|
Softpanorama |
May the source be with you, but remember the KISS principle ;-)
|
Tivoli gateway can hang (pretty typical if it works with Tivoli Gateway Proxy) and that critically affect the TMR functionality.
An unreliable gateway with 1,000 endpoints logged into it affects the ability of the administrator to effectively manage those 1,000 machines. In an environment of 10,000 endpoints, that represents a 10% failure rate. The same is true for endpoint manager. An unreliable endpoint manager has an even more crippling effect, because management of the aforementioned 10,000 endpoints becomes frustrating if the endpoint manager process (ep_mgr) dies unexpectedly during critical operations.
Gateway has logs and it should be analysed first
Every gateway method is initially spawned via an oserv thread. The gateway method request is handled by a gateway RPC thread that is spawned by the aforementioned oserv thread. When the gateway RPC thread is spawned, gateway_method_in() is called. This routine/function determines what type of method call is being made.
There is a max_gateway_rpc_threads parameter, which allows flexibility when managing gateway processes.
Also, each gateway RPC thread uses an oserv thread (and also an operating system thread). A good rule of thumb when using this parameter is:
max_gateway_rpc_threads <= 60% of rpc_max_threads (of oserv)
If the TMR Server is also acting as a gateway, then consideration must be given to the number of oserv threads being used by the endpoint manager and the gateway respectively. In this case, the recommended number of treads are:
(max_epmgr_rpc_threads + max_gateway_rpc_threads) <= 60% of rpc_max_threads (of oserv)
There are two types of gateway threads that execute gateway methods:
If gateway_method_in() determines that the method call is a jobq method, it calls new_job(), which places the method on a job queue and closes the gateway RPC thread and, in turn, the oserv thread is also closed. Therefore, the active RPC gateway thread count is reduced by one, and one of those 250 invaluable gateway RPC threads is released. This process is very fast, which means that jobq methods have a minimal impact on the RPC thread limit.
The method request (the job) on the job queue is processed by pthreads, of which there are max_concurrent_jobs in number. The pthreads work on any jobs placed on the queue. The default number of pthreads available to service the job queue is 200, this being the default value of max_concurrent_jobs. There is no limit on the number of pthreads that can be used, except that eventually we could run out of memory if too many are spawned.
A good rule of thumb is:
threads = max_concurrent_jobs = number of endpoints on hosting gateway >= 200 <= 1000 (to the next or nearest 100)
Therefore, the recommended value for max_concurrent_jobs is:
| If there are less than 200 endpoints on a gateway, max_concurrent_jobs = 200 (default) |
| If there are 570 endpoints on a gateway, max_concurrent_jobs = 600 |
| If there are 1,200 endpoints on a gateway, max_concurrent_jobs = 1000 |
Monitoring memory usage of the gateway process is advisable after changing this value. As pthreads are spawned, the gateway process will grow while it does the work of completing the jobs.
Monitoring the number of endpoint manager/TMR methods being spawned at the TMR Server from gateway method calls is also important. Upcalls are placed on the jobq and are handled by pthreads. As more pthreads are made available, more upcalls will be concurrently processed. If each upcall results in a method cascade that involves a process on the TMR Server, then it is possible that the TMR Server may get more method requests than it was tuned to handle concurrently. You may then have very efficient gateways, but an overloaded TMR Server.
If gateway_method_in() determines that the method call is a non-jobq method, it calls tmf_dispatch(). This routine handles the execution of the method while keeping the oserv thread open for the duration of its execution. During that period of time, one of the available 250 gateway RPC threads is being used, as well as one oserv thread. Therefore, these methods can be expensive, and become more so if they run for a long time.
Figure 13-4 may help to further clarify the gateway's method handling routines.
Now that we have described the two types of gateway threads, we will categorize the gateway methods so we can understand the impact they have on the gateway's performance and stability.
