Table of Contents
Scenario Commands
Scenario Commands can be called upon in scenarios. They can return nothing, a list variable or a hash variable. When they fail, the <error>-variable is set.
Several of the commands will benefit from the support of 'here documents' as a way to avoid creating trivial templates. Please see the section on Here documents
Commands
cmd_exec
Execute command line-by line on cli of a NetYCE managed node.
Mandatory options:
-n node hostname of the node
-f cmd_file file containing the commands you want to execute
Optional options:
[-o <origin>] use the 'origin' node as the context node. Commands are executed on the -n node.
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
[-q] sets quick-mode: skip config save/backup/nccm
[-e] cancels the normal stop-on-error behaviour
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
The -e option overrides the default behaviour of aborting all remaining commands when an error was detected. When used, the device response is not checked for errors or warnings and execution proceeds with the next command.
The logs will not indicate an error was encountered.
cmd_exec_basic
Execute command line by line on cli of a node in the CMDB table
Mandatory options:
-n node hostname of the node
-f cmd_file file containing the commands you want to execute
-a <node_addr> the address of the node
-d <domain> the management domain of the node
-v <vendor_type> the vendor type of the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
resched_job
Note This command is not available for releases prior to 7.2.0.
Schedule the running job to execute again. The resched_job command allows for recurring jobs. It is included in a scenario to reschedule the same job for a later day and/or time. Since the rescheduled job will execute the same scenario, this job will again be rescheduled for a later moment. In effect, the task will become a repeated job.
To stop the job from rescheduling itself it needs to be manually 'cancelled' using the “Operate - Jobs” tool.
Mandatory options
It is mandatory to specify the -i (interval) or -t (daytime) option.
-i interval reschedule this job at the "interval" after current start time
"interval" format in 'd' 'h' 'm' or a combination like "1d 2h 30m' or '72h"
no 'now' supported
-t daytime reschedule this job at "daytime"
"daytime" format like "tomorrow 5:05" or "sunday 0:00"
no 'now' or date supported
A maximum of eight (8) days can be scheduled ahead of the current time using either the 'interval' or 'daytime' option. The valid values for the 'day' in the “daytime” format are: today, tomorrow, monday, tuesday, wednesday, thursday, friday, saturday, sunday, or their three-letter abbreviations.
Using a '-t' option like 'today 18:00' will result in a job that repeats itself at 18:00 the same day, but since at that time this 18:00 will be past, the re-scheduling will stop. No error is raised for this. If the '-t' is 'tomorrow 18:00' the rescheduled job will again be scheduled for the next day as its targeted time will forever be in the future.
Optional
[-s] set server selection to '-auto-' instead of using same server
Job parameters are reused from the scheduler and job files: - reuse scenario file - reuse command file - reuse parameters file - operator will stay the same - reuse the change-id - reuse the queue - reuse the description - jobtype stays the same - reuse job-rule validations / approvals
The NetYCE server where the job will be rescheduled is the same as the current server by default. Only when the '-s' option is used will the server set to -auto-. It is not possible to assign a different server directly.
Configs
config_create
Generate commands from template.
Mandatory options:
-n node hostname of the node
Optional options:
[-t <template>] overrides the default main-template of the node
[-f <out_file>] override the default output file '<node>.cmd'
[-p parameter=<value>] define or overwrite variables for use in template. it will not overwrite parameters
It should be quoted if the value has whitespaces.
[-i <interface-list>] generate the config(s) for the portnames listed
[-x] extend (append) the commands to the output file if one exists
Parameters already known within the context are not required to be provided manually.
Multiple '-p' arguments may be provided.
Configuration can also be created 'inline' using the following syntax. And pushing the configuration using cmd_exec:
config_create -n <node> -f <node>_ntp.cmd <<EOT ntp server <ntp_server1> |<ntp_server2>|ntp server <ntp_server2> EOT cmd_exec -n <node> -f <node>_ntp.cmd
The here documents describe this functionality in more detail.
config_startup
Upload full config file and make startup.
Mandatory options:
-n node hostname of the node
-f cfg_file file containing the config you want to upload
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
config_save
Save config on device and download to nccm.
Mandatory options:
-n node hostname of the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
[-m <message>] adds a message to the nccm as the cause
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
config_diff
Fetch running config and save it into the nccm
Mandatory options:
-n node hostname of the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
This command will return an error if its backup cannot be retrieved or saved.
config_restore
Upload config from NCCM to the node as startup-configuration.
Mandatory options:
-n node hostname of the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
[-s <select>] specifies the NCCM configuration to select
may be one of:
'previous' select most recent config prior to a command job.
it selects the latest poll or pre-config backup available.
this is the default action if no -s option is provided
'last' select the most recent NCCM backup available
'poll' select the most recent NCCM polled backup
'marked' select the NCCM config manually identified ('marked')
using the NCCM 'Config diff' tools.
the mark will NOT be cleared after execution.
<jobid> select the NCCM config created by specified jobid.
jobs can create a pre-config backup if a config change was detected
before a config change is made, but will also create a post-config backup.
the config selected will be the final (post-config) backup the job executed.
<polltime> select the (first available) NCCM config specified by date and timestamp.
the date (yyy-mm-dd) must included in full, the time may be partial (hh:mm:ss).
a valid option value is therefore "2018-04-25 14:28".
<nccmid> select the NCCM config using the specified nccmid.
this option is only useful when direct database access is available.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
Parsing
parse_cmd
Run a show command on a node, and parse its output. Returns a hash variable with its result, trying to put the results in a regular list variable will fail. (<%variable>, instead of <variable>).
Mandatory options:
-n node hostname of the node
-t <template> a command parsing template. More information on [[howto:command_parsing_templates|command parsing templates]]
-r <request> a request which command to execute to the node, for example "show interface brief"
-c <client_type> the client type of the node. Use this if you have multiple nodes with the same hostname (for example in the CMDB and YCE database)
Some vendors allow the option to specify pipes '|' in the request. This is also allowed to be passed through the command parsing, though you do need to be careful if you wish to use multiple pipes.
An example is shown for Junos, which allows multiple pipes with multiple commands. In this case it is required to have them escaped (escaped with '\').
Before 7.1.0 the syntax was <@cmd>
<%cmd> := Parse_cmd -n <node> -t parsing_template -r "show configuration routing-instances <Vrf_name> \| display set \| match interface"
Device
reboot_node
Restart the device.
