 |
» |
|
|
|
 Turn on the power to your
system. See the system manual for your system's startup procedure
and any problem solving information.  Edit the /etc/rc.log file and search for the string hpapa. This takes
you to the beginning of the HP APA start-up section. If there are
failover groups configured on your system, the following message indicates
an error: ERROR: lanapplyconf failed |
If there are no failover groups configured on your system,
you can ignore the error message.If you
see error messages about link aggregate attributes, do the following: Stop HP APA with the following
commands: # /sbin/init.d/hplm stop
# /sbin/init.d/hpapa stop |
Correct the attribute in the
appropriate configuration file. Restart HP APA with the following
commands: # /sbin/init.d/hpapa start
# /sbin/init.d/hplm start |
Search for the string hplm in
the /etc/rc.log file. If there are failover groups
configured on your system, the preceding message indicates an error.
If there are no failover groups configured on your system, you can
ignore the error message. Also look for any driver and network errors.  Do the following: Verify that the network card
is seated correctly and that it is operational. Check that your network physical
port connectors between the NIC and the switch (or wall plug) are
fully connected.
 Verify that the NIC has
been claimed by the operating system. Issue the following command: If the driver has been claimed,
the driver name appears in the Driver column and CLAIMED appears in the S/W State column
for the lan class and hardware path. See “Scanning the System Hardware” for more information
on ioscan. If the software state is UNCLAIMED, do the following: If the class is unknown and driver is UNKNOWN, the interface driver has
not been generated into the kernel. Verify that the /stand/vmunix file contains the appropriate keyword for the NIC you are using
with the following command: # what /stand/vmunix | grep -i driver-name |
For example, if you
are using the 1000Base-T NIC, the igelan keyword
appears. If it does not, you must edit the file and create a new kernel.
For information on how to do this, see the Installing and
Administering LAN/9000 Software manual. Verify that the NIC is seated
correctly and that it is operational. Reboot the system.
 Verify that the NIC is
operating. Enter the following command: # nwmgr -v [-c lanx | -S driver-name] |
If the Interface State is DOWN, do the following: Check the status of the Link
LED. If the LED is off, check the connection
to the switch. Make sure that the switch is configured in the correct
mode and is autonegotiating, if necessary. Reset the NIC with following
command: Make sure the link speed and
duplex mode match the settings on the switch. If they do not, set
the correct link speed and duplex mode and enable autonegotiating,
if necessary. Then, reset the NIC with the nwmgr command Verify that
your interface state is UP by issuing the nwmgr command with no arguments. If it is not up, perform the steps for NIC claimed?.
-
 Verify that the network
daemon (inetd) is running. Enter the following
command: If no inetd daemon is running, start it using the following command: # /sbin/init.d/inetd start
|
 Verify that the HP APA
depot is installed by issuing the following command: If the product is installed, the following
message is displayed: J4240AA B.11.31.10 Auto-Port Aggregation Software |
If the product is not installed, install it by
using the swinstall command. See Chapter 2 and the Release Notes
for more information.  Go to “Solving Link Aggregate Problems (MANUAL Mode)”for MANUAL mode
link aggregates, “Solving Link Aggregate Problems (LACP Mode)” for LACP_AUTO mode link aggregates, “Solving Link Aggregate Problems (FEC Mode)” for FEC_AUTO
mode link aggregates, or “Solving Failover Group Problems” for failover groups.Verify that the software
is configured in the kernel by issuing the following command: # what /stand/vmunix | egrep -i hp_apa |
Output similar to the following
displays: $Revision: hp_apa: HP Auto-Port Aggregation (APA): B.11.31.10 Nov 10 2007 11:30 |
If nothing is displayed, rebuild the kernel.
