly,
The routers attached to the network are functioning
The LIF associated with the subnet is in a working state.

If the problem persists, report this problem to your service provider and include an online processor dump.

module: Bad Interface Type = interface-type

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`interface-type`	is the value of the unsupported interface type.

Cause The Address Resolution Protocol (ARP) module of the NonStop TCP/IPv6 subsystem detected an invalid interface type during ARP logic processing.

Effect An internal NonStop TCP/IPv6 problem occurred. You may observe some abnormal TCP/IP behaviors such as connection drops or failures to connect.

Recovery This event indicates an internal problem either in the routing logic or in the ARP logic. Report the incident to your service provider, supplying an online processor dump and a copy of the routing tables.

module: Bad Gateway Values, SA Family = sa-value, SA length = sa-len, Route Destination Address ip-addr.

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`sa-value`	is the socket address family.
`sa-len`	is the length of the socket address family.
`ip-addr`	is the route destination IP address in dotted decimal notation.

Cause The Address Resolution Protocol (ARP) module of the NonStop TCP/IPv6 subsystem detected either an invalid socket family or an invalid socket-family length.

Effect A link-level route failed was not created. Any attempt to communicate with ip‑addr address will fail until the problem is resolved.

module: Unable To Create ARP Entry, IP Address ip-addr, Route Dest Address route-addr.

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`ip-addr`	is the IP address related to ARP creation in dotted decimal notation.
`route-addr`	is the route destination address in dotted decimal notation.

Cause The Address Resolution Protocol (ARP) module of the NonStop TCP/IPv6 subsystem either failed to create or failed to look up a new address in the ARP table.

Effect Because the ARP cache lookup failed, a new attempt to connect will also fail, and existing connections might be temporarily lost. If the problem continues, all the existing connections will be lost.

Recovery Examine the command that failed to create the route and subsequently generated this event. If a route associated with the ARP entry already exists, then this event is just for informational purposes. If no route associated with the ARP cache exists, send the information about the routing tables to your service provider for analysis.

module: Route Not Freed, Route route-name, Reference count -cnt.

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`route-name`	is the name of the route.
`cnt`	is the negative reference count of this route.

Cause An attempt to free a routing structure has failed because the reference count of the route entry was negative.

Effect The route associated with this event becomes unusable. The connections through this route may experience abnormal behavior. A a memory leak might also occur.

Recovery This event is an indication that the reference count of that route was incremented or decremented incorrectly. Make a copy of the routing tables in the processor where this event occurred and send the information to your service provider for further analysis.

module: Trace buffer is full on CPU cpu-num.

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`cpu-num`	is the number of the processor in which the event was reported.

Cause One of the QIO buffers used for TCP6MON/TCPLIB trace buffering does not have enough room for another trace record.

Effect None

Recovery This event is for information. If the problem persists, please report this problem to your service provider.

module: TCP6MON failed to update a record to CONFIG DB error: err-code, Record-Type: type on CPU cpu-num.

module

is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.

err-code

is an error code returned by a configuration management routine.

type

is the record type. The currently defined record types are:

180	Config_DB_TYPE_TCP_PROCESS
181	Config_DB_TYPE_TCP_SUBNET
182	Config_DB_TYPE_TCP_ROUTE
183	Config_DB_TYPE_TCP_ENTRY
184	Config_DB_TYPE_TCP_SUBNET_2
185	Config_DB_TYPE_TCP_ROUTE_2
186	Config_DB_TYPE_TCP_IPV6_ADR
187	Config_DB_TYPE_TCP_IPV6_PREFIX

cpu-num

is the number of the processor in which the event was reported.

Cause An internal error occurred during configuration database management.

Effect This error may lead to inconsistent subnet or route structures among processors.