The methods that are queued are:
Note that if the gateway has an insufficient number of pthreads (max_concurrent_jobs) the following operations are affected:
Methods that are not queued are:
| get_data_mp This method extracts the details that are stored in the database about this gateway. The data returned includes: binary location, lcf_bundle directory location, protocol, port, hostname, and session timeout. |
| get_data Returns the same data as get_data_mp, but without the default values. |
| get_net_aliases This method is used by the endpoint manager to get the interface data from the gateway. The gateway would return all the interfaces that the oserv has bound to, and therefore, all the interfaces that the gateway has bound to in that machine. |
| new_endpoint This method adds new endpoint data to a gateway's database, but the data is not encrypted. This is for backward compatibility with 3.6 gateways because they did not implement the encrypted method. |
| new_endpoint_encrypted This method adds new endpoint data to a gateway's database. This would happen during an initial isolation (this gateway would be receiving the endpoint) and orphaned logins. Used by epmgr to inform a gateway that this endpoint is assigned to the gateway. |
| delete_endpoints This method is spawned by wdelep on the gateway to delete an endpoint from the gateway's database. It is also spawned during a migration of an endpoint from a gateway, so it is the method used to remove an endpoint account/record from a gateway database |
| login_policy This is the method that runs the login policy script for each endpoint during a normal login. This method is not run using a tmf_dispatch thread, but is instead run as a command, so it is run using oserv-h (oserv helper). |
| update_endpoint_cache This method is used to update a gateway's database with new information about an endpoint that it is already managing, or is already in its database. |
Summary of methods not queuedNote that these methods are bound by the gateway's RPC thread limit of 250, except for the login_policy.
|
Understanding gateway resource usageTo help with recognition of the workload that a gateway is handling, logstatus_interval method is used. This method determines the time interval between writing logstatus information in the gatelog. This is irrespective of the debug level of the gatelog. The logstatus information is composed of:
These logstatus values illustrate the workload and the workflow through the gateway. Figure 13-5 illustrates the workflow.
Figure 13-5 Logstatus information workflow |
wgateway gateway_name logstatus
Status Data ------
Note: If Jobq_threads in queue parameter is above 100 that usually means that gateway is experiencing problem with "hang" threads and needs to the recycled.
Jobq_threads in queue : 0
Jobq_threads running : 1
Reader_threads in queue: 0
GW methods running : 1
Number of Endpoints : 250
Endpoints connected : 245
wgateway nti2171-gw describe
# wgateway nti2171-gw describe Gateway Object : 1034227612.1.639#TMF_Gateway::Gateway# Hostname : tmrserver debug_level : 0 session_timeout : 300 Protocols : TCPIP TCPIP Port : 9494 Start Time : 2007/11/07 15:09:25 +05 no_endpoints : 250 log_dir : /opt/TMF/db/tmrserver.db log_size : 1024000 rpc_maxthreads : 500 max_concurrent_jobs : 400 max_concurrent_logins : 100 method_trace_time : 3600 logstatus_interval : 3600 login_interval : 270 epcheck_interval : 3600 epcheck_atboot : Enabled httpd : Disabled mcache_bwcontrol : Disabled wol_broadcast : Enabled, Port=9 wol_direct : Disabled send_userlink_file : Enabled crypt_mode : NONE ifs_ignore : NILNote:
rpc_maxthreads : 500 # Usually
you should set the rpc_max_threads
to as big as OS will allow (default is 250)
§Viewing this parameter§odadmin get_rpc_max_threads ####§Setting this parameter§odadmin set_rpc_max_threads ####
wepstatus -a -- list health conditions of both gateways and endpoints
How many gateways are installed
Gateway parameters
wgateway gateway_name describe
Starting and stopping gateway
wgateway <gateway_name> stop && wgateway <gateway_name> start
Re: Gateway down, please help, it's kindda urgent
Posted: Jan 30, 2007 09:06:50 AMin response to: shadow's post
[Nov 11, 2007] RE TMR 4.1.1 - gw_tune.pl
This script analyses gateway parameters and propose new settings. Later version is available from IBM Tivoli Group.