Solving Link Aggregate Problems (MANUAL Mode) | |
 Do the following:  Verify that the link aggregate
is configured by issuing the nwmgr command. Link
aggregate names begin with lan900 and have the
subsystem and interface type as hp_apa. If the link aggregate is not configured, use the System
Management Homepage to configure it. See “Configuring a Link Aggregate” for more information.  Verify that the link aggregate
is enabled by issuing the following command: where xxx is the link aggregate instance number (for example, 900).If the aggregate is enabled, the APA State is Up. If the APA State is Down, do
the following: Check the cabling between
the NIC and the switch. Verify that the link partner
is configured to run manual trunks. Make any changes, if necessary.
Then, verify that the link aggregate is enabled. Verify that all ports in the
aggregate are set to the same MANUAL mode by using the following command: where xxx is the link aggregate interface name (for example, lan900). If they are not, use the System Management Homepage
to modify the link aggregate. See “Modifying a Link Aggregate” for more information.
 Verify the ports in the
link aggregate by issuing the following command: The
Membership column shows the ports that compose the link aggregate.If the link aggregate contains ports that you
do not want in the aggregate, delete the ports by issuing the following
command: # nwmgr -d -A links=port-number -c lanxxx |
where port-number can be a list
of multiple ports numbers separated by commas and xxx is the instance number of the link aggregate (for example, 900).If any ports you wanted in the link aggregate
are members of another link aggregate, delete the ports in the other
aggregate by issuing the following command: # nwmgr -d -A links=port-number -c aggregate-instance |
If there are ports you wanted in the link aggregate
that are not members of the link aggregate, do the following: If FEC or LACP is enabled
on the port, stop autonegotiation with the following command: # netmgr -s -A mode=MANUAL -c lanxxx -S apa |
where xxx is the instance number for the port. Verify the type, speed (including
duplex), and MTU size of each port you want with the following command: All ports must have the same
values. If they do not, make any changes as necessary on the server
and the switch. Check if there is an IP address
or a VLAN on the port.
 Verify the active ports
in the link aggregate by issuing the following command: where xxx is the
link aggregate instance number.The Active
Port(s) field lists the active ports. If there are ports in the Not Ready Port(s) field,
verify the port's operation status by issuing the following command: where xxx is the interface number.If the Operation Status is UP, check the speed
(including duplex) and MTU for the port and another active port in
the link aggregate with the following command: # nwmgr -A speed, mtu -c lanxxx |
If they do not match, correct
the settings on the server and the switch. See the steps for Link partner configured?. If they match, reset the port.If the status is DOWN, check the cabling from
the NIC to the switch and see the steps for NIC operating?. On the switch, verify that all ports you expect
to be in the link aggregate are in the same MANUAL trunk. | | | |  | NOTE: If you are using the HOT_STANDBY
load distribution algorithm, be sure the switch side port is not in
any trunk. | | | | |
 If a remote host's
network is not reachable, the following message is displayed in response
to the ping command: Complete the following steps: Ensure that the network devices
are configured properly on the local host, using the netstat -i command. Verify that the routing tables
on the local host are correct, using the netstat -r command.
 If a host is not known,
the following message is displayed: Complete the following steps: Verify that the user is trying
to reach the remote host using a valid host name. Verify that the remote host
is in another name domain and that the user specified the full domain
name. If your site uses the Domain
Name System (DNS) for name-to-address translation, look in the /etc/nsswitch.conf file to see if nis is specified as a service for the hosts database
entry. If it is not, add it. Also, verify that the DNS service has
information about the remote host. If your site uses the NIS
name service for name-to-address translation, look in the /etc/nsswitch.conf file to see if nis is specified as a service for the hosts database
entry. If it is not, add it. Also, verify if the NIS service has information
about the remote host. If your /etc/nsswitch.conf file lists files as the only name-to-address
translation mechanism, the /etc/hosts file does
not have information on the remote host. See nsswitch.conf(4) for more information.