Recovery Examine the subnet and routing topology in all processors running NonStop TCP/IPv6. If any inconsistency exists among the processors, disable the configuration database for NonStop TCP/IPv6 and re-configure the subnets and routes.

module: TCP6MON failed to fetch record from CONFIG DB error: err-code, Record-Type: type on CPU cpu-num.

module

is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.

err-code

is an error code returned by a configuration management routine.

type

is the record type. The currently defined record types are:

180	Config_DB_TYPE_TCP_PROCESS
181	Config_DB_TYPE_TCP_SUBNET
182	Config_DB_TYPE_TCP_ROUTE
183	Config_DB_TYPE_TCP_ENTRY
184	Config_DB_TYPE_TCP_SUBNET_2
185	Config_DB_TYPE_TCP_ROUTE_2
186	Config_DB_TYPE_TCP_IPV6_ADR
187	Config_DB_TYPE_TCP_IPV6_PREFIX

cpu-num

is the number of the processor from which the event was reported.

Cause An internal error occurred during configuration database management.

Effect This error could lead to inconsistent subnet or route structures among processors.

module: TCP6MON Unexpected Sock Migration Request, originated from CPU cpu-num, Address ip-addr, from port port

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`cpu-num`	is the number of the processor from which the event was reported.
`ip_addr`	is the remote IP address associated with this migration request and notated in dotted decimal notation.
`port`	is the remote port number associated with this migration request.

Cause The application issuing the socket “ACCEPT_NW2” function call has cancelled the request.

Effect The migration request is discarded and the socket is not migrated to a new processor.

Recovery Informational message only; no corrective action is needed.

module: TCP6MON Sock Migration Request Get MBUF failed, originated from CPU cpu-num, Address ip-addr, from port port

`module`	is the function routine in the NonStop TCP/IPv6 subsystem from which this event is generated.
`cpu-num`	is the number of the processor from which the event was reported.
`ip_addr`	is the remote IP address associated with this migration request and displayed in dotted decimal notation.
`port`	is the remote port number associated with this migration request.

Cause The NonStop TCP/IPv6 subsystem failed to obtain memory for processing a migration request.

Effect The migration request will be dropped and the socket will not be migrated to a new processor.

Recovery Informational message only; no corrective action is needed.

200

TCP6MAN Unable to start backup process Error: err-code, Error-detail err-detail

`err-code`	is the error code indicating the reason why the backup did not start. See Appendix D for information about the specified error code.
`err-detail`	is the detailed information about the error.

Cause The TCP6MAN backup process cannot be started. To determine the cause, see err-code and error-detail.

Effect The backup TCP6MAN process is not created.

Recovery The primary TCP6MAN process continues to run and attempts to restart the backup until the backup process starts. If the problem persists, contact your service provider and provide the following:

Descriptions of the problem and accompanying symptoms
Details from this message and any other relevant messages
Supporting documentation such as Event Management Service (EMS) logs, trace files, and a processor dump, if applicable

If your local operating procedures require contacting the Global Mission Critical Solution Center (GMCSC), supply your system number and the numbers and versions of all related products.

201

TCP6MAN Checkpoint Error. Entity rept_entity, Chkpt Type #chkpt_type, Cause cause.

rept_entity

is the TCP6MAN check-point destination object type. The values for rept_type are:

0	A configuration change checkpoint failed. An SCF command did not checkpoint correctly to the backup TCP6MAN process.
1	A process change checkpoint failed. The TCP6MAN backup started, but it did not receive all the sync data from the primary process.
2	An OPEN checkpoint failed. The TCP6MAN backup did not receive an OPEN checkpoint from the TCP6MAN primary.
3	A Manager checkpoint failed. The TCP6MAN backup did not receive a sync data checkpoint from the primary process.

chkpt_type

is the type of checkpoint that failed. The values for chkpt_type are:

5001	ZCOM_CMD_ABORT
5003	ZCOM_CMD_ADD
5005	ZCOM_CMD_ALTER
5007	ZCOM_CMD_DELETE
5014	ZCOM_CMD_START
5015	ZCOM_CMD_STOP
5017	ZCOM_CMD_SWITCH
5018	ZCOM_CMD_TRACE

cause

specifies why this error occurred. The values forcause are:

1	The backup process received a subdevice number that is not configured.
2	The backup process received checkpoint data of the wrong size.
3	READLINK failed when reading checkpoint data.
4	The backup could not obtain memory for checkpointing additional control blocks.
5	The backup received a checkpoint with an unexpected type.

