Driver Support Document
| SYSOID Mapping | ||
| SYSOID | MODEL | OS VERSION |
| 1.3.6.1.4.1.15497.1.1 | C170, C370, C670, C680, M680, M690 | 7.1.4, 7.5.2, 8.5.6, 9.0 |
| 1.3.6.1.4.1.15497.1.2 | S160, S100V, S300V, S360, S390, S660, S670, S680, S690, M1070 | 7.0.0, 7.5.0, 7.5, 8.0.8, 8.8.0, 9.0 |
Ironport devices support CLI methods for capturing the configuration, as well as Transport methods [SCP, FTP, etc]. Due to an apparent device bug, occational carriage returns are missing from the CLI-formatted data, causing sequential XML elements to be elided into a single line. Parsing and deployment of the configs is unaffected, but spurious configuration changes can be seen if NA switches between the CLI and transport snapshot methods.
Ironport devices use the 'saveconfig' command to store the running configuration to a file, to be transported by SCP or FTP. Some devices may prompt requesting that 'cluster' mode be entered, and if answered 'no' the file will not be saved. To allow cluster mode to be entered, set the device access variable 'allow_cluster' to 'true'. If cluster mode is required but the variable is not set, transport protocol methods for snapshot will be skipped.
This note applies only for drivers that use HTTP requests for driver functions. HTTP proxy operations are supported by setting the device access variable "http_proxy" to "ip:port", replacing with the IP and port values of the proxy server. SNI-requiring devices (e.g CloudGenix & Cisco Meraki) can be supported by using the device access variable "alternate_host" to contain the DNS name of the host. The host name will be used rather than the normal management IP address for all HTTP requests, effectively supporting SNI.
Discovery tasks for Javascript drivers handle More prompts by using timeouts, which can cause problems with the third-party SSH client code, which interprets the timeout as a disconnection. There are two options to work around the problem. Setting the RCX option [<option name="Driver/Discovery/UsePollRead">true</option>] in site_options.rcx will effect the workaround for all affected devices. Alternatively, it could be applied to a single device by setting the device access variable "PollRead" to "true".
Discovery tasks for Javascript drivers use wakeup characters are sent during device connection, to ensure that the device is responding. Normally, these characters do not echo to the console, but some devices may echo them. In this case, this causes the prompt detection phase to fail, which in turn can cause More prompts to not be handled properly, and discovery may fail. If these characters are echoed from the device [check the session log to see this], then set the device access variable "skip_ctrl_u" to skip the sending of the wakeup characters. Note that setting this option on a previously working device could cause discovery tasks to fail, but it only affects CLI discovery. SNMP discovery is unaffected.