Mandatory options:
-n node hostname of the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
clear_console
Reset the console line on the terminal server
Mandatory options:
-n node hostname of the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
[-t <term_srv>] terminal server name
[-l <line>] line number on terminal server to reset
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
File transfer
file_get
Transfer a file TO the named node. The transfer uses SFTP or TFTP depending on the available support from the vendor and the connectivity available.
Mandatory options:
-n node hostname of the node
-s source location and name of the existing file on the NetYCE system
-t target location and name of the new file on the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
file_put
Transfer a file TO the NetYCE system. The transfer uses SFTP or TFTP depending on the available support from the vendor and the connectivity available.
Mandatory options:
-n node hostname of the node
-s source location and name of the existing file on the node
-t target location and name of the new file on the NetYCE system
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-v <vendor>] overrides the vendor type of the node
[-u <operator>] use NetYCE <operator> credentials for first attempt to logon to the device.
The -u option refers to a NetYCE user account. The username and password of this account will be used
to login to the device. When available, the Device password value will be used as the password instead of the login password of the account. This NetYCE account will be used as the first set of credentials to login to the device. The rme_user and local_user credentials will still be attempted should login fail.
OS Upgrades
Also see page on OS-repo scenario calls
os_files
retrieves a list of os files from the OS repository. Returns an error if no os files are found for these criteria.
Optional options:
[-n <name>] the name of the os image",
[-v <vendor>] the vendor type of the os image",
[-s <status>] the status of the os image (production, historic, planned, unknown; default is production)",
[-t <device_type>] the device type of the os image",
[-f <version>] the version of the os image",
[-d <domain>] the node type of the os image",
[-y <node_type>] overrides the node type of the os image",
[-o <file_type>] the file type of the os files: (os-image, boot-image, license, other, unknown; default is os-image)",
os_image_select
retrieve os repo image data. The returned hash variable has the following attributes: Id, Error, Status, Vendor_type, Device_type, Os_name, Os_version, Os_path, Domain, Node_type, Pre_activation_cmd, Post_activation_cmd.
If the options result multiple os images, the image with status production will be returned. If there are more than that, the one with the latest timestamp will be returned. If you want to guarantee that no more than one os image will be selected, then provide a filter for vendor type and device type, since only one os image in production status can exist for each vendor type/device type combination,
Optional options:
[-n <name>] the name of the os image
[-v <vendor>] the vendor type of the os image
[-s <status>] the status of the os image (production, historic, planned, unknown; default is production)
[-t <device_type>] the device type of the os image
[-f <version>] the version of the os image
[-d <domain>] the domain of the os image
[-y <node_type>] the node type of the os image
os_file_select
retrieve the files of an os repo image as a list of hash variables. Each hash variable has the following attributes: Os_file_name, Vendor_type, Type, Status, File_size, File_date, File_md5, File_crc, Storage_min
If the options result multiple os files, the one with the newest timestamp will be returned.
Optional options:
[-i <id>] the os image id the files belong to
[-v <vendor_type>] the vendor type of the file
[-n <name>] the name of the file
[-t <type>] the file type of the file (os-image, boot-image, license, other, unknown). Default is os-image
DNS
dns_add_host
This command is available only with the Infoblox integration operational.
Optional options:
[-n <node>] hostname of the node to add to DNS.
Mandatory if '-f' is not used
Do not use the fqdn here.
[-f <fqdn>] the full qualified host name.
Mandatory if '-n' is not used.
Overrides any associated with '-n'.
[-a <ipv4>] the ipv4-address of the DNS record. Overrides any associated with '-n'.
The ipv4-address must include a /prefix in the format 'address/prefix'.
Use a /32 prefix to bypass the subnet-size check.
[-6 <ipv6> an optional ipv6-address for 'host' records only. Cannot be used without an ipv4 address.
The ipv6-address must include a /prefix. Use prefix /128 to bypass subnet-size test.
[-t <type>] dns record type: 'host' or 'arec'.
Default to 'host'.
[-v <ipam_view>] specify the Infoblox IPAM-view. The default is configured in 'etc/<server>_dhcp.conf'
which is normally 'Default'.
[-d <dns_view>] specify the Infoblox DNS-view. The default is configured in 'etc/<server>_dhcp.conf'
which is normally 'Intern'. The DNS-view is hierarchically dependent on the IPAM-view
[-r] flag to renew the existing ip-address of the DNS record.
Requires the '-a' option to provide the new address
[-i <string>] information string added to the DNS record.
The dns_add_host will create by default a DNS 'host' record for the node using its management ip-address. If the node exists in the YCE database, only the -n option is needed:
dns_add_host -n <node>
If the node is present in the CMDB table, the fqdn is retrieved from this table, but the ip-address must be included explicitly using the -a since it obviously cannot be resolved using the DNS.
# requires subnet '10.1.2.16/29' to exist in IPAM dns_add_host -n <node> -a 10.1.2.17/29
It is mandatory to specify the ip-address with its subnet's prefix (1-32). The DNS policies in place require the IPAM subnet to exist or adding a DNS a-record or host is not permitted.
The only exception to this policy is where a /32 (single address) subnet is targeted. In that case, the address is tested to exist and is available (unused), but the subnets prefix needs not to be explicitly known.
# only requires the address to exist in IPAM, but subnet-size is ignored dns_add_host -n <node> -a 10.1.2.17/32
The '-n' option can be replaced with the -f option to specify the full-qualified-domain-name (fqdn).
# create the DNS host using the fqdn dns_add_host -f new-host.my.domain -a 10.1.2.17/29
If the fqdn is available in the YCE database, the management address of the node is located and used, but otherwise the address option is needed too. The -f option allows a different domain name to used than registered in the YCE database.
Policies demand that a domain name must pre-exist as a 'zone' in the IPAM or the request will be denied. This applies to the explicit domains given in the -f option, but also the implicit domains used with the -n option.
By default a 'host' type record will be created. Setting the -t option with the value 'arec' will create an A-record instead. The -t can be either 'host' or 'arec'. Both record types are checked when adding a new record to verify it does not already exist. The ERROR flag is set when either exists or when any of the many criteria and policies are not met.
The Infoblox environment requires the use of IPAM-views and DNS-views. The DNS-views are associated with the IPAM-views. Both views have defaults assigned in the 'etc/<server>_dhcp.conf' configuration file. Use the -v and -d options to override these defaults.
When creating a host, the fqdn must be unique and not exist, the domain-name must exist in the DNS-view, and the subnet must exist in the IPAM-view. A 'host' record must have at least one ipv4-address assigned, regardless of any ipv6-addresses. The ipv4 address and ipv6 address must reside in the same IPAM-view.