Cause The backup process cannot process a checkpoint.

Effect No immediate problems result.

Recovery Halt and dump the processor of the TCP6MAN backup process; then reload the processor to restart the TCP6MAN backup process.

If the problem persists, do not reload the backup; contact your support representative and provide the following:

Descriptions of the problem and accompanying symptoms
Details from this, and any other relevant messages
Supporting documentation such as Event Management Service (EMS) logs, trace files, and a processor dump, if applicable

If your local operating procedures require contacting the Global Mission Critical Solution Center (GMCSC), supply your system number and the numbers and versions of all related products.

202

TCP6MAN Switched from CPU from_cpu to CPU to_cpu.

`from_cp`	is the processor from which the TCP6MAN process switched.
`to_cpu`	is the processor to which the TCP6MAN process switched.

Cause An operator requested a processor switch through SCF.

Effect The TCP6MAN is running in a different processor.

Recovery Informational message only; no corrective action is needed.

203

TCP6MAN Error notification. Function notif-func, Error notif-err. Info1 notif-info1, Info2 notif-info2, Line line-num, Module mod-name [ , Name notif=name ].

notif-func

is the name of the function that failed.

notif-err

is the completion code returned by Guardian. The values for notif-err are:

1	Normal, voluntary termination with WARNING diagnostics. For example, if the process is a compiler, the compilation terminated with WARNING diagnostics after building a complete object file.
2	Abnormal, voluntary termination with FATAL errors or diagnostics. For example, if the process is a compiler, the compilation terminated with FATAL diagnostics and either an object file was not built or, if built, might be incomplete. A complete listing is generated.
3	Abnormal, voluntary, but premature termination with FATAL errors or diagnostics. For example, if the process is a compiler, the compilation terminated with FATAL diagnostics, and either no object file was built or an incomplete object file was built and an incomplete listing was generated (the compiler quit compiling prematurely).
4	The process never got started. This completion code exists primarily for the use of the command interpreter or other command language interpreters that can act as the executor process of a batch job. This code allows the executor process to detect that a process associated with a RUN statement never got started. In that sense, this completion code is a “fake” completion code. The command interpreter acts as though it received a termination message from the process that it tried to create, when in fact it received an error returned by the procedure or OSS function that launched the process. The command interpreter then makes the completion code and the error returned by the procedure or OSS function that launched the process available for evaluation, for example, by a batch job executor process.
5	Process calls PROCESS_STOP_ (with abnormal termination specified) or ABEND on itself. This code is the default completion code for the PROCESS_STOP_ procedure (when abnormal termination is specified) and for the ABEND procedure.
6	An external, but unauthorized process called PROCESS_STOP_, STOP, or ABEND to delete a process. The system includes this completion code in the process deletion message. If the process cannot be stopped, the request is saved so that when the process calls SETSTOP this completion code is sent with the process deletion message. The user id, the PCBCRAID (CAID), and the process ID of the process that caused the termination are included in the termination message.
7	Restart this job. This completion code is used by the NetBatch scheduler and an executor process. The executor process sets its completion code to this value upon termination; the scheduler interprets this completion code and restarts a “restartable” job.
8	Code 8 is the same as code 1, normal termination, except that you must examine the listing file to determine whether the results are acceptable. Completion code 8 is typically used by compilers.
9	The `kill()` or `raise()` function generated a signal that stopped the process. The termination information provides the signal number.

notif-info1 and notif-info2

provide context pertaining to the error.

line-num

is the line number in the module reporting the error notification.

mod-name

is the name of the TCP6MAN module reporting the error notification.

notif-name

if present, is the entity in TCP6MAN associated with the error.

Cause The cause depends on the notif-func and notif-err.

Effect TCP6MAN continues to run.

Recovery Contact your service provider and provide all relevant information.

