Problem
The features of the bptestnetconn network diagnostic and debugging tool have evolved continuously since its introduction in NetBackup 6.5.6 (with EEBs for 6.5.4 and 6.5.5). The information below is intended to make the tool easier to use by detailing
- what the syntax requirements are for each argument,
- which arguments are valid at each release level,
- what output is produced at each release level,
- what output is produced by each argument at the various verbosity levels,
- what the most useful argument combinations are, and
- how to enable debugging for troubleshooting.
Solution
Usage : (as of NetBackup 8.1)
usage 1: bptestnetconn [-v[verbosity]] -h | -b | -l
usage 2: bptestnetconn [-v[verbosity]] [-i] [-q] [-m] [-s][-H hostname]
usage 3: bptestnetconn [-v[verbosity]] [-a | [-f][-r]]
[-p [-R local_ip]] [-s][-H hostname]
usage 4: bptestnetconn [-v[verbosity]] -c[service/obj]
[-o tv] [-t tv] [-P] [-R local_ip] [-d reps] -s | -H hostname
usage 5: bptestnetconn --afcheck
usage 6: bptestnetconn --basic_networking_check
usage 7: bptestnetconn [-v[verbosity]] -w[webappname]
[-O port] [-T timeout] [-e retrycount] -s | -H hostname
usage without arguments: -v6 --all --prefnet --servers
where:
-a or --all perform forward and reverse resolution; -f -r (7.0)
-c or --connect connect to specified or default service/object (7.0)
-f or --flkup perform forward hostname lookup (7.0)
-h or --help displays this usage syntax (7.0)
-H or --hostname target hostname or IP address (7.0)
-i or --ipservers list IP address of host or servers (7.0)
-l or --listservers list all servers in the config (7.0)
-o or --objconntimeout object connect timeout (sec) (7.0)
-r or --rlkup perform reverse ip lookup (7.0)
-s or --servers target all servers in the config (7.0)
-t or --orbconntimeout TCP/PBX connect timeout (sec) (7.0)
-v or --verbose output details setting; 1-6 (7.0)
-b or --confchecker display NB networking config w/default values (7.0.1)
-6 or --afcheck check IP_ADDRESS_FAMILY requirements (7.1)
-p or --prefnet display PREFERRED_NETWORK status (7.1)
-R or --override_ri source connection from this local IP address (7.1)
-d or --reps repetitions for various ops, such as connect (7.5)
-m or --domainname display the domain name of hostname or servers (7.5)
-q or --fqdn display the FQDN of hostname or servers (7.5)
-P or --profile display forward profile for the target service (7.6)
-n or --basic_networking_check of local configuration (7.6)
-e or --retrycount web service connection retry count (8.0)
-O or --port web service port to connect (8.0)
-T or --wsconntimeout web service connection timeout (sec) (8.0)
-w or --web report web service responsiveness (8.0)
Note: There are additional arguments for internal/installation use by NetBackup, they are not intended for external users and are not documented here.
Example CORBA and web services: (for the connect and web arguments)
(default): nbsl/HSFactory
nbpem: nbpem/PEM
nbpem/PEMJL
nbjm: nbjm/JM
nbjm/JOBINST
nbstserv: nbstserv/StorageService.StorageServiceMgr
nbstserv/StorageService.ExpirationMgr
nbstserv/ImpMgr_SVC
nbrb: nbrb/ResBroker used by nbjm
nbrb/ReqMgmt
nbemm: EMM/EMMServer
nbevtmgr: nbevtmgr/Event.EventMgr
nbevtmgr/VRTS_NBU_PEM_Channel used by nbpem
nbsl: nbsl/HSFactory
nbsvcmon: nbsvcmon/NBSvcMon
nbaudit: nbaudit/nbaudit
nbvault: nbvault/VaultService
nbvault/VaultRuntimeServant
nbrmms: nbrmms/DiskPollingService.DPS
nbrmms/STSEventService
FSM: NBDSMFSM/FSM.FSMServiceMgr
NBFSMCLIENT/FSM.ClientClusterMgr
NBDSMFSM/DiskService.DiskServiceMgr used by nbrb
nbftsrvr: nbFTServer/FATServer.FATServerMgr
nbftclnt: nbFTClient/FATClient.FATClientMgr
7.6 master: nbars/DARSManagement
dedupe clnt: ProxyV9SVC/Proxy.EventChannelManager
ProxyV9SVC/Proxy.StorageServerManager
7.6 client: DiscoveryService/DiscoveryService
8.0 client: nbwmc/netbackup
nbwmc/security
Command Line Guidelines
Several of the command line arguments require a value that can be separated with white space, e.g.