 If a remote host is not
reachable, the following message is displayed: Complete the following steps: Inspect the cabling between
the local host and the link partner. Perform a loopback test on
your own system by using the ping command. If this
is successful, your system is operating properly to the Network Layer
(OSI Layer 3). Verify the remote host is
running, using the ping command. If the remote
host does not respond, ask the host's system administrator to
start the host. For additional information about the ping command, see “Testing Access to Internet Network Hosts”. Verify that an entry exists
for the remote host in your system's ARP cache by entering the
following command: If the entry is wrong or incomplete, enter the
correct station address by using the arp command.
See arp(1M) for more information. Make sure the network devices
are configured properly on the local host, using the netstat
-i command. Verify that the routing tables
on the local host are correct, using the netstat -r command. Use the ping command to determine whether
the IP router is reachable. Verify that the local host's
address-to-name translation for the remote host is correct by using
the nslookup command. Inspect the routers along
the path to the remote host to determine whether they have security
features enabled that prevent you from reaching the remote host.
 If a file cannot be accessed
using the rcp or rsh command,
the following message is displayed: Complete the following steps: Verify that the user is intended
to have access to the remote host. The remote host might be intentionally
preventing remote access. Verify that the correct host
and user definitions exist in the user's .rhosts file on the remote host. Verify that the /etc/hosts.equiv file is set up correctly. Verify that the directory
and file protection on the files to be copied or the .rhosts file on the remote system are correct. If you are using NFS, see
the appropriate NFS documentation.
 If the telnet command is not successful, there is a problem with the Transport
Layer (OSI Layer 4). Do the following: Verify whether you have a
problem with a pseudoterminal driver (pty) on your system by issuing
the ftp command to the remote host. If the command
is successful, there is a problem with a pty on your system. Contact
your HP representative. See “Reporting Problems ” for more information. Check the /etc/protocols file on both hosts to ensure that TCP is installed and configured.
Neither telnet nor ftp will
work if TCP is not configured on both sides of the connection. If it is not, install TCP and issue the telnet command again. Transfer a file to a different
remote host on the network. Use the netstat command
to check for lost packets. If network congestion
does not appear to be the cause, contact your HP representative. See “Reporting Problems ” for more information.
  Problem still exists? Report
your problem to HP. See “Reporting Problems ”.If the connection terminates
abnormally or a network application appears to hang, complete the
following steps: Test the network to determine
whether the problem is on the local host, remote host, or a host on
the path between the two. After you identify the host
with the problem, do the following: Confirm that the NIC is properly
configured. Verify that the broadcast address and address mask for
the local host are correct. Make sure the local host's /etc/hosts file has the correct IP address for the local
host. Make sure the cabling from
the local host to the network is intact and properly connected. If connected over a local
area network (LAN), verify that the Address Resolution Protocol (ARP)
entries are correct and that the system is properly connected to the
LAN. Check the /var/adm/nettl.LOG000 file to see if there are any errors (for example, cable disconnection
messages). Format the nettl.LOG000 file using
the following command:
Solving Link Aggregate Problems (LACP Mode) | |
Verify that the switch
is configured correctly with the appropriate ports enabled and autonegotiation
set to ON. See the appropriate switch documentation to determine if
the switch ports connected to the server are configured correctly
for link aggregation (trunking) and support LACP. Verify that the link aggregate
is configured by issuing the nwmgr command. Link
aggregate names begin with lan900 and have the
Subsystem and Interface Type as hp_apa. If the link aggregate is not configured, use the System
Management Homepage to configure it. See “Configuring a Link Aggregate” for more information. Verify that the link aggregate
is enabled by issuing the following command: where xxx is the
link aggregate instance number (for example, 900).If the aggregate is enabled, the APA State is Up. This also indicates that any LACP automatic protocol negotiation
was successful. If the APA State is Down, do
the following: Verify that all ports in the
aggregate are set to the same mode by using the following command: # nwmgr -A mode,key -I xxx -S apa |
where xxx is the port instance number (for example, 1, 2, or 6). If they are not, use the System Management Homepage
to modify the link aggregate. See “Modifying a Link Aggregate” for more information. Verify that the link partner
is configured to run the IEEE 802.3ad LACP protocol. Make any changes,
if necessary, and wait 10 seconds. Then, verify that the link aggregate
is enabled.
