Running diagnostics
You can run selected diagnostics on the nodes in an HCP system to help analyze and resolve issues related to interactions between the nodes and other components in the HCP environment. The available diagnostics let you:
- Ping a specified device (ping)
- Display the network path being used for communications between the node and a specified device (traceroute)
- Query the DNS for the records that match a specified IP address or domain name (dig)
- Display the routing table for the node (route)
- Display the NFS exports table for a specified device (showmount)
- On HCP SAIN systems, display information about the logical volumes that are mapped to the node on the Fibre Channel SAN arrays (fchbainfo)
- Display the ethtool command output command for both front-end interfaces (ethtool)
You run diagnostics from the system console for a node. To use the system console, you need a keyboard and monitor.
To run diagnostics, you use the Appliance Diagnostics menu. To display this menu in the system console, press Alt+F8.
Appliance Diagnostics ====================================== [1] ping - Check Network Connectivity [2] traceroute – Check Routing [3] dig - Check DNS Resolution [4] route - List the Kernel Routing Table [5] showmount - Show Exports from NFS Servers [6] fchbainfo - List Fibre Channel HBA and LUN Mapping Information [c] clear - Clear the Screen and Redisplay the Menu Enter a selection:
To run diagnostics from the Appliance Diagnostics menu:
- To execute a diagnostic command, enter the option number for that command. If the command requires an argument, HCP prompts you to enter a valid value for that argument. HCP then executes the command and displays the results on the screen.
- To cancel a command before it finishes, press Ctrl+C.
- If the information displayed in response to a command is longer than the space available on the screen, use Shift+PgUp and Shift+PgDn to scroll the display up or down, respectively.
- To return to the Appliance Diagnostics menu after viewing the information displayed by a command, press Enter.
- To clear the screen and redisplay the Appliance Diagnostics menu at the top, use option c from the menu.
- To return to the console login prompt from the Appliance Diagnostics menu, press Alt+F1.
In addition to being displayed on screen, the results of each diagnostic command that you execute are sent to the syslog servers that you specify on the Syslog page in the HCP System Management Console.
For more information about the diagnostic commands that you can execute from the Appliance Diagnostics menu, see the man pages for the commands on a Linux system.
ping
The ping command pings a device that you identify by IP address or fully qualified domain name (FQDN). You might run this diagnostic, for example, to determine whether the node you’re on can communicate with a specific DNS server, external time server, or external storage device.
The ping command pings the specified device 10 times, displaying the result of the ping each time. However, if the device is unreachable, the command displays only the summary of the ping attempts after the 10 tries.
To execute the ping command:
Procedure
From the Appliance Diagnostics menu, enter 1.
A prompt for an IP address or FQDN appears.This command tries to ping (count=10) an address or FQDN. Please enter the address or FQDN:
Enter the target device IP address or FQDN.
The results of the ping command are displayed on the screen.After viewing the display, press Enter to return to the Appliance Diagnostics menu.
The following is a sample response to the ping command:Pinging 172.20.33.40... PING 172.20.33.40 (172.20.33.40) 56(84) bytes of data. 64 bytes from 172.20.33.40: icmp_req=1 ttl=63 time=0.331 ms 64 bytes from 172.20.33.40: icmp_req=2 ttl=63 time=0.443 ms 64 bytes from 172.20.33.40: icmp_req=3 ttl=63 time=1.93 ms 64 bytes from 172.20.33.40: icmp_req=4 ttl=63 time=2.20 ms 64 bytes from 172.20.33.40: icmp_req=5 ttl=63 time=3.23 ms 64 bytes from 172.20.33.40: icmp_req=6 ttl=63 time=5.28 ms 64 bytes from 172.20.33.40: icmp_req=7 ttl=63 time=0.247 ms 64 bytes from 172.20.33.40: icmp_req=8 ttl=63 time=0.254 ms 64 bytes from 172.20.33.40: icmp_req=9 ttl=63 time=2.87 ms 64 bytes from 172.20.33.40: icmp_req=10 ttl=63 time=4.38 ms --- 172.20.33.40 ping statistics --- 10 packets transmitted, 10 received, 0% packet loss, time 9008ms Rtt min/avg/max/mdev = 0.247/2.120/5.287/1.732 ms
traceroute
The traceroute command displays the network path through which the node you’re on communicates with a device that you identify by IP address or FQDN. You might run this diagnostic, for example, if communication between the node and another device, such as a DNS server, is taking longer than expected.