$ bptestnetconn -a -H server1
But the verbose (-v), connect (-c), and web (-w) arguments take an optional value and must be followed by whitespace to signify a NULL value or the value without any leading whitespace. In these examples, the middle two commands are equivalent once the optional whitespace is removed. The bottom command is invalid because it contains whitespace that is not permitted.
good$ bptestnetconn -v -c -t 10 -o 5 -H mymm
good$ bptestnetconn -v6 -cnbsl/HSFactory -t 10 -o 5 -H mymm
good$ bptestnetconn -v6 -cnbsl/HSFactory -t10 -o5 -Hmymm
bad$ bptestnetconn -v 6 -c nbsl/HSFactory -t10 -o5 -Hmymm
Further, the verbose (-v) argument accepts the optional values 1...6 which may be confused with the afcheck (-6) argument if combined, so the latter must be specified alone, without any other arguments.
$ bptestnetconn -6
The basic networking check (-n) argument must similarly be specified without other arguments.
$ bptestnetconn -n
When performing name resolution or connectivity checks, the target can be a host (-H) and/or the entire servers list (-s).
The target host can be
- a hostname
- an IPv4 address
- an IPv6 address
- the name of a file containing a list of such names and/or addresses; one per line
- '-' for stdin with the hostnames and addresses each on a separate line
Overall it is best to specify each argument separately and to use whitespace where permitted and add the servers and/or host at the end of the line, e.g.
$ bptestnetconn -v6 -a -p -cnbsl/HSFactory -t 10 -o 5 -P -s -H mymm
Examples
The value for the verbose (-v) argument can cause varying amounts of output from the other arguments and those outputs can be redundant and overwhelming. The examples below demonstrate optimal verbose settings for various critical tests using NB 7.6 or 8.1, depending on feature being demonstrated. See the notes in the usage section above and the output notes below if utilizing an older version of NetBackup.
Example #1: Checking Name Resolution
To check for forward vs. reverse resolution mismatches use the all (-a) or forward (-f) and reverse (-r) arguments with a host and/or servers list. The output will also identify which hostnames resolve to local interface and if those interfaces are associated with a cluster or required interface setting.
Beginning with NB 7.0.1, slow name resolution will be flagged. The default threshold is 5 seconds and can be adjusted using VNET_OPTIONS field 8 (slow DNS threshold).
$ bpgetconfig VNET_OPTIONS
VNET_OPTIONS = 120 3600 200 40 3 1 30 5 1793 32 0 0
$ bptestnetconn -v -b | grep 'slow DNS'
VNET_OPTIONS (slow DNS threshold) = 5 [5]
Note: Beginning with NB 7.1.0.4, link-local IP addresses will be noted.
Note: Beginning with NB 7.6, stale host cache entries will be displayed.
Note: It is not a mismatch if the reverse lookup of a shortname is a fully qualified domain name.
$ bptestnetconn -a -s -H myhost
------------------------------------------------------------------------
FL: myhost -> 1.1.1.1 : 1 ms
FL: mymaster-bk -> 192.168.0.11 : 0 ms [local][cluster/ri]
FL: mymaster-bk.fqdn -> **LKUP FAIL** : 2 ms
FL: mymm-bk -> 10.82.105.15 : 2 ms
FL: mymm-private -> 169.254.1.1 Warning: LINK-LOCAL (APIPA) [NOT USED] : 0 ms
FL: mymm2-bk -> 10.81.40.61 : 4 ms
FL: [STALE_CACHE] mymm2-bk -> 192.168.0.61 : 3 ms
------------------------------------------------------------------------
RL: 1.1.1.1 -> notmyhost : 0 ms MISMATCH
RL: 192.168.0.11 -> mymaster-bk : 0 ms
RL: 127.0.0.1 -> localhost : 0 ms
RL: 10.82.105.15 -> **LKUP FAIL** : 4 ms
RL: 169.254.1.1 -> mymm-private : 0 ms
RL: 10.81.40.61 -> mymm2-bk.fqdn : 8003 ms SLOW (> 5 sec)
---------------------------------------------------------------------
Slow (>5 sec) or/and failed forward lookups:
mymaster-bk.fqdn : 0 sec [FAILED]
--------------------------------------------------------------------
Slow (>5 sec) or/and failed/mismatched reverse lookups:
myhost : 0 sec [MISMATCH] -> notmyhost
mymm2-bk : 8 sec
--------------------------------------------------------------------
If [STALE_CACHE] resolution differs from normal resolution, consider
clearing the host cache using bpclntcmd -clear_host_cache.