204

TCP6MAN Process Abended. Error = #err-code, Cause: #reason, Line: line-num, Module: module-name.

err-code

is the file system error that occurred. See Appendix B for information about the specified error.

reason

indicates the reason that TCP6MAN abended during initialization. The possible values of reason are:

1	An error was encountered in the system configuration database.
2	Too few start-up arguments.
3	Invalid start-up argument.
4	Specified backup processor invalid.
5	Invalid queue name specified.
10	Checkpoint error.
11	TCP6MAN received an unexpected request.
12	A process that was being monitored abended.
13	An internal processing error occurred during backup takeover.

line-num

is the line number in the module where the internal error was detected.

module-name

is the name of the TCP6MAN module that detected the internal error.

Cause The TCP6MAN process detected an error during process initialization and cannot proceed.

Effect TCP6MAN terminates abnormally.

Recovery Restart TCP6MAN. If the process abends again, contact your service provider and provide the following:

Descriptions of the problem and accompanying symptoms
Details from the message or messages generated
Supporting documentation such as Event Management Service (EMS) logs, trace files, and a processor dump, if applicable

If your local operating procedures require contacting the Global Mission Critical Solution Center (GMCSC), supply your system number and the numbers and versions of all related products.

205

Master TCP6MON Process tcp6mon-name started on CPU cpu-num.

`tcp6mon-name`	is the name of the master TCP6MON process.
`cpu-num`	is the number of the processor in which the master TCP6MON process started.

Cause No master TCP6MON exists.

Effect None. This event is informational only.

Recovery This is an informational message only; no corrective action is needed.

206

TCP6MON Process tcp6mon-name started on CPU cpu-num.

`tcp6mon-name`	is the name of the TCP6MON process.
`cpu-num`	is the number of the processor in which the TCP6MON process started.

Cause The operator started a TCP6MON or brought up a processor in which a TCP6MON was configured.

Recovery This is an informational message only; no corrective action is needed.

207

TCP6MON Process tcp6mon-name stopped on CPU cpu-num.

`tcp6mon-name`	is the name of the TCP6MON process.
`cpu-num`	is the number of the processor in which the TCP6MON process was stopped.

Cause An operator stopped the monitor using SCF.

Effect The monitor stops.

Recovery This is an informational message only; no corrective action is needed.

208

Master TCP6MON switched from process tcp6mon-name1 on CPU cpu-number1 to process tcp6mon-name2 on CPU cpu-number2.

`tcp6mon-name1`	is the name of the TCP6MON process that was formerly the Master TCP6MON.
`cpu-number1`	is the number of the processor in which the former Master TCP6MON process resided.
`tcp6mon-name2`	is the name of the new Master TCP6MON process.
`cpu-number2`	is the number of the processor in which the new Master TCP6MON process resides.

Cause The old Master TCP6MON was stopped.

Effect The default filter is moved to the new Master TCP6MON.

Recovery None

209

No Master TCP6MON Process on system.

Cause An operator stopped either the NonStop TCP/IPv6 subsystem or the only TCP6MON in the system.

Effect No NonStop TCP/IPv6 subsystem is running.

Recovery Restart the TCP6MON processes.

211

Subnet subnet Duplicate IPv6 address ipv6-addr detected for addr-type.

`subnet`	is the name of the subnet.
`ipv6-addr`	is the IPv6 address that is the subject of the report.
`addr-type`	contains the word “Interface” or “Address.”

Cause The neighbor discovery protocol has detected another node on the network that is using the same IPv6 address.

Effect The address will be unusable on this machine until the problem is resolved.

Recovery Determine which machine is assigned the wrong IPv6 address and change the configuration.

212

module: TCP6MON Unexpected Sock Migration Request, originated from CPU cpu-num, Address ipv6-addr, from port port

`module`	is the function routine in the NonStop TCP/IPv6 subsystem that generated this event.
`cpu-num`	is the number of the processor in which this migration request was generated.
`ipv6-addr`	is the remote IPv6 address.
`port`	is the remote port number associated with this migration request.

Cause The NonStop TCP/IPv6 subsystem detects an unexpected migration request. The application issuing the socket “ACCEPT_NW2” function call has cancelled the request.

Effect The migration request will be dropped and the socket will not be migrated to new processor.

Recovery This is an informational message only; no corrective action is needed.