Verify the ports in the
link aggregate by issuing the following command: # nwmgr -A all -c lanxxx -S apa |
where xxx is
the link aggregate instance number (for example, 900).The Membership column shows the ports that compose the
link aggregate. If the link aggregate contains ports that you
do not want in the aggregate, delete the ports by issuing the following
command: # nwmgr -d -A links=port-number -c lanxxx |
where port-number can be a list
of multiple ports numbers separated by commas and xxx is the instance number of the link aggregate (for example, 900).If any ports you wanted in the link aggregate
are members of another link aggregate, delete the ports in the other
aggregate by issuing the following command: # nwmgr -d -A links=port-number -c aggregate-instance |
If there are ports you wanted in the link aggregate
that are not members of the link aggregate, do the following: Verify the type, speed, duplex
mode, MTU size, and checksum offload capability of each port you want
with the following command: where xxx is the instance number
for the port. All ports must have the same values. | | | | | NOTE: You cannot aggregate a port that has no CKO capability
with a port that has CKO capability, even if its CKO capability is
currently disabled. | | | | |
To change an attribute, do the following: Stop LACP on the port with
the following command: # nwmgr -s -A mode=MANUAL -c lanxxx |
Set the attribute value with
the following command: # nwmgr -s -A attribute=value -c lanxxx |
Start LACP on the port with
the following command: # nwmgr -s -A mode=LACP_AUTO -c lanxxx |
Verify that the port is in
LACP mode with the following command: # nwmgr -A mode -c lanxxx |
If the port is not in LACP mode,
possible reasons are: The port has IP addresses or VLANS configured on it.
Unconfigure them from the port and start LACP again. The port is not in full-duplex mode. Make sure the
port on the server and on the switch are set to full-duplex or to
autonegotiate. Start LACP again. The port is already in a link aggregate, failover
group, or in FEC_AUTO mode. Find another port to use. The port is not an Ethernet port or is not supported
by HP APA. The port is down. Wait for the port to come up. Verify
that the port is connected to the switch correctly. If it is, reset
the port.
Save the current configuration
with the following command: # nwmgr -s -A all --fr cu --sa -S apa |
Verify that the link aggregate
has a key value specified with the nwmgr -A key -c lanxxx command, where xxx is the instance number for the link aggregate (for example, 900
for lan900). If it does not, set it with the following command: # nwmgr -s -A key=value -c lanxxx |
Verify that the port's
key value matches the value of the link aggregate. If it does not,
do the following: Set the port mode to MANUAL
with the following command: # nwmgr -s -A mode=MANUAL -I port-number -S apa |
Set the key value and mode
with the following command: # nwmgr -s -A key=value, mode=LACP_AUTO -I port-number -S apa |
The ports then
negotiate with the link partner (approximately 10 seconds) and join
the link aggregate. Repeat this step for each port you want in the
aggregate. Verify that the ports are
in the link aggregate with the nwmgr -S apa command. Save the current configuration
with the following command: # nwmgr -s -A all --fr cu --sa -S apa |
Verify the active ports
in the link aggregate by issuing the following command: where xxx is
the link aggregate instance number.The
Active Port(s) field lists the active ports. If there are ports in the Not Ready Port(s) field,
verify the port's operation status by issuing the following command: where xxx is the interface number.If the Operation Status is UP, see the steps for Link partner configured?. If the status is down, check the cabling from
the NIC to the switch and see the steps for NIC operating?. If a remote host's
network is not reachable, the following message is displayed in response
to the ping command: Complete the following steps: Ensure that the network devices
are configured properly on the local host, using the netstat -i command. Verify that the routing tables
on the local host are correct, using the netstat -r command.