Stale cache entries may continue to cause connectivity and permission failures
after the original cause has been fixed.
------------------------------------------------------------------------
Total elapsed time: 0 sec
Note: The output is not as well formatted in NB 7.0GA.
$ bptestnetconn -a -H myhost -s
Forward lookup times:
host: myhost[10.82.104.249]: 1 ms
host: mymaster-bk.min.veritas.com[10.82.105.11]: 0 ms
host: mymm[10.81.40.61]: 0 ms
Reverse lookup times:
host: myhost.min.veritas.com[10.82.104.249]: 1 ms
host: mymaster-bk.min.veritas.com[10.82.105.11]: 0 ms
host: mymm.min.veritas.com[10.81.40.61]: 0 ms
Example #2: Testing Connectivity
To check the CORBA connectivity between the local host and the target host and/or the servers list, use the connect (-c) argument. Be aware that the default service and object has evolved considerably since introduction.
nbrmms/STSEventService (6.5.6)
<none, user must supply> (7.0)
nbrmms/STSEventService (7.0.1)
nbsl/HSFactory (7.1+)
It is recommended to change the TCP/orb timeout (-t) to a value that is longer than the object timeout (-o). This makes it possible to more easily distinguish a connectivity issue ( TCP, firewall, or listener ) from a problem with the remote service. The default values of 30 seconds each are indistinguishable, and because NetBackup makes multiple connection attempts it can take a long time to fail repeatedly when checking multiple servers.
Using '-t 10 -o 5' allows a failure to be detected more quickly. An error after 9-10 seconds indicates a TCP connectivity problem to the target host, while an error after 4-5 seconds indicates either a name resolution problem before the TCP connection or a PBX-to-service problem after the TCP connection is established. If the remote host is on the far side of the world, '-t 20 -o 10' allows for sufficient delay.
Beginning with NB 7.1, add the override_ri (-R local_ip) argument to request 'local_ip' as the source interface. This will cause any Preferred Network, Required Interface, Required Network, or Cluster Name settings to be ignored during the test.
Beginning with NB 7.5, add the reps (-d count) argument to repeat the connectivity test multiple times.
Beginning with NB 7.6, add the profile (-P) argument to display the forwarding profile returned by the remote PBX to be used when connecting to the service.
Add the verbose (-v) argument to check connectivity to vnetd [and prior to NB 8.1 also to bpcd] daemon ports. Those tests results are reported independent of the CORBA connection results. NB 8.1+ no longer connects directly to the bpcd daemon port, so that test is no longer made or displayed. This sample output is from NB 8.1 which shows the connection address:port pairs and hostID certificates if routed through the vnetd secure comm proxy.
$ bptestnetconn -v -c -t 10 -o 5 -P -s -H newFTclient-bk
adding hostname = newFTclient-bk
...snipped target host list...