For each pair of points in the network path, the traceroute command displays the round-trip time between the two points for each of three probes. You can use this information, for example, to determine whether the network topology is configured correctly.
To execute the traceroute command:
Procedure
From the Appliance Diagnostics menu, enter 2. A prompt for an IP address or FQDN appears.
This command tries to trace routing to an address or FQDN. Please enter the address or FQDN:
Enter the target device IP address or FQDN. The results of the traceroute command are displayed on the screen.
After viewing the display, press Enter to return to the Appliance Diagnostics menu.
The following is a sample response to the traceroute command:Traceroute to 192.168.100.45 (192.168.100.45), 30 hops max, 60 byte packets 1 172.20.43.253 (172.20.43.253) 0.286 ms 0.368 ms 0.425 ms 2 2001:0db8::101 (2001:0db8::101) 0.531 ms 0.604 ms 0.669 ms 3 192.168.100.45 (192.168.100.45) 0.339 ms 0.343 ms 0.330 ms
dig
The dig command queries the DNS for records that match an IP address or FQDN that you specify. You might run this diagnostic with an HCP domain name, for example, if you have user-defined networks that are not working as expected.
When you execute the dig command, you are prompted for arguments. Although you can specify any dig command arguments, this document describes only IP addresses and FQDNs.
To execute the dig command:
Procedure
From the Appliance Diagnostics menu, enter 3. A prompt for dig command arguments appears.
This command runs dig with he arguments you provide. Enter dig arguments:
Enter one of these arguments:
- -x followed by an IP address (for example, -x 172.20.33.40)
- An FQDN
In response to the confirming prompt, enter y to confirm your entry or n to try again. When you enter y, the results of the dig command are displayed on the screen.
After viewing the display, press Enter to return to the Appliance Diagnostics menu.
The following is a sample response to the dig command with the argument hcp.example.com:;<<>> DiG 9.7.4-P1-RedHat-9.7.4-2.P1.fc14 <<>> hcp.example.com ;; global options: +cmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NXDOMAIN, id: 49028 ;; flags: qr aa rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 1, ADDITIONAL: 0 ;; QUESTION SECTION: ; hcp.example.com IN A ;; AUTHORITY SECTION hcp.example.com. 825 IN NS hcp.example.com hcp.example.com. 825 IN NS hcp.example.com hcp.example.com. 825 IN NS hcp.example.com hcp.example.com. 825 IN NS hcp.example.com ;; ADDITIONAL SECTION: d9-014.hcp.example.com. 825 IN A 172.20.37.14 d9-015.hcp.example.com. 825 IN A 172.20.37.15 d9-016.hcp.example.com. 825 IN A 172.20.37.16 d9-017.hcp.example.com. 825 IN A 172.20.37.17
Route
The route command displays the routing table for the node you’re on. This table shows the subnets that the node belongs to. The information for each subnet includes the subnet mask and the network interface. For the [hcp_system] network (shown as default) and the network used for replication, the subnet information also includes the network gateway IP address.
This diagnostic is useful, for example, for exposing network connectivity problems such as missing interfaces. If you’ve created multiple networks in HCP, the display from the route command should show the subnet and interface for each network for which the node has an IP address.
To execute the route command:
Procedure
From the Appliance Diagnostics menu, enter 4.
The results of the route command are displayed on the screen.After viewing the display, press Enter to return to the Appliance Diagnostics menu.
The following is a sample response to the route command:Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface default 172.20.43.254 0.0.0.0 UG 0 0 0 eth0 132.17.10.0 * 255.255.255.0 U 0 0 0 eth0.0010 Link-local * 255.255.0.0 U 1002 0 0 eth0 Link-local * 255.255.0.0 U 1003 0 0 eth1 172.20.42.0 * 255.255.254.0 U 0 0 0 eth0 172.21.52.0 * 255.255.255.0 U 0 0 0 eth1
showmount
The showmount command displays the NFS exports table for a device that you identify by IP address or FQDN or for the node you’re on. You might run this diagnostic, for example, to check whether shares associated with external storage volumes are properly exported on the external storage device.