If a host is not known,
the following message is displayed: Complete the following steps: Verify that the user is trying
to reach the remote host using a valid host name. Verify that the remote host
is in another name domain and that the user specified the full domain
name. If your site uses the Domain
Name System (DNS) for name-to-address translation, look in the /etc/nsswitch.conf file to see if nis is specified as a service for the hosts database
entry. If it is not, add it. Also, verify that the DNS service has
information about the remote host. If your site uses the NIS
name service for name-to-address translation, look in the /etc/nsswitch.conf file to see if nis is specified as a service for the hosts database
entry. If it is not, add it. Also, verify if the NIS service has information
about the remote host. If your /etc/nsswitch.conf file lists files as the only name-to-address
translation mechanism, the /etc/hosts file does
not have information on the remote host. See nsswitch.conf(4) for more information.
If a remote host is not
reachable, the following message is displayed: Complete the following steps: Inspect the cabling between
the local host and the link partner. Perform a loopback test on
your own system by using the ping command. If this
is successful, your system is operating properly to the Network Layer
(OSI Layer 3). Verify the remote host is
running, using the ping command. If the remote
host does not respond, ask the host's system administrator to
start the host. For additional information about the ping command, see “Testing Access to Internet Network Hosts”. Verify that an entry exists
for the remote host in your system's ARP cache by entering the
following command: If the entry is wrong or incomplete, enter the
correct station address by using the arp command.
See arp(1M) for more information. Make sure the network devices
are configured properly on the local host, using the netstat
-i command. Verify that the routing tables
on the local host are correct, using the netstat -r command. Use the ping command to determine whether
the IP router is reachable. Verify that the local host's
address-to-name translation for the remote host is correct by using
the nslookup command. Inspect the routers along
the path to the remote host to determine whether they have security
features enabled that prevent you from reaching the remote host.
If a file cannot be accessed
using the rcp or rsh command,
the following message is displayed: Complete the following steps: Verify that the user is intended
to have access to the remote host. The remote host might be intentionally
preventing remote access. Verify that the correct host
and user definitions exist in the user's .rhosts file on the remote host. Verify that the /etc/hosts.equiv file is set up correctly. Verify that the directory
and file protection on the files to be copied or the .rhosts file on the remote system are correct. If you are using NFS, see
the appropriate NFS documentation.
If the telnet command is not successful, there is a problem with the Transport
Layer (OSI Layer 4). Do the following: Verify whether you have a
problem with a pseudoterminal driver (pty) on your system by issuing
the ftp command to the remote host. If the command
is successful, there is a problem with a pty on your system. Contact
your HP representative. See “Reporting Problems ” for more information. Check the /etc/protocols file on both hosts to ensure that TCP is installed and configured.
Neither telnet nor ftp will
work if TCP is not configured on both sides of the connection. If it is not, install TCP and issue the telnet command again. Transfer a file to a different
remote host on the network. Use the netstat command
to check for lost packets. If network congestion
does not appear to be the cause, contact your HP representative. See “Reporting Problems ” for more information.
If the connection terminates
abnormally or a network application appears to hang, complete the
following steps: Confirm that the NIC is properly
configured. Make sure the cabling from
the local host to the network is intact and properly connected. Test the network connectivity
between the link aggregate and other hosts on the same subnet using
the ping command. Verify the link aggregate
membership. See the steps for Link aggregate
enabled?. Verify that there is an IP
address configured on the link aggregate and that the IP interface
is up with the ifconfig command. If it is not up,
bring it up with the following command: # ifconfig interface-name up |
Test the connection with the ping command. Test the network connectivity
between another interface on the server and other hosts on the same
subnet. If this fails, there is a network problem. Examine the traffic over the
link aggregate. Heavy traffic might cause the LACP system to lose
communication, which can cause the link aggregate to lose connection
with other ports or cause the LACP link aggregate to go down temporarily.
Reduce the traffic on the link aggregate.
