Tags: trigger/trigger
Tags
Fixup of rc5 bugs (#307) * Fixup of rc5 bugs * TextFSM processor bug which made it so only one set of parsed commands made it into the results. * Twister bug which regressed support for less than average ssh server implementations by not initialising keyexchange * Made some modifications after observing a different kind of failure with the last revision. Looks like kexinit is for the server side of things. * Applied fix to twister2 which is also affected * Fix import bug. * Update changelog and version control
Release 1.5.10 (#281) Bug Fixes --------- - Extended prompt detection for IOS-like devices to include interstitial space (`\s`) or carriage return (`\r`) characters which is sometimes seen on Arista EOS devices, and would cause asynchronous execution to sometimes hang and result in a `~trigger.exceptions.CommandTimeout` error. - Bugfix in `bin/load_acl` that prevents `queue.complete()` from being called when using the `--no-db` flag. Previously, an `AttributeError` attribute error was raised due to attempting to call `complete` on `queue`, which is set to `None` when passing `--no-db`. - Cleaned up network.utils.ping, removed hostname validation code - Bugfix in `~trigger.utils.network.ping()` where a file descriptor wasn't closed cleanly. - Bugfix in `bin/gnng` that printed device names before any tables, resulting in potentially confusing results. Devices names are now printed with the corresponding table. - Bugfix in `bin/gnng` that allows the `--filter-on-type` to function as expected. - Update documentation of `gnng`'s `-N`/`--nonprod` flag. - Bugfix in `bin/gnng` that allows `gnng` to fail gracefully when a device isn't found. - Bugfix in `bin/gnng --all` that was causing many device vendors to be skipped entirely because the filter was too specific. This vendor filter has been removed and will now fallback to `~trigger.cmds.NetACLInfo()` internal knowledge of supported platforms.
Release 1.5.9 Bug Fixes --------- + Bugfix in `~trigger.gorc` where init commands would be sent before a prompt was even available on the remote device. + Bugfix in `~trigger.utils.core.pretty_time()` where `pytz` was being referenced but not imported. + Extended prompt detection for IOS-like devices to include backspace characters (`\b` or `\x08`) which is sometimes seen on Cisco NX-OS devices, and would cause asynchronous execution to sometimes hang and result in a `~trigger.exceptions.CommandTimeout` error. * Improved the internal grouping logic for `load_acl` to be more permissive and if grouping fails it will just not group devices. * Fixed a bug that would prevent ACL staging from working when using default global settings. * Fixed bugs in the default global callables for `get_current_oncall()` and `create_tm_ticket()` that would prevent `lod_acl` from working. They now default to a disabled state that does not require customization just to utilize core load_acl functionality. * Updated the sample `settings.py` (`conf/trigger_settings.py`) to utilize the updated default callables. * Fixed a bug in default global callable for `get_tftp_source()` to properly perform lookup of `settings.VIPS` * Fixed a bug in default global callable for `stage_acls()` to properly perform lookup of `settings.FIREWALL_DIR` and `settings.TFTPROOT_DIR`.
Release 1.5.7 Enhancements ------------ - Added TextFSM parser to process unstructured CLI output. - Added a new prompt pattern to `settings.CONTINUE_PROMPTS`. - New continue prompts no longer need to be lower-cased. - Clarified the error text when an enable password is required but not provided when connecting to a device to make it a little more clear on how to proceed. Bug Fixes --------- - Bugfix in `~trigger.contrib.commando.plugins.config_device` causing an unhandled `NameError`. - Bugfix in `bin/gnng` that would cause a crash when using `--sqldb` or `--csv` output flags.
Release 1.5.6 Bug Fixes --------- * Added `-f/--force-cli` to `run_cmds` to allow CLI execution on Juniper devices or any vendor platform where API support is enabled by default, fixing an underlying bug where CLI output would result in a crash. * Multiple commands can now be sent to Juniper devices w/ `run_cmds`. * Updated the Juniper CLI prompt pattern to work when a hostname isn't set that would result in a `CommandTimeout` causing execution to fail.
PreviousNext