[Nov 11, 2007] IBM - Headers in the gw_tune.pl script What do they mean
Technote (FAQ) Problem gw_tune.pl : What does the different headers in the gw_tune.pl script mean? Cause Not listed in manuals Solution RECOMMENDED SETTINGS
===================================================================
GW_LABEL| o_rpc eps jq lq g_rpc md2 mem pktsz nt_ld
===================================================================
tiv-gwy| 250 217 200 100 250 55 2048 16 500
recommend| ** 300 -- 450 OK OK -- --
===================================================================
oserv threads: 750 epmgr threads: OK login_limit: 150
o_rpc
oserv_rpc threads (default 250)
The number of OS File Descriptors limits the number of all other RPC Threads in Tivoli. Main threads within Tivoli are the oserv_rpc threads which control the limit of the epmgr rpc_treads and gw rpc treads.
eps
number of endpoints=217
This setting is used for the consideration in the calculations of all the other settings
jq -- jobq threads
If gateway_method_in() determines that the method call is a “jobq” method, it calls new_job(), which places the method on a job queue and closes the gateway RPC thread and also the oserv thread is also closed.
This process is very fast, which means that jobq methods have a minimal impact on the RPC thread limit.
The method request (the job) on the job queue is processed by pthreads, of which there are max_concurrent_jobs in number. The pthreads work on jobs placed on the queue.
The default number of pthreads available to service the job queue is 200, the default value of max_concurrent_jobs.
If the gateway has an insufficient number of pthreads (max_concurrent_jobs) the following operations are affected:
Downcalls to endpoints.
Excessive isolation logins. Upcalls and logins from endpoints.
Delayed distributions. The rpt method waits for a free pthread to spawn.
lq -- login queue threads
g_rpc -- gw rpc threads
Because a gateway RPC Thread uses an oserv RPC thread, care must be taken to keep the ‘oserv’ functional.
A good rule of thumb to adopt is:
rpc_maxthreads (of gateway) <= 60% of rpc_max_threads (of oserv)
Prior to Framework FP2, this parameter was not available.
Recall that the oserv RPC threads is a global parameter and this in itself is governed by the number of OS File Descriptors.
md2 -- Mdist2 sessions[total of high+medium+low]
mem -- memory on gateway
net_load=KB_per_second. Specifies the maximum amount of network bandwidth (in kilobytes per second) that the repeater is allowed to use. If slow_link is set to TRUE, the network bandwidth is measured in bytes per second. This allocation is shared among all active distributions. This option is used with target_netload.
pktsz -- mdist2 packet size
[Aug 14, 2007] IBM - Gateways Explained Gateway functionality, tuning and troubleshooting
Tivoli Field Guide - The Tivoli Management Enterprise Endpoint Gateway: A Technical Look at the Internals
April 4, 2007 An unstable endpoint manager or endpoint gateway can have a catastrophic ripple effect on the reliability and effectiveness of the Tivoli Management Environment. This is primarily because of the number of management units that are affected by their instability. An unreliable gateway with 1,000 endpoints logged into it affects the ability of the administrator to effectively manage those 1,000 machines. In an environment of 10,000 endpoints, that represents a 10% failure rate, which is unacceptable...
Syntax
wgateway
wgateway gateway_name {add_gatewayproxy node_name |
get_gatewayproxies | remove_gatewayproxy node_name |
reset_gatewayproxies}
wgateway gateway_name [add_protocol protocol | rm_protocol
protocol] [set_ipx_port port] [set_nbi_port port]
[set_nbi_name name ] [set_protocols protocol_list]
[set_sna_address address] [set_sna_mode mode] [set_sna_port
port] [set_tcp_port port]
wgateway gateway_name {dbcheck | epact_dbcheck}
wgateway gateway_name describe
wgateway gateway_name get option
wgateway gateway_name logstatus
wgateway gateway_name {restart | start | stop}
wgateway gateway_name run option value
wgateway gateway_name set option value
The wgateway command starts or stops a gateway, lists or modifies the properties of a gateway, controls access to the gateway HTTP server, or defines logging characteristics. Without options, the wgateway command lists the object identifier, label, and status of all gateways in the local Tivoli region. The status value for a gateway can be one of the following:
b The gateway is booting.
d The gateway is down and not running.