 Problem still exists? Report
your problem to HP. See “Reporting Problems ”.If you suspect or know
that you are experiencing packet loss over your connection, do the
following: LACP mode link aggregates
require time to negotiate with the switch. You might drop some packets
during this time. Check the trunk is configured
correctly on the switch. Make sure that ports connected to the ports
on the server are assigned to the LACP trunk. If the LACP trunk is
assigned to any other ports on the server, this can cause packets
to be dropped. Check the switch port configuration.
Remove any port that is not supposed to be in the trunk. Verify the link aggregate
membership. See the steps for Link aggregate
enabled?. If the number of ports in the link aggregate
is greater than the maximum number of ports in the LACP trunk on the
switch, delete the extra ports from the LACP link aggregate.
Solving Link Aggregate Problems (FEC Mode) | |
Verify that the switch
is configured correctly. See the appropriate switch documentation
to determine if the switch ports connected to the server are configured
correctly for link aggregation (trunking) and support the FEC protocol. Verify that the link aggregate
is configured by issuing the nwmgr command. Link
aggregate names begin with lan900 (lan900, lan901, ...) and have the
Subsystem as hp_apa. If the link aggregate is not configured, use the System Management
Homepage to configure it. See “Configuring a Link Aggregate” for more information. Verify that the link aggregate
is configured by issuing the lanscan command. Link
aggregate names begin with LinkAgg. If the link aggregate is not configured, use SAM to
configure it. See “Configuring a Link Aggregate” for more information. Verify that the link aggregate
is enabled by issuing the following command: where xxx is
the link aggregate instance number.If the
aggregate is enabled, the APA State is Up. This
also indicates that the FEC automatic protocol negotiation was successful. If the APA State is Down, do
the following: Verify that all ports in the
aggregate are set to the same mode using the following command: # nwmgr -A mode,gc -I xxx -S apa |
where xxx is the port instance number (for example, 1,4, and 10). If they are not, use the System Management Homepage
(SMH) to modify the aggregate. See “Modifying a Link Aggregate” for more information. Verify that the link partner
is configured to run the Cisco Fast EtherChannel protocol. Make any
changes, if necessary, and wait 10 seconds. Then, verify that the
link aggregate is enabled.
Verify the ports in the
link aggregate by issuing the following command: # nwmgr -A all -c lanxxx -S apa |
where xxx is
the link aggregate instance number.The
Membership column shows the ports that compose the link aggregate. If the link aggregate contains ports that you
do not want in the aggregate, delete the ports by issuing the following
command: # nwmgr -d -A links=port-number -c aggregate-instance |
where port-number can be a list of multiple ports numbers separated by commas and aggregate-instance is the instance number of the link
aggregate (for example, 900).If any ports you wanted in the link aggregate
are members of another link aggregate, delete the ports in the other
aggregate by issuing the following command: # nwmgr -d -A links=port-number -c aggregate-instance |
If there are ports you wanted in the link aggregate
that are not members of the link aggregate, do the following: If the port is set to LACP
or FEC mode, set the port mode to MANUAL. Note the speed, duplex mode,
MTU size, and checksum offload capability of each port you want. All
ports must have the same values. Verify that the link aggregate
has a group capability value specified. If it does not, set it with
the following command: # nwmgr -s -A gc=value -I aggregate-instance -S apa |
Verify that the port's
group capability value matches the value of the link aggregate. If
it does not, set it and the mode with the following command: # nwmgr -s -A gc=value -A mode=FEC_AUTO -I port-number -S apa |
The ports then negotiate with the link
partner (approximately 10 seconds) and join the link aggregate. Verify that the ports are
in the link aggregate with the nwmgr -S apa command. Save the current configuration
with the following command: # nwmgr -s -A all --fr cu --sa -S apa |
Verify the active ports
in the link aggregate by issuing the following command: where xxx is
the link aggregate instance number.The
Active Port(s) field lists the active ports. If there are ports in the Not Ready Port(s) field,
verify the port's operation status by issuing the following command: where xxx is the port number (for example, 1, 4, or 10).If the Operation Status is UP, see the steps for Link partner configured?. If the status is down, check the cabling from
the NIC to the switch and see the steps for NIC operating?. If a remote host's
network is not reachable, the following message is displayed in response
to the ping command: Complete the following steps: Ensure that the network devices
are configured properly on the local host, using the netstat -i command. Verify that the routing tables
on the local host are correct, using the netstat -r command.