213

module: TCP6MON Sock Migration Request Get MBUF failed, originated from CPU cpu-num, Address ipv6-addr, from port port.

`module`	is the function routine in the NonStop TCP/IPv6 subsystem that generated this event.
`cpu-num`	is the number if the processor in which this migration request was generated.
`ipv6-addr`	is the remote IPv6 address.
`port`	is the remote port number associated with this migration request.

Cause No memory is available to format the migration request.

Effect The migration request will be dropped and the socket will not be migrated to a new processor.

Recovery This is an informational message only; no corrective action is needed.

214

name: Filter Register For IPv6 Address ipv6-addr Failed, Error= err-code, Filter Type = type, Local Port = port.

name

is the name of an interface (subnet) or the name of a filter/

ipv6-addr

is the IPv6 address.

err-code

is the filter registration error code. The values for err-code are:

200	Filter exists
201	Invalid owner
202	No filter tree
203	No filter
204	Invalid parent filter
205	Bad LIF header
206	Filter other
207	Device inaccessible
208	Invalid parameter

type

is the type of filter registration. The possible values of type are listed in Table 116-1:

port

is the local port number.

Cause A filter registration failed on subnet name for ipv6-addr. This event could be due to an improperly configured NonStop TCP/IPv6 subsystem or an internal NonStop TCP/IPv6 subsystem problem.

Recovery Determine the type and the cause of the failed registration from the err‑code value and take appropriate action.

215

if-name: Filter Register For IPv6 Address ipv6-addr Failed, Timed Out.

`if-name`	is the subnet name.
`ipv6-addr`	is the IPv6 address.

Cause A register timeout occurred on subnet if-name for ipv6-addr.

Effect None

Recovery The NonStop TCP/IPv6 subsystem will try the filter registration at a later time. No operator invention is required.

216

name: Filter Register For IPv6 Address ipv6-addr Failed, Already In Use, Filter Type = type, Local port = l-port, Foreign port = f-port

name

is the name of an interface (subnet) or the name of a filter/

ipv6-addr

is the IPv6 address.

err-code

is the filter registration error code. The values for err-code are:

200	Filter exists
201	Invalid owner
202	No filter tree
203	No filter
204	Invalid parent filter
205	Bad LIF header
206	Filter other
207	Device inaccessible
208	Invalid parameter

type

is the type of filter registration. The values for type are listed in Table 116-1.

l-port

is the local port number.

f-port

is the foreign port number.

Cause A filter register is already in use on name for ipv6-addr. This event could be due to an improperly configured NonStop TCP/IPv6 subsystem or an internal subsystem problem.

Effect No retry will be done by the NonStop TCP/IPv6 subsystem. The application will receive a proper error code.

Recovery Determine the type and the cause of the failed registration from the err‑code value and take the appropriate action

217

name: Filter De-register For IPv6 Address ip-addr Failed, Error = err-code, Filter Type = type Local Port = port.

name

is the name of an interface (subnet) or the name of a filter.

ipv6-addr

is the IPv6 address for which the filter de-registration failed.

err-code

is the filter registration error code. The values for err-code are:

200	Filter exists
201	Invalid owner
202	No filter tree
203	No filter
204	Invalid parent filter
205	Bad LIF header
206	Filter other
207	Device inaccessible
208	Invalid parameter

type

is the type of filter registration. The possible values for type are listed in Table 116-1.

port

is the port number.

Cause A filter de-registration failed on subnet name foripv6-addr. This event could be due to an improperly configured NonStop TCP/IPv6 subsystem or an internal NonStop TCP/IPv6 subsystem problem.

Effect None

Recovery Determine the type and the cause of the failed registration from the err-code value and take the appropriate action.

218

Device lif-name MAC Address Changed, Operator Intervention Required, Subnet will remain in the STARTING state.

lif-name

is the LIF name.

Cause A LAN adapter was replaced, but the operator did not first use SCF to STOP the subnet that was using the LIF on the adapter.

Effect The subnet is unusable until the operator takes corrective action.

Recovery Using SCF, the operator must first STOP the subnet, then DELETE it, re‑ADD it, and START it again.