Each line in the exports table shows a path that can be mounted by other devices. For an HCP node, the exports table includes the paths to namespaces for which the NFS protocol is enabled.
To execute the showmount command:
Procedure
From the Appliance Diagnostics menu, enter 5.
A prompt for an IP address or FQDN appears.This command runs ‘showmount -e’ for the node you specify by address or FQDN. Please enter the address or FQDN:
Enter the target device IP address or FQDN. For the FQDN of the node you’re on, use localhost.
The results of the showmount command are displayed on the screen.
After viewing the display, press Enter to return to the Appliance Diagnostics menu.
The following is a sample response to the showmount command:
Export list for 172.20.43.150: /hcp_shares/share2 172.20.43.151 /hcp_shares/share1 172.20.43.151
fchbainfo
For SAIN systems only, the fchbainfo command displays information about the logical volumes that are mapped the node you’re on for all the Fibre Channel SAN arrays to which the node is connected. You might run this command, for example, to diagnose issues with zero-copy failover.
The information displayed by the fchbainfo command is broken out by HBA port. For each logical volume that maps to the port, the display includes the volume number and size.
To execute the fchbainfo command:
Procedure
From the Appliance Diagnostics menu, enter 6.
The results of the fchbainfo command are displayed on the screen.After viewing the display, press Enter to return to the Appliance Diagnostics menu.
The following is a sample response to the fchbainfo command:
General Fibre Channel Host Adapter Information: 04:00.0 Fibre Channel: Emulex Corporation Zephyr-X LightPulse Fibre Channel Host Adapter (rev 02) 04:00.1 Fibre Channel: Emulex Corporation Zephyr-X LightPulse Fibre Channel Host Adapter (rev 02) Host Adapter Information for ID: 0 Adapter Type: Emulex LPe11002-M4-H FV2.82A4 DV8.3.25 HBA Port WWN: 0x10000000c973d7a2 Maps to Remote Port: 0x200b00051e7f1649 Maps to Remote Port: 0x21fc00051e7f1649 Maps to Remote Port: 0x50060e8010252593 SCSI Device:0:0:0:0 Block Device:sda Vendor:HITACHI Model:DF600F Size:31GB Partition: sda2 Size: 1MB Partition: sda3 Size: 3GB Partition: sda4 Size: 1KB Partition: sda5 Size: 2GB Partition: sda6 Size: 2GB Partition: sda7 Size: 11GB Partition: sda8 Size: 11GB SCSI Device:0:0:0:1 Block Device:sdb Vendor:HITACHI Model:DF600F Size:104GB SCSI Device:0:0:0:129 Block Device:sdh Vendor:HITACHI Model:DF600F Size:104GB SCSI Device:0:0:0:130 Block Device:sdi Vendor:HITACHI Model:DF600F Size:104GB SCSI Device:0:0:0:192 Block Device:sdj Vendor:HITACHI Model:DF600F Size:52GB SCSI Device:0:0:0:193 Block Device:sdk Vendor:HITACHI Model:DF600F Size:52GB SCSI Device:0:0:0:2 Block Device:sdc Vendor:HITACHI Model:DF600F Size:104GB SCSI Device:0:0:0:64 Block Device:sdd Vendor:HITACHI Model:DF600F Size:52GB SCSI Device:0:0:0:65 Block Device:sde Vendor:HITACHI Model:DF600F Size:52GB Host Adapter Information for ID: 3 Adapter Type: Emulex LPe11002-M4-H FV2.82A4 DV8.3.25 HBA Port WWN: 0x10000000c973d7a3 Maps to Remote Port: 0x200b00051e80a7ff . . .
ethtool
The ethtool command displays information about the NIC parameters and device drivers for both front-end interfaces. You might run this command, for example, to diagnose issues with 10 gigabyte Ethernet network configurations.
To execute the ethtool command:
Procedure
From the Appliance Diagnostics menu, enter 6.
The results of the ethtool command are displayed on the screen.After viewing the display, press Enter to return to the Appliance Diagnostics menu.