# remove a host from the IPAM-view 'Internet' using the DNS-view 'Exposed' dns_clear_host -n <node> -f "foo.acme.com" -t host -v Internet -d Exposed # add a host to the IPAM-view 'Internet' using the DNS-view 'Exposed' dns_add_host -n <node> -f "foo.acme.com" -t host -a 172.122.34.20/24 -6 1a04:bc8:3001::20/120 -v Internet -d Exposed
Note: currently the ipv6 support is limited to 'host' records. An extension to support AAAA-records is planned.
For additional details, see the article on “Add_host” in Infoblox DNS API plugin. It is this API call that will be executed when using the dns_add_host command.
These are the four new DNS commands and rely on the NetYCE - Infoblox integration. Please note that in these early-release DNS commands there is no support for the Infoblox Extended Attributes.
dns_clear_host
Remove DNS host. This command is available only with the Infoblox integration operational.
Optional options:
[-n <node>] hostname of the node to add to DNS.
Mandatory if '-f' is not used
Do not use the fqdn here.
[-f <fqdn>] the full qualified host name.
Mandatory if '-n' is not used.
Overrides any associated with '-n'.
[-t <type>] host or arec, overrides default host type record
[-v <ipam_view>] specify the Infoblox IPAM-view. The default is configured in 'etc/<server>_dhcp.conf'
which is normally 'Default'.
[-d <dns_view>] specify the Infoblox DNS-view. The default is configured in 'etc/<server>_dhcp.conf'
which is normally 'Intern'. The DNS-view is hierarchically dependent on the IPAM-view
[-c] also remove cnames
dns_add_alias
Add DNS alias.
This command is available only with the Infoblox integration operational. Edit the configuration file /opt/yce/etc/<servername>_dhcp.conf to define GridMaster, policies and defaults.
Use the type ('-t') 'alias' to add an alias to a 'Host'-record or the type 'cname' to create a 'Cname'-record to point to a canonical host or a-record.
The use of the '-n' option implies the node name must exist in the YCE or CMDB environments of NetYCE. Otherwise the '-f' option should be used to provide the full-qualified-domain-name by which the host or canonical a-record record can be located in Infoblox. The node or fqdn cannot be ipv4 or ipv6 addresses.
Aliases should be included as a fqdn. However, should it look like a hostname, and a host domain is available, the alias will be extended with use the host domain to find it. Multiple aliases may be specified in one call.
Options:
-n <node> the hostname of the node in question, either the hostname or the fqdn is required
-f <fqdn> the the fqdn of the node in question, either the hostname or the fqdn is required
-a <alias> alias name(s) for <node> as fqdn or hostname (where the <node> domain will be used)
[-t <type>] 'alias' or 'cname', default is 'cname' type record
[-d <dns_view>] specify the Infoblox DNS view name - default set in etc/<server>_dhcp.conf
[-i <info>] add a comment to the record
[-e <ext_attr>] define an extended attributes (use "param=value" format)
dns_clear_alias
Remove DNS alias.
This command is available only with the Infoblox integration operational. Edit the configuration file /opt/yce/etc/<servername>_dhcp.conf to define GridMaster, policies and defaults.
The behaviour when removing an alias from a 'Host'-record is different from removing a 'Cname'-record that points to a canonical host- or a-record.
When type ('-t') is 'alias', the aliases ('-a') will be removed from an existing 'Host'-record. The use of node ('-n') or fqdn ('-f') option is then mandatory. The included aliases will only be removed if they belong to that host record.
If the type is 'cname', the use of the node ('-n') or fqdn ('-f') is optional. When included, this information is used to verify the canonical name matches the cname. If not, the cname is not removed. Omitting the node and fqdn will result in removing the cname without verification.
The use of the '-n' option implies the node name must exist in the YCE or CMDB environments of NetYCE. Otherwise the '-f' option should be used to provide the full-qualified-domain-name by which the host or canonical a-record record can be located in Infoblox. The node or fqdn cannot be ipv4 or ipv6 addresses.
Aliases should be included as a fqdn. However, should it look like a hostname, and a host domain is available, the alias will be extended with use the host domain to find it. Multiple aliases may be specified in one call.
Options:
-n <node> the hostname of the canonical node, either the hostname or the fqdn is required for host type
-f <fqdn> the fqdn of the canonical node, either the hostname or the fqdn is required for host type
-a <alias> alias name(s) for <node> as fqdn or hostname (where the <node> domain will be used)
[-t <type>] 'alias' or 'cname', default is 'alias' type record
[-d <dns_view>] specify the Infoblox DNS view name - default set in etc/<server>_dhcp.conf
dns_clear_ip
This command is available only with the Infoblox integration operational.
The dns_clear_ip command will search the DNS for any Host or A-record referring to an ipv4-address and remove it. Any CNAME records pointing to these removed records will be removed as well.
Since an Infoblox Host construct can be associated with multiple ip-records, only the targeted ip-address is removed. Should such a Host constrict at that point have no ip-addresses left, then the entire Host is deleted. CNAMES pointing to Hosts constructs are preserved if ipv4-addresses remain.
Optional options:
[-a <ipv4>] the ipv4-address of the DNS records to search for and remove (without prefix).
Overrides -n and -f option.
[-n <node>] nodename of the node whose management ip-address is to be removed.
Do not use the fqdn here, the nodename is assumed to be a known NetYCE object.
[-f <fqdn>] full qualified host name of a DNS record to be removed using its resolved ip-address.
Only the first ip-address that the DNS returns is searched for.
By default, the DNS view searched is 'Intern' (as defined in the NetYCE 'DNS Infoblox API' plugin).
Examples:
dns_clear_ip -a 10.106.16.10 dns_clear_ip -n ams-cr01 dns_clear_ip -f server-1234.acme.com
External calls
The external calls is a group of traps, signals and other exec calls.
trap_nms
The trap_nms command sends a SNMP trap to the NMS management stations. The management stations are specified
using the -a option, one for each station.
The trap can include message details that require the -e option with the SNMP enterprise ID and a -v
with the variable OID (object-id). Then by including a -m option with a message, this message gets assigned to the variable OID.
By including more -m messages, each is then assigned to the next OID by auto-incrementing the last digit of
the OID. The order of the -m's determine which OID is used. The default -v is '1.0' which is used for the first -m, the second would then be 1.1 and so on. By defining your own -v <varoid> you have full control over the object-identifier.
