Chapter 4. Connectivity 135
IST1802I P4 CURRENT = 0 AVERAGE = 0 MAXIMUM = 0
IST1221I DATA DEV = EB03 STATUS = ACTIVE STATE = N/A
IST1724I I/O TRACE = OFF TRACE LENGTH = *NA*
The DYNAMICXCF configuration created a SAMEHOST TRLE named IUTSAMEH. The
related TRLE status can be displayed, as shown in Example 4-23.
Example 4-23 D NET,TRL,TRLE=IUTSAMEH
D NET,TRL,TRLE=IUTSAMEH
IST097I DISPLAY ACCEPTED
IST075I NAME = IUTSAMEH, TYPE = TRLE 491
IST1954I TRL MAJOR NODE = ISTTRL
IST486I STATUS= ACTIV, DESIRED STATE= ACTIV
IST087I TYPE = LEASED , CONTROL = MPC , HPDT = YES
IST1715I MPCLEVEL = HPDT MPCUSAGE = SHARE
IST1717I ULPID = TCPIPA
IST314I END
4.4.6 Problem determination
Isolating network problems is an essential step to verify a connectivity problem in your
environment. This section introduces commands and techniques that can use to diagnose
network connectivity problems related to a specific interface.
The following diagnostic commands are available for either the z/OS UNIX environment or
the TSO environment:
? PING - The PING command can be very useful to determine if a destination address can
be reached in the network. Based on the results, it is possible to define whether the
problem is related to the interface being tested, or whether it is a network-related problem.
Using PING we can verify the following:
? The directly attached network is defined correctly.
? The device is properly connected to the network.
? The device is able to send and receive packets on the network.
? The remote host is able to receive and send packets.
When a PING command is issued, you can receive any of the responses listed in Table 4-4.
Table 4-4 Using PING command as a debugging tool
PING command (direct
network)
PING response Possible cause and actions
ping 10.10.2.1 (intf
osa2080lnk
C S V 1 R 7 : P i n g i n g h o s t 1 0 . 1 0 . 2 . 1
sendMessage(): EDC8130I Host
cannot be reached.
The interface being tested has
a problem. Use the Netstat
command to verify the interface
status.
ping 10.10.2.1 (intf
osa2080lnk
CS V1R7: Pinging host 10.10.2.1
Ping #1 timed out
It means the ICMP packet has
been sent to the network, but
the destination address is either
invalid or it is not able to answer.
Correct the destination address
or verify the destination host
status. This problem should be
verified in the network.
136 Communications Server for z/OS V1R7 TCP/IP Implementation, Volume 1 - Base Functions, Connectivity, and Routing
? Netstat - We can use the Netstat command to verify our TCP/IP configuration. The
information provided in the output from the Netstat command should be checked against
the values in our configuration data sets for the TCP/IP stack. To verify our connectivity
status from an interface perspective, we can use the following Netstat options:
– Netstat HOME/-h shows all defined interfaces and their IP addresses, even those
interfaces dynamically created, as shown in Example 4-24.
Example 4-24 NETSTAT HOME command results
D TCPIP,TCPIPA,N,HOME
EZD0101I NETSTAT CS V1R7 TCPIPA 109
HOME ADDRESS LIST:
LINKNAME: STAVIPA1LNK
ADDRESS: 10.10.1.230
FLAGS: PRIMARY
LINKNAME: OSA2080LNK
ADDRESS: 10.10.2.232
FLAGS:
LINKNAME: IUTIQDF4LNK
ADDRESS: 10.10.4.234
FLAGS:
LINKNAME: EZASAMEMVS
ADDRESS: 10.20.10.100
FLAGS:
LINKNAME: IQDIOLNK0A140A64
ADDRESS: 10.20.10.100
FLAGS:
LINKNAME: LOOPBACK
ADDRESS: 127.0.0.1
FLAGS:
INTFNAME: LOOPBACK6
ADDRESS: ::1
TYPE: LOOPBACK
FLAGS:
12 OF 12 RECORDS DISPLAYED
– Netstat DEVLINKS/-d shows the status of each interface, physical and logical, defined
in the TCP/IP stack, as shown in Example 4-25 (only one interface shown as an
example).