D A communication error occurred. The object dispatch is
down. However, the entire computer could be down.
u The gateway is up and running.
To start, stop, or restart (stop and start) a gateway, use wgateway(1) the start,
stop and restart options, respectively.
Note: For NetWare gateways, the start and stop options cannot be used. For the start option, use the oservrun command. For the stop option, use the oservend command. Both of these commands must be run from the NetWare console.
To list all the properties of a specific gateway, use the describe option. If you need to change any of these properties, you can modify many of these properties using the set option and its suboptions.
For endpoint method logging, use the get method_trace_time option to return the value of the set resolution time and the set method_trace_time option to change the resolution time. To return the status of logging, use the logstatus option. To change the amount of time between status checks, use the set logstatus_interval option.
To ensure that access is secure to the gateway HTTP server, you can enable or disable the server by using the set httpd option. After the gateway HTTP server is enabled, anyone connecting to this service is prompted to authenticate. No one can access the gateway Web pages until the user names and passwords are created. After setting the user names and passwords, access to the gateway pages for those users is allowed. User names and passwords must be predefined for each user using the set httpd_passwd option. Each administrator with the admin authorization role can define his or her user name and password.
Note: The user name and password can be any value. They do not default to the system administrator user name or password. They do not default to the Tivoli administrator user name and password.
For NetWare gateways, you need at least one gateway proxy. The gateway proxy runs endpoint policy scripts. Any managed node can be a gateway proxy. The following options can be used to manage gateway proxies:
add_gatewayproxy
To add a managed node to the gateway proxy list.
get_gatewayproxies
To list the gateway proxies for a specific
NetWare gateway.
remove_gatewayproxy
To remove a managed node from the gateway proxy
list.
reset_gatewayproxies
To remove all entries from the gateway proxy
list.
For NetWare gateways and other gateways that support the IPX/SPX protocols, you
can add IPX/SPX support after creating the gateway. To add IPX/SPX support, use
the add_protocol IPX option and use the set protocols IPX option to set the protocols
for the gateway. If you no longer need IPX/SPX support, use the rm_protocol IPX
option. If you do not want to use the standard TCP/IP or IPX/SPX listening ports,
use the set_tcp_port to set the TCP/IP listening port or set_ipx_port option to
set the IPX and SPX listening ports.
Because each gateway maintains a database file and has a method cache, it could become unsynchronized with the one on the Tivoli server or become corrupted. To synchronize the method caches between these computers, use the dbcheck option. To check the consistency of the endpoint activity database (the epact.bdb file), use the epact_dbcheck option.
add_gatewayproxy node_name
For NetWare gateways only. Adds a managed node
as an entry to the gateway proxy list.
add_protocol protocol
Adds a supported protocol for the specified
gateway. Supported protocols are TCPIP (the
default) and IPX.
dbcheck Synchronizes the method cache on the gateway
with the method cache on the Tivoli server.
describe Lists the properties of the specified gateway.
epact_dbcheck
Checks the consistency of the endpoint activity
database (the epact.bdb file).
get option Returns the value that is set for the specified
option. You can specify the following suboptions
with the get option:
continue_eplogin_onerror
Returns a value indicating whether the
gateway allows exeptions from the
boot_method method, the login_policy
script, or both, during normal login
attempts by endpoints.
debug_level
Returns the level of message information
that is logged by the gateway. The
levels are as follows:
epcheck_atboot
Returns a value indicating whether the
gateway checks the status of all
endpoints that are assigned to it each
time the gateway starts up.
epcheck_interval
Returns the interval, in seconds, at
which the gateway checks the status of
endpoints that it communicates with.
eplogin_timeout
Returns the current value of the
endpoint login timeout.
epupgrade_by_gateway
Returns a value indicating whether the
gateway upgrades any endpoints assigned
to it during the normal login process.