Example:
[scenario] Description <node> trap-test trap_nms -a 172.17.10.21 -a 172.17.10.28 -e 1006 -v 3.2.1 -m "hostname:<hostname>" -m "domain:<domain>" -m "somevar:someval" end
The traps sent will in this case include a 'varbindlist' with the following information:
{
'1.3.6.1.4.1.1006.3.2.1' => 'hostname:ACME-GV-023',
'1.3.6.1.4.1.1006.3.2.2' => 'domain:pub.acme.com'
'1.3.6.1.4.1.1006.3.2.3' => 'somevar:someval',
},
Mandatory options:
-a <nms_addr> send the trap to this address
Optional options:
[-e <enterprise>] the enterprise OID to use (default 696)
[-v <var_oid>] the variable OID to use (default 1.0)
[-m <message>] the OID message text
[-s <specific>] the specific trap type to use (default 0)
trap_node
The trap_node command sends a SNMP trap to the NMS management stations but 'spoofs' the node as the originating address. The management stations are specified using the -a option, one for each station.
The trap can include message details that require the -e option with the SNMP enterprise ID and a -v
with the variable OID (object-id). Then by including a -m option with a message, this message gets assigned to the variable OID.
By including more -m messages, each is then assigned to the next OID by auto-incrementing the last digit of
the OID. The order of the -m's determine which OID is used. The default -v is '1.0' which is used for the first -m, the second would then be 1.1 and so on. By defining your own -v <varoid> you have full control over the object-identifier.
Example:
[scenario] Description <node> trap-test trap_node -n <node> -a 172.17.10.21 -e 1006 -v 3.2.1 -m "hostname:<hostname>" -m "domain:<domain>" -m "somevar:someval" end
The traps sent will in this case include a 'varbindlist' with the following information:
{
'1.3.6.1.4.1.1006.3.2.1' => 'hostname:ACME-GV-023',
'1.3.6.1.4.1.1006.3.2.2' => 'domain:pub.acme.com'
'1.3.6.1.4.1.1006.3.2.3' => 'somevar:someval',
},
Mandatory options:
-n <node> hostname of the node
-a <nms_addr> send the trap to this address
Optional options:
[-e <enterprise>] the enterprise OID to use (default 696)
[-v <var_oid>] the variable OID to use (default 1.0)
[-m <message>] the OID message text
[-s <specific>] the specific IOD to use (default 0)
signal_json
A RESTFUL JSON call can be added to the scenario using signal_json.
Mandatory options:
-t <target> Define the target application (server and url)
Optional options:
-p "variable=value" Provide variable-value pairs to include in Json post
-n <node> Provide the nodename
-m <message> Provide the message you wish to include
-a <action>] Provide the action value
-s <status> Provide the status
This call is dependent on the configuration file /opt/yce/etc/signal_json.conf and can be modified using the edit configs page under system. If the configuration file doesn't exists yet, it will be created the first time the signal_json command is executed from a scenario.
This configuration file must be setup to define the various target applications. These targets represent for
example the ticketing system, the monitoring tool or the cmdb. Each of these targets have attributes like: host, URL path and authentication details.
since these details may vary per NetYCE server, these targets have their attributes defined per NetYCE hostname, or when they are identical to all servers, use the 'default'.
Example signal_json.conf file:
our $json_config = { 'ticketing' => { 'default' => { host => 'http://1.2.3.4:8080', url => '/api/v1/services', # provide for websevice basic authentication username => undef, password => undef, # provide for webservice bearer authentication webtoken => 'abcdef', }, 'genesis' => { host => 'http://1.2.3.4:8888', url => '/api/v1/services', # provide for websevice basic authentication username => 'my_username', password => 'my_passwd', # provide for webservice bearer authentication webtoken => undef, }, }, 'cmdb' => { 'default' => { host => 'http://11.12.13.14:8080', url => '/api/v1/services', # provide for websevice basic authentication username => undef, password => undef, # provide for webservice bearer authentication webtoken => 'abcdef', }, }, };
The JSON dataset has a few pre-defined attributes (node, message, status, action) but can easily be extended using the -p options. These -p options allow you to add any number of additional attributes.
To allow nesting of attributes (called hashes), the parameter name in the -p option may be using dots (.). These dots separate the key names used in the hash.
A simple example may clarify their use:
signal_json -t ticketing -n <node> -m "going down" -p "moo.one='one value'" -p "moo.two=two value" -p moo.three="three value" -p "moo.four.one = 'four value=none"
Results in:
using Json config for target 'ticketing' and server 'default'
#--- JSON -----------------------------------------
{
"jobid" : "0619_0031",
"target" : "ticketing",
"action" : "",
"status" : "",
"node" : "ASD-CR02001",
"message" : "going down",
"moo" : {
"three" : "three value",
"one" : "one value",
"two" : "two value",
"four" : {
"one" : "four value=none"
}
}
}
Submit to webservice: http://1.2.3.4:8080/api/v1/services
st_exec
ST_EXEC allows the job to execute a Service-type from the scenario. Traditionally, changes are first prepared before scheduling a job. The preparation might include using Service-types to automate it. But in some cases the sequence of events must be reversed (e.g. deleting an object from the database before removing it from the network), or the choice and parameterization of the Service-type is dictated by the job/scenario status.
Any Service-type can be used with 'st_exec' provided the operator has sufficient permissions for the Client_type, or the call will be rejected at run-time.
If the Service-type uses API 'custom' variables, these MUST be included using '-p “var=value”' options or the call will be rejected at run-time. The logs, list the located custom vars of the Service-type, and any missing will be logged in the error message.
[-b <client_type>] client_type of Service-type to execute. Overrides client_type of node
[-c <service_class>] service_class of Service-type to execute. Overrides service_class of node
[-s <service_type>] service_type of Service-type to execute. Overrides service_type of node
-t <service_task> mandatory service_task of Service_type to execute
[-n <node>] optional nodename (YCE nodes only) to provide the -b, -c, -s options.
Also sets 'current' service, client and node
[-a <ipv4_subnet>] optional ipv4 subnet/prefix in service-key or global ipv4-subnet-id.
Sets 'current' subnet. Requires -n, excludes -6
[-6 <ipv6_subnet>] optional ipv6 subnet/prefix in service-key or global ipv6_subnet-id.
Sets 'current' subnet. Requires -n, excludes -a
[-p <parameter=value>] define Service-type 'custom' parameters. Mandatory for every custom used
in Service-type. Use multiple '-p'
The simplest call uses only the '-n node' and '-t service_task' options. By including the '-n node' the st_exec will use the client_type, service_class and service_type from the service the node is located in. If any of -b, -c or -s is specified, it will overrule the values found for the node.