Example 4-25 NETSTAT DEVLINKS command results
DISPLAY TCPIP,TCPIPA,N,DEV,INTFN=OSA2080LNK
EZD0101I NETSTAT CS V1R7 TCPIPA 180
DEVNAME: OSA2080 DEVTYPE: MPCIPA
DEVSTATUS: NOT ACTIVE
LNKNAME: OSA2080LNK LNKTYPE: IPAQENET LNKSTATUS: NOT ACTIVE
NETNUM: N/A QUESIZE: N/A
IPBROADCASTCAPABILITY: NO
CFGROUTER: NON ACTROUTER: UNKNOWN
ACTMTU: UNKNOWN
SECCLASS: 255
BSD ROUTING PARAMETERS:
ping 10.10.2.1 (intf
osa2080lnk
CS V1R7: Pinging host 10.10.2.1
Ping #1 response took 0.000
seconds.
This is the expected response.
The interface is working.
PING command (direct
network)
PING response Possible cause and actions
Chapter 4. Connectivity 137
MTU SIZE: 1492 METRIC: 00
DESTADDR: 0.0.0.0 SUBNETMASK: 255.255.255.0
MULTICAST SPECIFIC:
MULTICAST CAPABILITY: UNKNOWN
GROUP REFCNT
----- ------
224.0.0.1 0000000001
LINK STATISTICS:
BYTESIN = 10892725
INBOUND PACKETS = 61972
INBOUND PACKETS IN ERROR = 117576
INBOUND PACKETS DISCARDED = 0
INBOUND PACKETS WITH NO PROTOCOL = 0
BYTESOUT = 1368
OUTBOUND PACKETS = 8
OUTBOUND PACKETS IN ERROR = 4
OUTBOUND PACKETS DISCARDED = 0
1 OF 1 RECORDS DISPLAYED
END OF THE REPORT
– Netstat ARP/-R (for OSA-Express devices) - Use this command to query the ARP
cache for a given address. Use this command when the remote host does not answer
as expected, to check if an ARP entry has been created for the remote host. It also
allows you to check if the relationship between the IP and MAC address is the
expected one. The resulting display is shown in Example 4-26.
Example 4-26 D TCPIP,TCPIPA,N,ARP command results
DISPLAY TCPIP,TCPIPA,N,ARP
EZD0101I NETSTAT CS V1R7 TCPIPA 266
QUERYING ARP CACHE FOR ADDRESS 10.10.2.224
LINK: OSA2080LNK ETHERNET: 00096B1A7490
QUERYING ARP CACHE FOR ADDRESS 10.10.2.222
LINK: OSA2080LNK ETHERNET: 00096B1A7490
QUERYING ARP CACHE FOR ADDRESS 10.10.2.244
LINK: OSA2080LNK ETHERNET: 00096B1A7490
QUERYING ARP CACHE FOR ADDRESS 10.10.2.242
LINK: OSA2080LNK ETHERNET: 00096B1A7490
QUERYING ARP CACHE FOR ADDRESS 10.10.2.232
LINK: OSA2080LNK ETHERNET: 00096B1A7490
QUERYING ARP CACHE FOR ADDRESS 10.10.2.1
LINK: OSA2080LNK ETHERNET: 00097B61E48A
QUERYING ARP CACHE FOR ADDRESS 10.10.3.245
LINK: OSA20A0LNK ETHERNET: 00096B1A74C2
The described commands can help you locate connectivity problems. If they do not, the next
step to debug a direct attached network problem would be to gather documentation that
would show more detailed information about traffic problems related to the interface and
network.
To get this detailed information, the z/OS V1R7.0 Communications Server typically uses the
component trace to capture event data and save it to an internal buffer, or writes the internal
buffer to an external writer, if requested. You can later format these trace records using the
Interactive Problem Control System (IPCS) subcommand CTRACE.
138 Communications Server for z/OS V1R7 TCP/IP Implementation, Volume 1 - Base Functions, Connectivity, and Routing
To debug a network connectivity problem you can use the Component trace with either of the
two specific components, as follows:
? SYSTCPIP component trace with options:
– VTAM, which shows all of the nondata-path signaling occurring between the devices
and VTAM
– VTAMDATA, which shows data-path signaling between the devices and VTAM,
including a snapshot of media headers and some data
? SYSTCPDA component trace, used with the VARY TCPIP,PKTTRACE command. You
can use the PKTTRACE statement to copy IP packets as they enter or leave TCP/IP, and
then examine the contents of the copied packets.
For more information about how to set up and activate a CTRACE, refer to Appendix A,
“Component trace (CTRACE)” on page 209.
Attention: Using this option slows performance considerably; therefore, it should be
used with caution.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.