------------------------------------------------------------------------
Connecting to 'nbsl/HSFactory'
CN: newFTclient-bk : 11 ms [SUCCESS] PBX: Yes VNETD: No
Addresses returned on attempted connection to newFTclient-bk/nbsl:
: ---Initial Request---
: 10.81.40.127
: ---Forward Profile---
: newFTclient
: 191.168.0.12
: newFTclient-bk
: fc44:53f9:cb30:50:20c:29ff:fe8d:5fea
: ---------------------
Connection is not routed via vnet proxy
CN: mymaster-bk : 11 ms [SUCCESS] PBX: Yes VNETD: Yes
Addresses returned on attempted connection to mymaster-bk/nbsl:
: ---Initial Request---
: 10.81.40.61
: ---Forward Profile---
: server1
: 10.81.40.61
: mymaster
: ---------------------
127.0.0.1:38437 -> 127.0.0.1:61288 PROXY 192.168.0.13:63982 -> 192.168.0.12:1556
Certificate Information:
local_cert_info: {
"certificate_subject_common_name": "ecddfaba-b477-4982-bd49-a4c7e94f4152",
"certificate_issuer_name": "/CN=broker/OU=root@mymaster.com/O=vx",
"netbackup_certificate_type": 1
}
peer_cert_info: {
"certificate_subject_name": "/CN=933e20db-033a-487c-9705-f8092d883156/OU=NBU_HOSTS/O=vx",
"certificate_subject_common_name": "933e20db-033a-487c-9705-f8092d883156",
"certificate_issuer_name": "/CN=broker/OU=root@mymaster.com/O=vx",
"certificate_issuer_org_unit_name": "root@mymaster.com",
"certificate_serial_number_name": "71931A7000000008",
"master_server": "mymaster.com"
}
------------------------------------------------------------------------
Total elapsed time: 5 sec
CORBA connections and communications can fail for many reasons. The table below shows the results at some of the known failure points in the connection and communication process. Because each CORBA status may have more than one failure cause it is beneficial to set the TCP/orb connection timeout (-t) to be longer than the object connection timeout (-o). The combination of status and time before the status is reported narrows the number of possible root causes.
CORBA connection results for: bptestnetconn -v -t 10 -o 5 -cservice/object -H hostname
Elapse CORBA results Process results through PBX Cause
------- ------------------ ----------------------------- --------------------------------
4 sec [TRANSIENT] PBX: No VNETD: No BPCD: No Hostname is not resolvable.
10 sec [TRANSIENT] PBX: No VNETD: No BPCD: No No route to/from port on hostname.
10 sec [TIMEOUT] PBX: No VNETD: No BPCD: No TCP SYN dropped by firewall.
4 sec [TRANSIENT] PBX: No VNETD: Yes BPCD: Yes PBX is not listening.
10 sec [TRANSIENT] PBX: Yes VNETD: Yes BPCD: Yes PBX is hung or busy.
4 sec [NO_PERMISSION] PBX: Yes VNETD: Yes BPCD: Yes Caller is not in SERVER list.
4 sec [TRANSIENT] PBX: Yes VNETD: Yes BPCD: Yes Service is not registered with PBX.
5 sec [COMM_FAILURE] PBX: Yes VNETD: Yes BPCD: Yes -cbpcd is not a CORBA service.
4 sec [COMM_FAILURE] PBX: Yes VNETD: Yes BPCD: Yes -cvnetd is not a CORBA service.
4 sec [OBJECT_NOT_EXIST] PBX: Yes VNETD: Yes BPCD: Yes Object is unknown to service .
X sec [COMM_FAILURE] PBX: Yes VNETD: Yes BPCD: Yes TCP connection dropped or service
exited/crashed after X seconds.
17 ms [SUCCESS] PBX: Yes VNETD: No BPCD: No vnetd & bpcd not registered w/PBX.
12 ms [SUCCESS] PBX: Yes VNETD: Yes BPCD: Yes Service/vnetd/bpcd reached via PBX.
ZZZ ms [SUCCESS] PBX: Yes VNETD: Yes BPCD: Yes TCP [re]transmission or service
was delayed for ZZZ ms.
This example captured an intermittent TCP connection problem on the 4th of 5 attempts.
$ bptestnetconn -c -t 10 -o 5 -d 5 -H newFTclient-bk
------------------------------------------------------------------------
CN: newFTclient-bk : 50 ms [SUCCESS]
CN: newFTclient-bk : 5 ms [SUCCESS]
CN: newFTclient-bk : 6 ms [SUCCESS]
CN: newFTclient-bk : 10 sec [TRANSIENT]
CN: newFTclient-bk : 7 ms [SUCCESS]
------------------------------------------------------------------------
This example shows a hung connection attempt because the service (nbsl) is hung or too busy to process the request. Some CORBA using processes set a maximum timeout to prevent indefinite hangs if TCP or the remote service are unresponsive; typically CLIENT_CONNECT_TIMEOUT or SERVER_CONNECT_TIMEOUT, but other values are possible.