If the endpoint is already at the most
current level, no upgrade is performed.
httpd Returns a value indicating whether the
gateway HTTP server is enabled.
httpd_passwd
Displays the user name and password
required to access the gateway HTTP
server to view information about
endpoints.
ifs_ignore
Displays a list of IP addresses that are
excluded from the gateway aliases list
when the gateway boots up.
ipx_port
Returns the port on which the gateway
listens for IPX packets.
log_dir Returns the directory on the local
system for the gatelog and all
gatelog.timestamp log files.
log_size
Returns the reference size of the
gatelog file, in bytes.
logstatus_interval
Returns the amount of time in seconds
between status checks.
max_concurrent_jobs
Returns the maximum number of jobs that
run concurrently.
max_concurrent_logins
Returns the maximum number of endpoints
that can log in at the same time.
mcache_bwcontrol
Returns a value that indicates whether
the gateway shares bandwidth control
between the MDist 2 service and the
method cache download.
method_trace_time
Returns the resolution time in seconds
for logging of endpoint methods.
nbi_name
Returns the NetBIOS name.
nbi_mode
Returns the NetBIOS name resolution
mode.
nbi_port
Returns the port on which the gateway
listens for NetBIOS packets.
protocols
Returns the supported protocol for the
specified gateway.
rpc_maxthreads
Returns the maximum number of threads.
send_userlink_file
Returns a value that indicates whether
the gateway is enabled to send the
userlink.htm file to an endpoint.
session_timeout
Returns the amount of time in seconds
that a gateway waits for a response from
an endpoint after sending a downcall.
sna_address
Returns the SNA address.
sna_mode
Returns the SNA mode.
sna_port
Returns the port on which the gateway
listens for SNA packets.
tcp_port
Returns the port on which gateway
listens for TCP packets.
wol_broadcast
Returns a value that indicates whether
the gateway broadcasts a Wake on Lan
packet to the subnet of the endpoint.
wol_direct
Returns a value that indicates whether
the gateway sends a Wake on Lan packet
directly to the IP address of the
endpoint.
get_gatewayproxies
For NetWare gateways only. Displays the list of
managed nodes that have been added to the named
gateway.
logstatus Returns logging status.
remove_gatewayproxy node_name
For NetWare gateways only. Removes a managed
node as an entry from the gateway proxy list.
reset_gatewayproxies
For NetWare gateways only. Clears the gateway
proxy list.
restart Stops and restarts the specified gateway.
rm_protocol IPX
Removes support of the IPX protocol from the
specified gateway. The TCP/IP protocol cannot be
removed.
run Runs the specified suboption. You can specify
the following suboptions with the run option:
sync_login_interval
Synchronizes the endpoint login interval
for the gateway with the endpoint
manager value. You do not have to reboot
the gateway for this change to take effect.
ifs_ignore_remove IP_address
Removes the specified IP address from
the list of excluded gateway aliases.
set option value
Sets the value for the specified option. You can
specify the following options and values with
the set option:
continue_eplogin_onerror {0 | 1 | 2 | 3}
Specifies whether the gateway allows
exeptions from the boot_method method,
the login_policy script, or both, during
normal login attempts by endpoints.
Specify one of the following values:
0 The gateway aborts the login
attempt if receives an exception
from the boot_method method or
login_policy script. This is the
default value.
1 The gateway allows exceptions
from the login_policy script.
2 The gateway allows exceptions
from the boot_method method.
3 The gateway allows exceptions
from both the login_policy
script and the boot_method
method.
debug_level level
Sets the level of message information
that is logged by the gateway. The
levels are as follows:
0 Errors. This is the default and
recommended level.
1 Errors and warnings.
2 Harmless exceptions.
3 Verbose communication
information.
5 Verbose boot, database check,
and endpoint login information.
6 Verbose upcall, downcall, and
repeater information.
7 Verbose job scheduler
information.
8 Verbose gateway cache
information.
Note: Level 4 does not exist.
epcheck_atboot {TRUE | FALSE}
Specifies whether the gateway checks the
status of all endpoints assigned to it
each time the gateway starts up. The
default is TRUE.
epcheck_interval seconds
Specifies the interval, in seconds, at
which the gateway checks the status of
endpoints that it communicates with. The
range for this option is 300 to 14400
seconds. The default is 3600 seconds (1
hour).
epupgrade_by_gateway [TRUE | FALSE]
Upgrades any endpoint assigned to this
gateway during the normal login process.