[scenario]
Description <node> Add_vpn by ST
st_exec -n <node> -t "Add_vpn"
if <error>
log -m "st_exec failed"
stop
endif
Service-types using 'current'
When including the '-n node', some of the service-type commands that match on 'CURRENT' - representing the selected object in the GUI - will be set. The 'current' values for the service, client, site, and node are thus defined and allow the Service-type to be executed despite the lack of a GUI selection.
Note however, that 'current' subnet, ipv6-net or dhcp objects are not available by setting -n alone. Add the '-a ipv4net/prefix' or '-6 ipv6net/prefix' to 'select' the current subnet if the Service-type requires it. The subnet will be searched for using the service-key of the node. If the -a or -6 value is the ipv4/ipv6 subnet-id, then this subnet will be recognized as the 'current' subnet. The -a and -6 options can only be used with the -n.
API Service-types
A Service-type may also be fully specified by including the -b, -c -s, and -t options. This form is commonly used to execute API based Service-types. Since they usually require custom variables, these variables must be included using -p options. Before execution of the Service-type, all customs are located and logged. If -p variables are missing, the st_exec will not start and produces an error.
The example below shows the -p options one separate lines following the st_exec call. This is a supported format and is used to improve readability. Options are automatically joined if the first part of the line is a recognizable option ('-a ', '-6 '). Comments cannot be included.
[scenario]
Description Add_dmz_frw to LB
st_exec -b 'DC' -c 'LB' -s "api" -t "Add_dmz_frw"
-p "frw_name = DC2-FW01"
-p "lb_name = DC2-LB02"
-p "pe_a_name = DC2-PE01"
-p "pe_b_name = DC2-PE02"
-p "vrf_id = 100"
-p "vrf_template = dmz_vrf"
if <error>
log -m "st_exec failed"
stop
endif
It is allowed and sometimes useful to include a '-n node' option even when all four Service-task specifying options are explicitly set. In this case the node will provide the 'CURRENT' values for this node to be used in the Service-type. Of course, the use cases for this setup are very limited.
ansible_exec
This calls upon the local ansible installation. It allows the launching of either an existing Playbook, or the launching of a NetYCE Scenariobook by Ansible. Where the Playbook option refers to a pre-existing playbook-file that can be used as-is, the Scenariobook refers to the NetYCE template-based generated Playbook that can be parameterised using the NetYCE modelling and database.
Optional options:
[-n <node>] the node in question
[-p <playbook.yml>] Use pre-existing Playbook YML file - include full path
[-s <scenariobook>] Use playbook created in a NetYCE job
[-o "-vvvv"] Use ansible options - https://docs.ansible.com/ansible/latest/cli/ansible-playbook.html"
Example, static yaml file:
ansible_exec -n <node> -p /opt/update_ntp.yml
Example, dynamic yaml file, which is generated by NetYCE:
config_create -n <node> -t update_ntp -f update_ntp.yml ansible_exec -n <node> -s update_ntp.yml
script_exec
This calls upon a local script. Additional arguments can be provided to the script and if needed the interpreter can be overridden.
Optional options:
-s script Script to execute - full path may be included [-i <interpreter>] Override interpreter to execute the script [-o <arguments>] The script arguments. multiple are allowed [-t <timeout>] The timeout in seconds before forcibly killing the script. defaults to 300
Example, of the python script execution:
<output> := script_exec -i /usr/bin/python2.6 -s /opt/scripts/some_script.py -o variable1 -o variable2 <output> := script_exec -s /opt/scripts/some_script.py -o variable1 -o variable2 -t 120
List Operations
calc
Executes the given calculation
Mandatory options:
-c <calculation> string containing the calculation, may be a variable
Example:
<valA> := 5 <newVal> := calc -c '5 + <valA>' # Output in the job log: Assignment: (valA -- result: (5)) Assignment: (newVal -- result: (10))
relation
Extract a variable (list) from a named relation, and returns this to a list variable. More information on relations.
Mandatory options:
-n <node> hostname of the node
-r <relation> name of the relation (context)
-v <column> the name of the relation column (variable) returned.
Optional options:
[-p <parameter=value>] provide additional relation variable=value pairs
[-f <column=value> ] define filter column=value pairs. The value supports wildcards ('*' and '?')
the relation row must match all filters to be included in the result
[-l <limit>] return the <limit> number of entries. The limit must be > 0
Example:
Using variable: <Cpe_hostname>
<CPE_Vrf_id> := relation -n <Cpe_hostname> -r Node_vrf -v Vrf_id -p "Hostname=<Cpe_hostname>"
replace
Replace searches for a specific string-value and replaces it with another string-value. The replace value can be empty. The match string is non case-sensitive.
Mandatory options:
-l <list> Provide the list value, it will use the first entry
-c <match> The string that needs to match
Optional options:
[-r <replace>] The match will be replace with this string
[-a] Provide this to replace every match
Example:
<valA> := "abc" <valB> := "def" <newVal> := replace -l <valA> -c 'bc' -r <valB> # Output in the job log: Assignment: (valA -- result: (abc)) Assignment: (valB -- result: (def)) Assignment: (newVal -- result: (adef))
sort
Takes one or more lists and sorts its elements ascending. Returns a list variable.
Mandatory options:
-l <list> a list variable
Optional options:
[-r] reverse sort
[-n] sort the list numerically
[-u] return unique elements
Example:
<listA> += "1"
<listA> += "2"
<listA> += "3"
<listB> += "3"
<listB> += "5"
<listB> += "2"
<listB> += "4"
# sort both lists numerically in reverse keeping only unique items
<sorted> := sort -n -u -r -l <listA> -l <listB>
<msg> = concat -s ", " -l <sorted>
log -m "items: <msg>"
# prints the log message:
# 2019-03-14 11:43:35 items: 5, 4, 3, 2, 1
concat
Takes one or more lists and concatenates its values in a single string. Each value is separated by the separator string specified with the '-s' option. Returns a list variable with a single string value.
Mandatory options:
-s <separator> separator string, default to a space (' ')
-l <list> a list variable
Example:
<listA> += "1"
<listA> += "2"
<listA> += "3"
<listB> += "one"
<listB> += "two"
<listB> += "three"
<msg> = concat -s ", " -l <listA> -l <listB>
log -m "items: <msg>"
# prints the log message:
# 2019-03-14 12:17:52 items: 1, 2, 3, one, two, three
like
The 'like' command was removed in version 7.1.1. This command is superseded by the 'grep' command which supports both regex and wildcard formats for the condition.
grep
Takes one or more lists and a condition, and runs that condition against a wildcard or a regex over the list. Returns all matching elements as a list variable.