219

ND6HOSTD Process Running: function-name, programfilename = program-filename, tcip = tcpip-name. resconf = resconf-filename, debug = debug-info

`function-name`	is name of the function in which the event was generated.
`program-filename`	name of the program file for the process.
`tcpip-name`	is the name of the TCPIPv6 process used
`resconf-filename`	is the name of the resconf file.
`debug-info`	indicates whether debugging information will be displayed. A value of 0 indicates that debugging is not enabled. A value of 1 indicates that debugging is enabled.

Cause A running process displays configuration information.

Effect None

Recovery This is an informational message only; no operator intervention is required.

220

ND6HOSTD Process Terminating. function-name, text

`function-name`	is name of the function in which the event was generated.
`text`	describes the reason the process is terminating. Possible values are: Operator Initiated and Process Abending

This event is NonCritical when the termination is operator-initiated. It is Critical when the process terminates abnormally.

Cause The ND6HOSTD process is terminating because of an unrecoverable error or because of an operator initiated request.

Effect The ND6HOSTD process is stopped.

Recovery If the ND6HOSTD is configured as a persistent process, the persistence manager restarts the process. If the process terminated due to an unrecoverable error condition, an EMS event message generated prior to this event describes the error condition.

221

ND6HOSTD Configuration Error: function-name: config-err-info, params

function-name

is name of the function in which the event was generated.

config-err-info

specifies the configuration error. The values are:

Text Message	Meaning
Invalid number of arguments	An invalid number of arguments was found in the STARTUPMSG.
Invalid argument	An unexpected argument or an invalid argument was found in the STARTUPMSG.
Resconf file not specified	The -f argument was specified but was not followed by the resconf file name.
TCPIP name not specified	The -t argument was specified but it was not followed by the TCP/IP process name.
TCPIP name invalid	The format of the TCP/IP name specified for the -t argument is invalid.
DEFINESETATTR( ) failed	The call to DEFINESETATTR to set the resconf file name for the DEFINE "=TCPIP^RESOLVER^NAME" failed. The format of the resconf file name may be invalid.

params

are values whose number and meaning depend on the value of config‑err‑info:

`config-err-info` Value	`params` Displayed	Description
Invalid number of arguments	`usage`	Displays the expected format of the start-up message.
Invalid argument	`usage`	Displays the expected format of the start-up message.
Resconf file not specified	`usage`	Displays the expected format of the start-up message.
TCPIP name not specified	`usage`	Displays name found in the STARTUPMSG.
TCPIP name invalid	tcpip=`tcipip-name`	Displays name found in the STARTUPMSG.
DEFINESETATTR() failed	`error=err-num`	Displays the error number returned by DEFINESETATTR()
	`attr=attr`	Displays the define attribute being initialized
	`value=value`	Displays the attribute value

Cause A configuration error was detected. The config-info field indicates the cause.

Effect The process terminates or stops.

Recovery Check for errors in the STARTUPMSG and restart the process.

222

ND6HOSTD Process Internal Error: function-name: int-err-info, params

function-name

is name of the function in which the event was generated.

int-err-info

indicates the nature of the internal error. The values are:

Text Message	Meaning
`guardian-proc` failed	A Guardian Procedure was called and an unexpected error was returned.
Process not started by super.group	The process was not started by a user who is a member of the super group.
Unexpected filenum	A Guardian File System I/O operation completed, and the file number returned was not expected.
malloc( ) failed	A call to malloc( ) to allocate memory failed.
Unknown DNS update operation	The DNS update operation was being performed but the operation was of an unknown type.
Unexpected AWATIOX tag	A Guardian File System I/O operation completed, but the tag returned for the AWAITIOX completion did not match the expected value.

params

are values whose number and meaning depend on the value of int-err-info:

int-err-info Value params Displayed Description

guardian-proc failed

guardian-proc

Specifies the name of the Guardian procedure called. Possible values are:

AWAITIOX( )

CANCELTIMEOUT( )

DEFINEADD( )

DEFINESETATTR( )

FILE_OPEN_( )

PROCESS_GETINFO_( )

PROCESS_NULLIT_( )

READX( )

SETMODE( )

SIGNALTIMEOUT( )

error=err-num Specifies the error returned by the call to the Guardian procedure (if available).For the call to AWAITIOX, a file number is displayed in place of an error number.For the call to SIGNALTIMEOUT( ), the wait time and tag value is returned.For the call to CANCELTIMEOUT( ), the condition code status is displayed.

detail=err-dtl Provides the detailed error information (if available).