If the endpoint is already at the most
current level, no upgrade is performed.
The default for all gateways, except
NetWare gateways, is FALSE. For NetWare
gateways, this value is set to TRUE. A
NetWare gateway cannot run a
login_policy script without first
defining a proxy managed node.
Note: The epupgrade_by_gateway option works only if the upgrade_mode parameter in the upgrade.cntl file is set to auto. The upgrade.cntl file is located in the $BINDIR/lcf_bundle.41100/upgrade directory on the gateway.
httpd [TRUE | FALSE]
Enables and disables the gateway HTTP
server. The option requires only the
admin role. TRUE enables the gateway
HTTP server; FALSE disables the gateway
HTTP server. The default is FALSE.
httpd_passwd user_name:password
Sets the user name and password required
to access the gateway HTTP server to
view information about endpoints. This
option requires only the admin role.
ifs_ignore IP_address
Specifies an IP address that is excluded
from the gateway aliases list when the
gateway boots up. Use this option on
multi-NIC gateways to specify network
cards that you do not want endpoints to
connect to. You can specify only one IP
address each time you run this command.
However, you can run this command
multiple times to exclude multiple IP
addresses.
log_dir directory
Specifies the directory on the local
system for the gatelog and all
gatelog.timestamp log files.
log_size size_in_bytes
Specifies the reference size of the
gatelog file, in bytes. When the
referenced size is reached, the gatelog
file is copied to a gatelog.timestamp
file and a new gatelog file is created.
The maximum file size is 10485760 bytes
(10 megabytes), and the default file
size is 1048576 bytes (1 megabyte).
logstatus_interval seconds
Sets the amount of time in seconds
between status checks. If set to zero,
logging is disabled. The range for this
option is zero to 3600 seconds. The
default is 3600 seconds (1 hour).
max_concurrent_jobs count
Specifies the maximum number of jobs
that run concurrently. The range for
this option is 200 to 2000 jobs. The
default is 200.
max_concurrent_logins count
Specifies the maximum number of
endpoints that can log in at the same
time. The range for this option is 100
to 500 endpoints. The default is 100.
mcache_bwcontrol [TRUE | FALSE]
Specifies whether to share bandwidth
control between the MDist 2 service and
the method cache download. The default
is FALSE.
method_trace_time seconds
Specifies the resolution time in seconds
for tracing endpoint activity (last
login time, last logout time, last
migration time, last downcall time, last
upcall time, and last endpoint check
time). When this value is set to zero,
endpoint tracing information is not
written to the epact.bdb database but is
stored in memory on the gateway. The
default is 3600 seconds (1 hour).
protocols protocol_list
Sets a supported protocol for the
specified gateway. TCP/IP is the
default.
rpc_maxthreads count
Specifies the maximum number of threads.
The range for this option is 250 to 2500
threads. The default is 250.
send_userlink_file {TRUE | FALSE}
Specifies whether the gateway sends the
userlink.htm file to an endpoint. When
this value is set to TRUE, the gateway
sends the userlink.htm file to an
endpoint after the endpoint completes
its first successful normal login. The
userlink.htm file is stored on the
endpoint in the /etc/Tivoli directory on
UNIX systems and the c:/etc/Tivoli
directory on Windows systems. When this
value is set to FALSE, the gateway does
not send the userlink.htm file to an
endpoint. The default is TRUE. This
option is designed for new endpoints.
The userlink.htm file might already be
installed on existing endpoints.
session_timeout seconds
Specifies the amount of time in seconds
that a gateway waits for a response from
an endpoint after sending a downcall.