Mandatory options:
-l <list> a list variable
-c <condition> a wildcard string: "some*ing" or a regex string: "/some.*ing/"
Other Options:
[-u] returns unique entries in the result. Duplicates are dropped, but only if the entries are really identical (case-sensitive).
NOTE: the difference for the regex. It requires the two '/' characters to mark the regex pattern (/pattern/). It will be always be case sensitive unless the 'i' modifier is used after the last slash: /pattern/i.
NOTE: Variables that were defined as [parameters] will be substituted with the '-c' argument. Run-time variables will not, but can be as a condition directly.
<filter> := grep -l <list> -c "*<var>*" # will only work when <var> is defined in the [parameter] section <filter> := grep -l <list> -c <var> # will work with run-time variable <var>
Example:
[parameters]
[scenario]
Description <node> grep test
# fill a list with relation entries
# (must use += because each relation entry will copy this line in the final task before the scenario starts)
<net_names> += <net_name@vlans>
# copy this list to another list
<all_names> := <net_names>
# append some entries to this list
<all_names> += "eenie"
<all_names> += "meenie"
<all_names> += "mynie"
<all_names> += "moo"
# filter the entries ending in "nie" using REGEX
<nie_names> := grep -l <all_names> -c "/.*nie$/"
# or using wildcard:
# <nie_names> := grep -l <all_names> -c "*nie"
# and log them one by one
foreach <name> in <nie_names>
log -m nie_name: <name>
endeach
end
The last lines of the log shows only ‘eenie’, ‘meenie’ and ‘mynie’ are grepped. The -c “nie” would have retrieved the same entries but might also have found ‘denied’.
match
Takes one or more lists and a condition, and runs that condition against a wildcard or a regex over the list. Returns a list with the matching substring(s) of each element.
The behavior of 'match' is similar to the 'grep' function but with an essential difference: where the 'grep' returns the actual element of the matching condition, the 'match' returns only the matching string. In essence, the 'match' is a grep function and a substring function rolled into one.
Mandatory options:
-l <list> a list variable
-c <condition> a wildcard string: "some*ing" or a regex string: "/some.*ing/"
Other Options:
[-u] returns unique entries in the result. Duplicates are dropped, but only if the entries are really identical (case-sensitive).
As with 'grep', the 'match' supports both wildcard conditions and regex conditions.
The wildcard condition support the '*' and '?' characters to match multiple or single characters. The wildcard condition uses a format like -c “va?ue*”. Wildcard conditions are case-insensitive.
The regex condition uses a format like -c “/val.ue.*/i”. The regular expression itself is positioned between slashes ('/…/') and optional modifiers are appended after the last slash. The 'i' modifier causes the matching to be case-insensitive.
Example:
This example illustrates the difference between the 'match' and the 'grep' functions. It uses the wildcard format for the condition. Grep only returns if there is a full match on the entire string. Match will get return the matched portion.
[scenario] Description test match function <list1> := "the first element" <list1> += "the second element" <list2> := "The third one" <list2> += "The last one" <grepped> := grep -l <list1> -l <list2> -c "*first*" foreach <entry> in <grepped> log -m "grep entry: <entry>" endeach <matched> := match -l <list1> -l <list2> -c "*first" foreach <entry> in <matched> log -m "match entry: <entry>" endeach end
Results:
The grep entries: "the first element" The matched entries "The first"
When using the regex condition a distinct feature can be used: multiple substring matches. Regular expressions have the capability to 'mark' matched strings using '()' and refer to them by position ($1, $2 and so on). The 'match' function allows you to use this feature to use these marks in the condition regex and have these marked strings returned as a separate entry in the result.
When using these '()' markers, only those matches are returned. Without these markers the matching part of the condition is returned as was the case in the example above.
Example:
The same dataset as above using the regex condition below: <matched> := match -l <list1> -l <list2> -c "/the (.*) (.*)/i" Results in the matched entries: "first" "element" "second" "element" "third" "one" "last" "one" And, when adding the optional '-u' creates a sorted list of unique values: <matched> := match -l <list1> -l <list2> -c "/the (.*) (.*)/i" -u "element" "first" "last" "one" "second" "third"
NOTE: Variables that were defined as [parameters] will be substituted with the '-c' argument. Run-time variables will not, but can be as a condition directly:
# this will only work when <var> is defined in the [parameter] section or # is a parameter that is defined in de <node> context (standard template variables). <found> := match -l <list> -c "*<var>*" # this will work with run-time variable <var> that was defined in the scenario section <var> = "value*" <found> := match -l <list> -c <var>
NOTE: Regular expressions support the use of 'shorthand' character groups. A commonly used one is \s to define a whitespace (being space or tab). When using these in a condition, you have to protect these backslashes with a backslash:
Using spaces: <found> := match -l <list1> -l <list2> -c "/the (.*) (.*)/i" Using \s: <found> := match -l <list1> -l <list2> -c "/the\\s(.*)\\s(.*)$/i"
split
Takes list(s) and a regex or wildcard separator to split the element strings into more elements; returns a list with the separated substring of each element.
Mandatory options:
-l <list> the list variable in question; multiple -l allowed
-c <separator> a wildcard separator string, 'var*ble' or a regex string '/\svar(.*)\s/i'"
Other Options:
[-n <number>] limit number of returned elements
[-u] return unique elements as a sorted list
Example:
<family> := "father, mother, son, daughter" <family> += "grandpa, grandma,uncle , aunt" # one way using regex <members> = split -l <family> -c "/\s*,\s*/" # another using wildcards <members> = split -l <family> -c ", " # same, but with a result limit of 1. <members> = split -l <family> -c ", " -n 1
Misc
pingable
Test if node is pingable
Mandatory options:
-n <node> hostname of the node
Optional options:
[-a <address>] override the node address
reachable
test if node is reachable using ssh or telnet
Mandatory options:
-n <node> hostname of the node
Optional options:
[-a <address>] override the node address
[-p <port_list>] override the default ports "22, 23"
hangup
Disconnect node connection(s).
Mandatory options:
-n <node> hostname of the node
Optional options:
[-c <connect>] use 'console', 'api' or 'management' connection
wait
Delay all actions for <time> seconds.
Mandatory options:
-t <time> time in seconds
update_rev
Update database with active template revision.
Mandatory options:
-n <node> hostname of the node
Database
shortest_path
Locate the devices that make up the shortest path between end points. Returns a variable list of all nodes making up the path including the start and end points.