Process not started by super.group access=creator‑access Specifies the creator access obtained from PROCESS_GETINFO_( ).

Unexpected filenum file=file-num Specifies the unexpected file number.

malloc() failed size=buffer-size Indicates the size of memory being allocated

Unknown DNS update operation if=interface Specifies the name of the interface.

op=operation Specifies the operation being performed.

host=host-name Specifies the host name.

ip=ip-addr Specifies the IP address.

Unexpected AWATIOX tag receieved=rec-tag Specifies the tag value returned from AWAITIOX.

expected=exp-tag Specifies the tag value expected.

cmd=cmd Specifies the command issued.

Cause ND6HOSTD detects an unrecoverable error due to an internal error condition.

Effect The ND6HOSTD process is terminated abnormally or stopped.

Recovery Contact your service provider.

223

ND6HOSTD Socket Operation Failed: function-name: socket-err-info, params

function-name

is name of the function in which the event was generated.

socket-err-info

indicates the nature of the socket error. The values are:

Text Message	Meaning
socket_nw( ) failed	The socket_nw( ) function failed with an error. The requested socket was not created.
nsk_setsock_opt( ) failed	The shell procedure to perform the setsocket option request failed with an error.
recvfrom_nw( ) failed	The recvfrom_nw( ) operation performed on a socket failed with an error.
Could not access tcpip	A socket operation was performed, but the operation failed because the TCPIP process was not accessible.
nsk_siocgifconf( ) failed	The shell procedure to retrieve the configuration interface for the interfaces failed with an error.
nsk_socket_ioctl( ) failed	The shell procedure to issue an ioctl( ) request failed with an error.

params

are values whose number and meaning depend on the value of socket‑err‑info:

socket-err-info Value params Displayed Description

socket_nw( ) failed errno=err-num Displays the error number returned by the socket library function

err-text Displays the error text describing the error

socket=socket

Identifies the socket being created. The possible values are:

INET6
Socket is used to receive RA.
IP6M
Socket is used to receive configuration information.
AF_ROUTE
Socket is used to receive configuration change messages.

tcpip=tcpip‑name Displays the TCP/IP process name

nsk_setsock_opt( ) failed opt=option Specifies the name of the socket option. The values are:ICMP6_FILTER IPV6_MULTICAST HOPS IPV6_MULTICAST LOOP IPV6_PKTINFO IPV6_RXHDR

value=op-value Specifies the value of the option

recvfrom_nw( ) failed error=err-num Displays the error returned from recvfrom_nw( )

socket=socket Identifies the socket in use

Could not access tcpip tcpip tcpip‑name Specifies the name of the TCPIP process in use

error=err-num Displays the error returned from AWAITIOX( )

nsk_siocgifconf( ) failed index=index Specifies the interface index

nsk_socket_ioctl( ) failed if=interface‑name Specifies the interface name (if known)

cmd = cmd Indicates the socket command issued. The values are:SIOCGIFTYPE SIOCGIFFLAGS SIOCIPV6GIFADDR SIOCIPV6GIFATTR

Cause A socket I/O operation was performed, and the operation completed with an error message.

Effect The process abnormally terminates, then restarts.

Recovery This is an informational message only; no corrective action is needed.