If a host is not known,
the following message is displayed: Complete the following steps: Verify that the user is trying
to reach the remote host using a valid host name. Verify that the remote host
is in another name domain and that the user specified the full domain
name. If your site uses the Domain
Name System (DNS) for name-to-address translation, look in the /etc/nsswitch.conf file to see if nis is specified as a service for the hosts database
entry. If it is not, add it. Also, verify that the DNS service has
information about the remote host. If your site uses the NIS
name service for name-to-address translation, look in the /etc/nsswitch.conf file to see if nis is specified as a service for the hosts database
entry. If it is not, add it. Also, verify if the NIS service has information
about the remote host. If your /etc/nsswitch.conf file lists files as the only name-to-address
translation mechanism, the /etc/hosts file does
not have information on the remote host. See nsswitch.conf(4) for more information.
If a remote host is not
reachable, the following message is displayed: Complete the following steps: Inspect the cabling between
the local host and the link partner. Perform a loopback test on
your own system by using the ping command. If this
is successful, your system is operating properly to the Network Layer
(OSI Layer 3). Verify the remote host is
running, using the ping command. If the remote
host does not respond, ask the host's system administrator to
start the host. For additional information about the ping command, see “Testing Access to Internet Network Hosts”. Verify that an entry exists
for the remote host in your system's ARP cache by entering the
following command: If the entry is wrong or incomplete, enter the
correct station address by using the arp command.
See arp(1M) for more information. Make sure the network devices
are configured properly on the local host, using the netstat
-i command. Verify that the routing tables
on the local host are correct, using the netstat -r command. Use the ping command to determine whether
the IP router is reachable. Verify that the local host's
address-to-name translation for the remote host is correct by using
the nslookup command. Inspect the routers along
the path to the remote host to determine whether they have security
features enabled that prevent you from reaching the remote host.
If a file cannot be accessed
using the rcp or rsh command,
the following message is displayed: Complete the following steps: Verify that the user is intended
to have access to the remote host. The remote host might be intentionally
preventing remote access. Verify that the correct host
and user definitions exist in the user's .rhosts file on the remote host. Verify that the /etc/hosts.equiv file is set up correctly. Verify that the directory
and file protection on the files to be copied or the .rhosts file on the remote system are correct. If you are using NFS, see
the appropriate NFS documentation.
If the telnet command is not successful, there is a problem with the Transport
Layer (OSI Layer 4). Do the following: Verify whether you have a
problem with a pseudoterminal driver (pty) on your system by issuing
the ftp command to the remote host. If the command
is successful, there is a problem with a pty on your system. Contact
your HP representative. See “Reporting Problems ” for more information. Check the /etc/protocols file on both hosts to ensure that TCP is installed and configured.
Neither telnet nor ftp will
work if TCP is not configured on both sides of the connection. If it is not, install TCP and issue the telnet command again. Transfer a file to a different
remote host on the network. Use the netstat command
to check for lost packets. If network congestion
does not appear to be the cause, contact your HP representative. See “Reporting Problems ” for more information.
- Problem still exists? Report
your problem to HP. See “Reporting Problems ”.