The shortest-path uses the topology (links between ports) of the nodes in the YCE database. Tracing a path is therefore limited to YCE nodes and the accuracy of the topology.
For finding the shortest path the Dijkstra greedy algorithm is used. The model uses any kind of topology between ports of different nodes as a valid path and takes into account the modelling of the hierarchy in that uplink directions have a cost of '1', downlink directions a cost of '3' and interconnections a cost of '2'. Thus resulting in a path where uplinks are preferred and core routes are not circumvented.
The algorithm will find exactly one path that has the lowest cost even is there are multiple paths due to redundancy.
Mandatory options:
-s <node-name> Starting YCE node name of path
-e <node-name> Ending YCE node name of path
Optional:
-a <avoid-string> String of ordered node-names defining links to avoid using a higher cost
-a [<avoid-list>] Converts the <avoid-list> variable in a separated string
The '-a' option is intended to find an alternate, secondary path between the start and end nodes once the primary shortest path was found. Using this option, the primary path is passed as an argument in a second call to 'shortest_path' where it is used to increase the cost of the links along this path. These higher costs will allow the algorithm to find an alternative path if it has a lower cost.
<path_pri> := shortest_path -s "switch1" -e "switch2" <path_sec> := shortest_path -s "switch1" -e "switch2" -a [<path_pri>]
The example shows the use of the special notation “-a [<path_pri>]” that is needed here. Where the regular form (“-a <path_pri>”) to pass an option to a command takes only the FIRST element of the variable, this form concatenates ALL elements of the variable into one string. The use of '[..]' simplifies passing all elements separately using multiple '-a' options or having to convert the list into a string.
When using the '-a' option, the nodes names must form an ordered path. Also when multiple '-a' options are used, the resulting names are combined in a single path. Any '-a' value may be a single node name or a separated string of node names where the separator may be a space, a comma, a semicolon or a vertical bar. If a topology link exists between two neighbouring node names, the cost of that link will be increased by 5.
Note that the increase in cost is unidirectional: the cost of a path in the opposite direction remains unaffected and could therefore still be included in the resulting secondary path. In this and similar cases where a specific link or direction is to be avoided as well, the given path can be extended by including additional sets of nodes. Although not part of the primary path this will result in a higher cost for the link between these two nodes in the given direction. This manipulation of costs can also be used to find the primary path.
In the example below the primary and secondary paths must avoid the link between “switch1” and “switch2” in either direction:
<path_pri> := shortest_path -s "switch1" -e "switch2" -a "switch3 switch4" -a "switch4 switch3" <path_sec> := shortest_path -s "switch1" -e "switch2" -a [<path_pri>] -a "switch3 switch4" -a "switch4 switch3"
db_query
Retrieve a variable(-list) from the database. Returns a variable list.
Mandatory options:
-t <table> table to retrieve information from.
-f <field> the field to put in the return variable
-w '<where>' where-arguments (for example: 'ClientCode="1000"'
db_update
Set a value in the database using a sql statement.
Mandatory options:
-t <table> table to retrieve information from.
-p '<parameter=value>' parameters that need to be set
-w '<where>' where-arguments (for example: 'ClientCode="1000"'
Make sure you put the entire part after -p and -w in quotes.
In case you wish to update Custom Attributes, the where would contain pipes ('|'). These would be interpreted as conditionals and need to be escaped using a '\'. Example:
db_update -t Par_vals -p "Var_value=<Var_value>" -w 'Par_key="<ClientCode>\|<SiteCode>\|<Hostname>" AND Var_name="var_name"'
Logging
For the logging, as well as the scenario in general and templates, it is possible to show/use the job_id using <jobid>.
log
Insert timestamped message in the log of the job.
Mandatory options:
-m "<message>" a string of text, in between quotes.
log_action
Add entry to action log.
Mandatory options:
-n <node> hostname of the node
Optional options:
[-a <action>] name the action type to list
[-m "<message text>"] the message to log. Make sure you use quotes around this variable.
send_email
Send email notification to the operator/group mail addresses, if known.
This behaviour can be adjusted using the send_email_to tweak.
Optional options:
[-s <subject>] the email subject
[-m <message_text>] the email message text
[-t <mail to>] the email address to send email to.
[-f <mail from>] the email address to send email from.
[-a <attachment-file>] filename to include as csv attachment. multiple -a are allowed
[-r <custom-report>] custom report to include as csv attachment. multiple -r are allowed
Multiple '-t <email-recipient>' options are allowed, as well as a single -t option with a (quoted) list of email-addresses.
Example
See the entry below on csv_file
csv_file
Create csv file of list variable(s)
Mandatory options:
-r <name> the csv file name to be created in the job directory. Spaces and special characters are replaced with '_', and '.csv' is added
-v <var_name>' variable name with a list of values to form a report column. add a '-v' for each column
Optional options:
[-s <separator>] separator character or string between columns. defaults to comma (',')
[-d] change file format from unix (lf) to dos (crlf)
The generated csv report will be created in the job-directory named after the job-id (/var/opt/yce/jobs/<job_id>)
and will receive the .csv extension. This file will be displayed in the pop-up window when viewing the job files. It includes a link to download the file.
Tip: The report file can be emailed directly from the job scenario to the operator by using the send_email command by attaching the report file to the email (the '-a <file>' option was added for this purpose).
Example
The example below uses data from the relation “vlans” to fill three list variables. The csv_report then creates a report from this data using these variables. Note that the report inserts the variable names as the csv-header using the casing as typed, in this case with a leading capital. Non-existing variables or empty variables included in the report generate no error but are included as a blank column.
The report name can include variables as well as is demonstrated. The -d option appends todays date to the report name.
The same report is then created as a local csv-file using csv_file. The '-d' option now converts the standard unix file into dos.
Finally the report is sent to the local 'yce' user using send_email. To demonstrate the send_email attachment file handling two more files are attached. Here, the '.csv' and '.log' extensions are tried to locate the report and log files in the job directory. But for other extensions the proper name must be used.