224

ND6HOSTD Update Failed: function-name: update-err-info, params

function-name

is name of the function in which the event was generated.

update-err-info

indicates the nature of the socket error. The values are:

Text Message	Meaning
Cannot allocate ns_updrec T_AAAA	Allocation of space for an update record failed.
Cannot allocate ns_updrec PTR	Allocation of space for an update record failed.
gethostname ( ) failed	The host name, which is required for the performance of the DNS update, was not defined or was not fully qualified.
res_init( ) failed	The host name, which is required for the performance of the DNS update, was not defined or was not fully qualified.
Cannot get fully-qualified domain name	The host name, which is required for the performance of the DNS update, was not defined or was not fully qualified.
Failed zone update T_AAAA	The DNS update operation for the record failed.
Failed zone update T_PTR	The DNS update operation for the record failed
Failed zone update T_AAAA for brother	The DNS update operation for the record failed
Failed zone update T_PTR for brother	The DNS update operation for the record failed

params

are values whose number and meaning depend on the value of update‑err‑info:

`update-err-info` Value	`params` Displayed	Description
Cannot allocate ns_updrec T_AAAA	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	op=`operation`	Identifies the update operation being performed. Possible values are:ADD DELETE MODIFY
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
Cannot allocate ns_updrec PTR	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	op=`operation`	Identifies the update operation being performed. Possible values are:ADD DELETE MODIFY
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
gethostname ( ) failed	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
res_init( ) failed	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
Cannot get fully-qualified domain name	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
Failed zone update T_AAAA	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	op=`operation`	Identifies the update operation being performed. Possible values are:ADD DELETE MODIFY
	rcode=`return‑code`	Specifies the code returned by the resolver update function
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
Failed zone update T_PTR	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	op=`operation`	Identifies the update operation being performed. Possible values are:ADD DELETE MODIFY
	rcode=`return‑code`	Specifies the code returned by the resolver update function
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
Failed zone update T_AAAA for brother	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	op=`operation`	Identifies the update operation being performed. Possible values are:ADD DELETE MODIFY
	rcode=`return‑code`	Specifies the code returned by the resolver update function
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed
Failed zone update T_PTR for brother	if=`interface`	Specifies the name of the interface
	host=`host-name`	Specifies the name of the host for which the record is being generated
	op=`operation`	Identifies the update operation being performed. Possible values are:ADD DELETE MODIFY
	rcode=`return‑code`	Specifies the code returned by the resolver update function
	ip=`ip-addr`	Specifies the IP address of the host for which the update is being performed

Cause Allocation of space for an update record failed.

Effect The update for the DNS record is not performed.

Recovery If update-err-info is Cannot allocate ns_updrec T_AAAA or Cannot allocate ns_updrec PTR, no operator intervention is needed. The process restarts.

If update-err-info is gethostname ( ) failed, res_init( ) failed, or Cannot get fully-qualified domain name:

Make sure that the host name was specified in the TCP6MON configuration.
Make sure that the resconf file specifies a domain name.

If update-err-info indicates a failed zone update, make sure that the DNS server is reachable and operating correctly.

Table 116-1 lists the filter registration/deregistration types appearing in the type field of messages 17,19, 20, 214, 216, and 217.

Table 116-1 Filter Registration/De-registration Types

Value	Type	Category		Value	Type	Category
1	ARP request	Subnet		17	UDP6 client	Port Level
2	ARP response	Subnet		18	UDP4 server with IP	Port Level
3	MMAC	Subnet		19	UDP6 server with IP	Port Level
4	SNAP	Subnet		20	UDP server any	Port Level
5	IP broadcast	Subnet		21	UDP6 server any	Port Level
6	IP default	Subnet		22	ICMP default	Subnet
7	IP alias	Subnet		23	IP fragmentation	Subnet
8	TCP default	Subnet		24	UDP Broadcast	Subnet
9	UPD default	Subnet		26	IP default (IPv6)	Subnet
10	TCP4 client	Port Level		27	TCP default (IPv6)	Subnet
11	TCP6 Client	Port Level		28	UDP default (IPv6)	Subnet
12	TCP4 server with IP	Port Level		29	IPv6 Fragmentation	Subnet
13	TCP6 server with IP	Port Level		30	ICMP6 Default	Subnet
14	TCP server any	Port Level	13	31	IPv6 MMMAC link local	Subnet
15	TCP6 server any	Port Level		32	IPv6 MMMAC solicited	Subnet
16	UDP4 Client	Port Level		33	IPv6 MMMAC node local	Subnet