$ bptestnetconn -cnbsl/HSFactory -t 10 -o 5 -H myMM
adding hostname = myMM
------------------------------------------------------------------------
CN: myMM ...output is hung...
See the Related Articles for debug log examples when a connection fails.
Note: The output is not as well formatted in NB 7.0GA.
$ bptestnetconn -v -cEMM/EMMServer -s -H master2
ping_corba host master2 object EMM/EMMServer elapsed time: 315 msec
ping_corba host mymaster object EMM/EMMServer elapsed time: 18 msec
ping_corba host master2mm object EMM/EMMServer
system exception, ID 'IDL:omg.org/CORBA/NO_PERMISSION:1.0'
TAO exception, minor code = 0 (unknown location; unspecified errno), completed = NO
-- No Permission! Detected after 8 seconds
Example #3: Checking the Source Binding Request and IPv4 vs IPv6 status
Note: This is only applicable to NB 7.1 and newer.
Use the prefnet (-p) argument to display the Preferred Network source binding configurations for outbound connections to a host and/or servers.
Add the -v2 argument to display the local interface list or the -v6 argument to also display the Preferred Network settings ordered by subnet and position in the configuration.
Add the override_ri (-R local_ip) argument to request 'local_ip' as the source interface. This will ignore the Preferred Network and related settings.
In the output below notice that NetBackup (IP_ADDRESS_FAMILY) is not enabled for IPv6 (AF_INET), that host1 is prohibited (by the ONLY), that one local interface is identified as being in the server list and also configured as a required or cluster interface, and that each of the configured servers is affected by a Preferred Network MATCH setting (none are ANY).
$ bptestnetconn -v6 -p -s -H host1
...snip...
------------------------------------------------------------------------
NBU IP_ADDRESS_FAMILY configured to use Remote Addresses: IPv4(yes) IPv6(no)
FL: host1 -> 10.81.40.61 : 0 ms FAST (< 5 sec) TGT PROHIBITED
FL: server2 -> 10.81.40.127 : 0 ms FAST (< 5 sec) [local] [cluster/ri] SRC: ANY
FL: server3 -> 10.82.105.15 : 1 ms FAST (< 5 sec) SRC: 192.168.252.12
FL: host4 -> 10.82.105.14 : 1 ms FAST (< 5 sec) SRC: 192.168.1.12
FL: host55 -> 10.82.104.249 : 1 ms FAST (< 5 sec) SRC: 192.168.0.192
------------------------------------------------------------------------
...snip...
Using configured hostname server2
Displaying all locally configured IP addresses
and if these addresses are found in the DNS
Address Local DNS Entry
------- ----- ----------
fc44:53f9:cb30:50:20c:29ff:fe8d:5fea Yes No - private/autoconfigured
192.168.252.12 Yes No - private/autoconfigured
192.168.1.12 Yes No - private/autoconfigured
192.168.0.192 Yes No - private/autoconfigured
192.168.0.12 Yes No - private/autoconfigured
10.81.40.127 Yes Yes
No PREFERRED_NETWORK directives.
...or...
PREFERRED_NETWORK rules are applied in this order:
[0] PREFERRED_NETWORK = 10.82.105.15 MATCH 192.168.252.12
[1] PREFERRED_NETWORK = 10.82.105.14 MATCH 192.168.1.12
[2] PREFERRED_NETWORK = 10.82.104.0/24 MATCH 192.168.0.0/24
[3] PREFERRED_NETWORK = 10.82.0.0/16 ONLY
[4] PREFERRED_NETWORK = <PSEUDO PREFERRED NETWORK> MATCH server2
Note: [3] is an automatic translation of 'REQUIRED_NETWORK = 10.82.0.0/16 '.
Note: [4] is an automatic translation of ' REQUIRED_INTERFACE = server2 '.
See the Related Articles for additional information about Preferred Networks.