[scenario] Description <node> mail test <net_names> += <net_name@vlans> <net_vlans> += <vlan_id@vlans> <net_addrs> += <net_address@vlans> csv_report -r "<node>_nets" -v Net_names -v Net_vlans -v Net_addrs -v Nope -d csv_file -r "<node>_nets" -v Net_names -v Net_vlans -v Net_addrs -v Nope -d send_email -a "<node>_nets" -a "<jobid>" -a "<jobid>.tsk" -t [email protected] end
When executing the following logs were produced:
28-Function (csv_report -r "ASD-CR98001_nets" -v Net_names -v Net_vlans -v Net_addrs -v Nope -d) saved report as 'ASD-CR98001_nets-20191217' 29-Function (csv_file -r "ASD-CR98001_nets" -v Net_names -v Net_vlans -v Net_addrs -v Nope -d) saving report to '/var/opt/yce/jobs/1217_0053/ASD-CR98001_nets.csv' 30-Function (send_email -a "ASD-CR98001_nets" -a "1217_0053" -a "1217_0053.tsk" -t [email protected]) including 'ASD-CR98001_nets.csv' as attachment including '1217_0053.log' as attachment including '1217_0053.tsk' as attachment send_email: '' sent to '[email protected]'
csv_report
Create custom csv report of list variable(s)
Mandatory options:
-r <name> the report name to be added to the custom reports
-v <var_name>' variable name with a list of values to form a report column. add a '-v' for each column
Optional options:
[-s <separator>] separator character or string between columns. defaults to comma (',')
[-d] add the creation date to the report name ('_yyyymmdd')
The generated csv report will be added to the Operate - Reports - View reports menu and is subject to the same cleanup and ageing process as the other reports. The report can be retrieved using the 'fetch_report' API call and/or the download URL.
Example
See the section above on csv_file
Error handling
All scenario commands can set the 'error' flag if their execution is considered failed. When this flag is set (an error occurred) it will normally have no effect. Only when the scenario checks for it can it affect the further execution of the scenario. To stop the scenario execution and mark it 'ABORTED' the stop command MUST be included.
When the scenario reaches the end command (that is always present at the bottom of the scenario) the job status will be 'SUCCESSFUL' regardless of any errors it encountered. Therefore a scenario that contains no error handling or a stop command will by definition be successful.
reachable -n <node> if <error> log -m "node <node> is non-responding" stop endif cmd_exec -n <node> -f <node>.cmd -v if <error> log -m "command failure" stop endif end
Since having to include explicit error checking for every command in a lengthy scenario can be quite tedious, the default error behaviour can be modified to automatically ABORT the scenario at the failing command. This is accomplished with the stop on-error command. The above could then be simplified as:
stop on-error reachable -n <node> cmd_exec -n <node> -f <node>.cmd -v end
Note that the automatic abort will ignore any error handling that is present in the scenario since the scenario will stop before it reaches the next line with the 'if error' test.
At the point where the normal error handling by the scenario must be resumed the stop default command can be inserted. This allows for explicit error handling where it is needed and skip it where it might be implicit.
stop on-error reachable -n <node> stop default cmd_exec -n <node> -f <node>.cmd -v if <error> log -m "command failure" stop endif end
Here documents
At times it is cumbersome having to create and define lots of trivial or very small templates. In those cases the scenario command can support HERE DOCUMENTS. This name is a reference to unix systems where multi-line input can be given for a command. In our case that input represents the content of a template.
A here document uses the < < and a terminator string that is defined at the end of the command line. The next line all the way up to the terminator string on a line by itself then forms the here document that will be used as the template text.
An example using the scenario command cmd_exec.
The string EOT is used here as terminator (end-of-text), but any string is allowed.
description <node> here document test
cmd_exec -n <node> -f here.cmd <<EOT
!
a sample command for hostname <node> of customer <ClientCode>
another command for the node in <SiteCode>
the full template syntax can be used in a here-document
| condition | {template}
!
EOT
This example will create the command file 'here.cmd' with the content below and then will execute these lines one-by-one on the node's cli:
! a sample command for hostname AMS-DC99001 of customer ACME another command for the node in AMS-DC99 the full template syntax can be used in a here-document !
Please note that in the example above the cmd_exec was invoked with the option -f here.cmd. An explicitly named output file is essential here since the generated cli commands from the here document will be stored in this file.
The second example shows how a more complex command file can be created using a few config_create commands, with and without the use of here documents. The idea is that the command file is created bit-by-bit over the execution the scenario using command or config parsing.
Since the default behaviour of the config_create command is to restart the config file on each invocation, the -x ('extend') option must be included to append new configuration lines.
description <node> here document test
config_create -n <node> -f my.cnf <<EOT
#
SOME COMMANDS FOR <node>
#
EOT
config_create -n <node> -f my.cnf -x -t clear_atm
config_create -n <node> -f my.cnf -x <<EOT
#
SOME MORE COMMANDS FOR <node>
{create_atm}
#
EOT
cmd_exec -n <node> -f my.cnf
In this case the finally resulting config file 'my.cnf' is then executed on the node.
Deprecated
These commands are still supported for backwards compatibility, but they are not meant to be used in new scenarios. When upgrading to version 7.0.2, a conversion was executed whereby the old commands and options were replaced with the new. Creating scenarios using the old commands is still supported for the time being.
import_cfg
Execute command line-by line on cli of a NetYCE managed node.
Mandatory options:
-n node hostname of the node
-f cmd_file file containing the commands you want to execute
Optional options:
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
basic_import
Execute command line by line on cli of a node in the cmdb table - does not support templates.
Mandatory options:
-n node hostname of the node
-f cmd_file file containing the commands you want to execute
Optional options:
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
gen_startup
Generate commands from template.
Mandatory options:
-n node hostname of the node
Optional options:
[-t <template>] overrides the default main-template of the node
[-p <parameter=value list>] define or overwrite parameters for use in template
[-i <interface-list>] generate the config(s) for the portnames listed
write_startup
Upload full config file and make startup.
Mandatory options:
-n node hostname of the node
-f cfg_file file containing the config you want to upload
Optional options:
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
save_config
Save config on device and download to nccm.
Mandatory options:
-n node hostname of the node
Optional options:
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
[-m <message>] adds a message to the nccm as the cause
diff_config
Fetch running config and compare to latest nccm config.
Mandatory options:
-n node hostname of the node
Optional options:
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
reload
Restart the device.
Mandatory options:
-n node hostname of the node
Optional options:
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
restore
Upload config from nccm and reload.
Mandatory options:
-n node hostname of the node
Optional options:
[-a <address>] overrides the address of the node
[-d <domain>] overrides the management domain of the node
LogAction
Add entry to action log.
Mandatory options:
-n <node> hostname of the node
Optional options:
[-a <action>] name the action type to list
[-m <message text>] the message to log
trap
Send snmp trap from yce server to nms (“1.3.6.1.4.1.696.1.0 - 0”).
Mandatory options:
-e <enterprise> the enterprise OID to use (default 696)
-v <var_oid> the variable OID to use (default 1.0)
-s <specific> the specific IOD to use (default 0)
Optional options:
[-m <message text>] the trap message