If the connection terminates
abnormally or a network application appears to hang, complete the
following steps: Test the network to determine
whether the problem is on the local host, remote host, or a host on
the path between the two. After you identify the host
with the problem, do the following: Confirm that the NIC is properly
configured. Verify that the broadcast address and address mask for
the local host are correct. Make sure the local host's /etc/hosts file has the correct IP address for the local
host. Make sure the cabling from
the local host to the network is intact and properly connected. If connected over a local
area network (LAN), verify that the Address Resolution Protocol (ARP)
entries are correct and that the system is properly connected to the
LAN. Check the /var/adm/nettl.LOG000 file to see if there are any errors (for example, cable disconnection
messages). Format the nettl.LOG000 file using
the following command:
Solving Failover Group Problems | |
Verify that the switch
is configured correctly. Refer to the appropriate switch documentation
to determine if the switch ports connected to the server are configured
correctly for link aggregation (trunking).  Verify that the failover
group is configured by issuing the following command: where xxx is the
aggregate instance number.If the failover
group is not configured (Mode field is Not_Enabled), use the System
Management Homepage to configure it. For more information, see “Configuring a Failover Group”. | | | | | NOTE: One of the ports in the failover group must have an IP address
configured on it. | | | | |
 Verify that the failover
group is up by issuing the following command: # nwmgr -A all -I xxx -S apa |
where xxx is
the failover group instance number.If the
failover group (LAN_MONITOR mode) is enabled, the APA State is Up. If the APA State is Down, do
the following: Check that all ports in the
failover group are connected to the switch. Verify that the link partner
is configured for no trunking. Make any changes, if necessary, and
wait 10 seconds. Then, verify that the failover group is up.
 Verify the ports in the
failover group by issuing the following command: where xxx is the failover group instance number.The Membership column shows the ports that compose the
failover group. If the failover group contains ports that you
do not want, delete the ports by issuing the following command: # nwmgr -d -A links=port-number -c lanxxx |
where port-number can be a list
of multiple ports numbers separated by commas and xxx is the instance number of the failover group (for example, 900).If any ports you wanted in the failover group
are members of another link aggregate, delete the ports in the other
aggregate by issuing the following command: # nwmgr -d -A links=port-number -c lanxxx |
If there are ports you wanted in the failover
group that are not members of the failover group, verify the type,
speed, duplex mode, MTU size, and checksum offload capability of each
port you want with the following command: where xxx is the instance number
for the port. All ports must have the same values. To change an attribute,
do the following:Stop the port (unless it is
a link aggregate) with the following command: # nwmgr -s -A mode=MANUAL -c lanxxx |
Set the attribute value with
the following command: # nwmgr -s -A attribute=value -c lanxxx |
Start the port (unless it
is a link aggregate) with the following command: # nwmgr -s -A mode=LAN_MONITOR -c lanxxx |
Verify that the port is in
LAN_MONITOR mode with the following command: # nwmgr -A mode -c lanxxx |
If the port is not
in LAN_MONITOR mode, possible reasons are: The port has IP addresses or VLANS configured on it.
Unconfigure them from the port and start LAN_MONITOR again. The port is not in full-duplex mode. Make sure the
port on the server and on the switch are set to full-duplex or to
autonegotiate. Start LAN_MONITOR again. The port is already in a link aggregate or failover
group. Find another port to use. The port is not an Ethernet port or is not supported
by HP APA. The port is down. Wait for the port to come up. Verify
that the port is connected to the switch correctly. If it is, reset
the port.
Save the current configuration
with the following command: # nwmgr -s -A all --fr cu --sa -S apa |
 Verify the active ports
in the failover group by issuing the following command: where xxx is the
failover group instance number.The Active
Port(s) field lists the active ports. If there are ports in the Not Ready Port(s) field,
verify the port's operation status by issuing the following command: where xxx is the interface number.If the Operation Status is UP, wait 30 seconds
to see if the port becomes ready or active. If it does not, see the
steps for Link partner configured?. Then, reset
the port with the following command and wait 30 seconds: | | | | | NOTE: Only reset physical ports, not link aggregates that are members
of the failover group. | | | | |
If the port does not appear in the Ready Port(s) list, delete
the port and add it with the following commands:# nwmgr -d -A links=port-number -c lanxxx -S apa
# nwmgr -a -A links=port-number -A mode=LAN_MONITOR -c lanxxx -S apa |
Save the current configuration with the following
command:
|