Example #4: Checking name resolution and connectivity all-in-one
Combine all of the arguments above to display the name resolution, source binding configuration, and connection results to all the defined servers. Include the forwarding profiles if desired.
$ bptestnetconn -v6 -a -p -c -t 10 -o 5 [-P] -s
...output snipped for brevity...
Example #5: Display the NetBackup network related configuration settings and defaults
Note: This is only applicable to NB 7.0.1 and newer.
Some of these configuration settings did not exist in older versions of NetBackup. See the NetBackup Administrator Guides, Volumes I and II, for details about each option.
There can be many connect option settings for many servers and clients. If noted in the output, review the configuration, including the client database, manually to see them all.
$ bptestnetconn -v -b
CLIENT_PORT_WINDOW (min) = 0 [0]
CLIENT_PORT_WINDOW (max) = 0 [0]
CLIENT_CONNECT_TIMEOUT = 300 [300]
CLIENT_READ_TIMEOUT = 300 [300]
SERVER_CONNECT_TIMEOUT = 30 [30]
DEFAULT_CONNECT_OPTIONS (bpcd port) = Reserved [Reserved]
DEFAULT_CONNECT_OPTIONS (callback port) = vnetd [vnetd]
DEFAULT_CONNECT_OPTIONS (daemon port) = Automatic [Automatic]
CONNECT_OPTIONS = [configured]
VNET_OPTIONS (timeout) = 120 [120]
VNET_OPTIONS (cache time) = 3600 [3600]
VNET_OPTIONS (successful cache size) = 200 [200]
VNET_OPTIONS (failed cache size) = 40 [40]
VNET_OPTIONS (TRY_AGAIN retries) = 3 [3]
VNET_OPTIONS (TRY_AGAIN delay) = 1 [1]
VNET_OPTIONS (TRY_AGAIN timeout) = 30 [30]
VNET_OPTIONS (slow DNS threshold) = 5 [5]
VNET_OPTIONS (endpoint selector) = New [New]
VNET_OPTIONS (cache checksum bits) = 32 [32]
VNET_OPTIONS (disable file cache) = 0 [0]
VNET_OPTIONS (unused 12) = 0 [0]
MAX_CORBA_PAYLOAD_SIZE = 262144 [262144]
OPEN_NBSL_NBNOS = NO [NO]
REQUIRED_INTERFACE = server2 [NULL]
ANY_CLUSTER_INTERFACE = NO [NO]
REQUIRED_NETWORK = 10.82.0.0/16 [NULL]
REVERSE_NAME_LOOKUP = ALLOWED [ALLOWED]
USE_VXSS = PROHIBITED [PROHIBITED]
Example #6: Display the fully qualified domain name and/or domain name
Note: This is only applicable to NB 7.5 and newer.
Use the fqdn (-q) and/or domainname (-m) arguments to display those values for this host, or another host and/or the servers list.
$ bptestnetconn -v -q -m [-s] -H server2
adding hostname = server2
FQDN: server2.rmnus.sen.symantec.com ShortName: server2 DomainName: rmnus.sen.symantec.com
rmnus.sen.symantec.com
Example #7: Sanity check the configured client name vs. O/S configuration
Note: This is only applicable to NB 7.1 and newer.
Use the afcheck (-6) argument to compare the configured NetBackup IP_ADDRESS_FAMILY, AF_INET (IPv4 only) or AF_UNSPEC (IPv4 or IPv6), to the configured client name and associated IP addresses, and to ensure at least one IP addresses is assigned to an active local interface.
$ bptestnetconn -6
bp.conf: IP_ADDRESS_FAMILY = AF_INET: default value, no update needed
...or... INFO: Host name <name> does not resolve to any IPv4 addresses.
DNS Lookup failed for host <name> error: errno_str!
INFO: Host/Client name <name> does not resolve to any IPv4 addresses.
...or (clarification beginning in NB 7.6)... Warning: Name resolution of local host myhost found no usable local IP addresses.
Possible ways to resolve this:
1. Ensure that the host can resolve the hostname noted above to an IP address assigned to one of the local network interfaces;
bpclntcmd -hn myhost
2. Correct either the hostname, network interface configuration, local name resolution configuration, or the DNS configuration as appropriate.
3. If the network interface associated with the local host name is disabled, it should be enabled before NetBackup is started.