The default is 300 seconds (5 minutes).
wol_broadcast {TRUE | FALSE}
Specifies whether the gateway broadcasts
a Wake on Lan packet to the subnet of
the endpoint. In broadcast mode, if the
IP address of an endpoint is unknown,
the gateway can wake up the endpoint by
sending a Wake on Lan packet to the
subnet of the endpoint. The default is
TRUE.
wol_direct {TRUE | FALSE}
Specifies whether the gateway sends a
Wake on Lan packet directly to the IP
address of the endpoint. The default is
FALSE.
set_ipx_port port
Specifies the port on which the gateway listens
for IPX packets. When an IPX port is assigned,
an SPX port is assigned. The SPX port is always
the assigned IPX port number less one. The
default IPX port number is 9494, and the default
SPX port is 9493.
set_nbi_name name
Specifies the NetBIOS name.
set_nbi_mode mode
Specifies the NetBIOS name resolution mode.
set_nbi_port port
Specifies the port on which the gateway listens
for NetBIOS packets.
set_protocols protocol_list
Sets the supported protocol for the specified
gateway. Supported protocols are TCPIP (the
default) and IPX.
set_sna_address address
Specifies the SNA address.
set_sna_mode mode
Defines the SNA mode.
set_sna_port port
Specifies the port on which the gateway listens
for SNA packets.
set_tcp_port port
Specifies the port on which the gateway listens
for TCP/IP packets. The default port number is 9494.
start Starts the specified gateway. For NetWare, use
the oservrun command to start the object
dispatcher from the NetWare console.
stop Stops the specified gateway. For NetWare, use
the oservend command to stop the object
dispatcher from the NetWare console.
gateway_name
Identifies the name of the gateway.
Authorization admin or senior
wgateway solgw describe Gateway Object : 1094978554.1.579#TMF_Gateway::Gateway# Hostname : packer debug_level : 9 session_timeout : 30 Protocols : TCPIP, IPX TCPIP Port : 41684 IPX Port : 41690 Start Time : 2003/10/24 11:26:05 +06 no_endpoints : 2 log_dir : /data/gem/1013/packer.db log_size : 1024000 rpc_maxthreads : 350 max_concurrent_jobs : 200 max_concurrent_logins : 200 method_trace_time : 1800 logstatus_interval : 1800 login_interval : 0 epcheck_interval : 300 epcheck_atboot : Enabled httpd : Enabled mcache_bwcontrol : Disabled wol_broadcast : Disabled wol_direct : Enabled send_userlink_file : Disabled crypt_mode : NONE ifs_ignore : NIL
wgateway liliana-gateway describe Gateway Object : 1849216852.1.580#TMF_Gateway::Gateway# Hostname : liliana debug_level : 9 session_timeout : 300 Protocols : TCPIP TCPIP Port : 21234 Start Time : 2003/09/26 16:36:03 +06 no_endpoints : 3 log_dir : /data/lorozco/Tivoli-1234/db/liliana.db log_size : 1024000 rpc_maxthreads : 350 max_concurrent_jobs : 200 max_concurrent_logins : 100 method_trace_time : 3600 logstatus_interval : 3600 login_interval : 270 epcheck_interval : 3600 epcheck_atboot : Enabled gateway_httpd : Disabled mcache_bwcontrol : Disabled gateway_crypt_mode : NONE wol_broadcast : Enabled wol_direct : Disabled Interface Ignore List : None send_userlink_file : Enabled
wgateway lux add_gatewayproxy lradner
wgateway lux get_gatewayproxies
wgateway lux run ifs_ignore_remove 9.99.99.98
wgateway lux set max_concurrent_logins 200See Also wcrtgate, wdelgate, wrpt
Copyright © 1996-2008 by Dr. Nikolai Bezroukov. www.softpanorama.org was created as a service to the UN Sustainable Development Networking Programme (SDNP) in the author free time. Submit comments This document is an industrial compilation designed and created exclusively for educational use and is placed under the copyright of the Open Content License(OPL). Original materials copyright belong to respective owners. Quotes are made for educational purposes only in compliance with the fair use doctrine.
Standard disclaimer: The statements, views and opinions presented on this web page are those of the author and are not endorsed by, nor do they necessarily reflect, the opinions of the author present and former employers, SDNP or any other organization the author may be associated with. We do not warrant the correctness of the information provided or its fitness for any purpose.
Last modified: March 15, 2008