4. If the hostname that NetBackup will use is a link-local address (169.254.0.0/16 or fe80::/64) and all the other NetBackup hosts are on the same network, then add this permission to the configuration;
echo "ALLOW_LINK_LOCAL_APIPA_ADDRS = TRUE" | nbsetconfig
Example #8: Basic network checking
Note: This is only applicable to NB 7.6 and newer.
Perform a superset of the afcheck, plus check the resolution of the hostname.
$ bptestnetconn --basic_networking_check
<no output if all checks are successful>
...or... INFO: Host name <hostname> does not resolve to any IPv4 addresses.
DNS Lookup failed for host <name> error: errno_str!
INFO: Host/Client name <name> does not resolve to any IPv4 addresses.
The host name: <gethostname> does not resolve to a local IP address.
Example #9: Check connectivity to web services on the master server
Note: This is only applicable to NB 8.0 and newer.
Use the web (-w) argument to check connectivity to web service and specific webapps on the master server. If using the servers (-s) option, a failure is expecting if the target host is not a master server and there may be considerable delays attempting to connect to web services on those hosts. Use the web service connection timeout (-T) to shorten the timeout from CLIENT_CONNECT_TIMEOUT to 10 seconds when diagnosing connectivity vs responsiveness delays. Similarily it is often useful to shorten the web service retry count (-e) from 5 to 0 or 1. The verbose (-v) option will display the names of the target hosts and target service to be checked, prior to making the connection attempt.
$ bptestnetconn -v -w -T 10 -e 1 -s
SERVER = mymaster.fqdn
SERVER = mymaster
SERVER = mymedia
SERVER = not-resolvable
SERVER = not-routable
Checking responsiveness of 'nbwmc/netbackup' webapp...
nbwmc/netbackup web service test for host: mymaster.fqdn : 75 ms [SUCCESS]
nbwmc/netbackup web service test for host: mymaster : 71 ms [SUCCESS]
nbwmc/netbackup web service test for host: mymedia : 80234 ms [FAIL]
nbwmc/netbackup web service test for host: not-resolvable : 20024 ms [FAIL]
nbwmc/netbackup web service test for host: not-routable : 20018 ms [FAIL]
$ bptestnetconn -v -wnbwmc/security -T 10 -e 0 -H mymaster
adding hostname = mymaster
Checking responsiveness of 'nbwmc/security' webapp...
nbwmc/security web service test for host: mymaster : 97 ms [SUCCESS]
Troubleshooting
Prior to NetBackup 7.1, the bptestnetconn program logs to OID 112. In NetBackup 7.1 the OID was changed to 396 (nbutils). This will show the application debug information, but not the debug information for the underlying connection libraries.
To see the connection details in the VxUL debug logs for the various processes (including FileID 112 or 396), also enable OID 137 (libraries) and OID 156 (ACE). These must be adjusted after adjusting OID 396. Be sure to note the previously existing log levels before making changes.
$ grep 'Level=' /usr/openv/netbackup/nblog.conf
Default.DiagnosticLevel=6
Default.DebugLevel=1
$ vxlogcfg -p nb -o 396 -a -s DebugLevel=6 -s DiagnosticLevel=6
$ vxlogcfg -p nb -o 137 -a -s DebugLevel=5 -s DiagnosticLevel=6
$ vxlogcfg -p nb -o 156 -a -s DebugLevel=4 -s DiagnosticLevel=6
When troubleshooting is complete, remember to either remove the OID specific logging levels or set them back to the previous values.
$ vxlogcfg -p nb -o 396 -r -s DebugLevel -s DiagnosticLevel
$ vxlogcfg -p nb -o 137 -r -s DebugLevel -s DiagnosticLevel
$ vxlogcfg -p nb -o 156 -r -s DebugLevel -s DiagnosticLevel
Applies To
NetBackup 7.x , 8.x , 9.x , 10.xWas this content helpful?
Translated Content
Please note that this document is a translation from English, and may have been machine-translated. It is possible that updates have been made to the original version after this document was translated and published. Veritas does not guarantee the accuracy regarding the completeness of the translation. You may also refer to the English Version of this knowledge base article for up-to-date information.