Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Release Notes Summary: Information about actions run from a plan are now logged at the notice level which is visible by default. This provides good default logging for progress during a plan. Release Notes: New Feature Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Michelle Fredette commented on BOLT-266 Re: Log run_* actions when called from a plan added a release note. Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Michelle Fredette updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Michelle Fredette Labels: docs_reviewed Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer created an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Issue Type: Improvement Assignee: Unassigned Created: 2017/12/18 12:09 PM Priority: Normal Reporter: Alex Dreyer Currently are no log messages at the default log level when a plan is run. Constraints: In addition to the stdout stream we have 5 log levels to work with in bolt DEBUG INFO WARN(default) ERROR FATAL(used before exiting) Loggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication WARN prefacing normal messages may be concerning to users. Current state: DEBUG: log connection information
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Currently are no log messages at the default log level when a plan is run. h3. Constraints: In addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting) Loggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplicationWARN prefacing normal messages may be concerning to users. h3. Current state: State In addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG - INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG :- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on node h3. Proposed changes: This involves adding a new log level notice and making it the default. DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failed Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. h3. Constraints:Loggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplicationWARN prefacing normal messages may be concerning to users.h3. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on nodeh3. Proposed changes: This involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failed Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. h3 h2 . Constraints : Loggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication . WARN prefacing normal messages may be concerning to users. h3 h2 . Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on node h3 h2 . Proposed changes : This involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failed Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubsc
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failed h2. Out of scope:- logging node failures by default. This would result in different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db)
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failedh2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db)
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback. h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failedh2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-s
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Michael Smith updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Michael Smith Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us) INFO:- Log when bolt starts run execution with object type and first argument(command, script, task or file)- Log when actions start on nodes- Log the results of each action on each node NOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failedh2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when bolt starts run execution with object type and first argument(command, script, task or file)- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is- log when a run_* call finishes and summarize how many nodes succeeded or failedh2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Jenny Mahmoudi commented on BOLT-266 Re: Log run_* actions when called from a plan Diagram of bolt output types https://docs.google.com/a/puppet.com/drawings/d/13YgP5aS0Ux3eaFvzr8rwOhYEusPFyEZjOqLtQV3Hrow/edit?usp=sharing Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Sprint: Bolt Irving One Point Oh Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a nodeINFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when bolt starts run execution with object type and first argument(command, script, task or file) {{ Starting bolt command run `echo 'hello world'` }} - Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is {{ Running task package }} - log when a run_* call finishes and summarize how many nodes succeeded or failed {{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Sprint: Bolt One Point Oh Irving Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Sprint: Bolt Irving One Point Oh Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Jenny Mahmoudi commented on BOLT-266 Re: Log run_* actions when called from a plan Alex Dreyer The proposed new notice level makes sense to me. I'm confused about how the info and debug levels are changing, though. 1. It looks like "log when actions are started on nodes" is duplicated in both info and debug - is that intentional? 2. I can't tell for sure what's the same between the current state and proposed changes because some items sound similar but are written with different language. Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node (as json result) INFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- log subactions- log connection information(This is controlled by net:ssh not us)INFO:- Log when bolt starts run execution with object type and first argument(command, script, task or file){{ Starting bolt command run `echo 'hello world'` }}- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node (as json result)INFO:- log whenever an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- - log subactions every sub-action taken on a node like changing a task file permissions - log connection information(This is controlled by net:ssh not us)INFO:- Log when bolt starts run execution with object type and first argument(command, script, task or file){{ Starting bolt command run `echo 'hello world'` }}- Log when actions start on nodes- Log the results of each action on each nodeNOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node (as json result) INFO:- log whenever when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log when actions are started on nodes- - log every sub-action taken on a node like changing a task file permissions - log connection information (This is controlled by net:ssh not us) INFO:- Log when bolt starts run execution with object type and first argument(command, script, task or file){{ code}} Starting bolt command run `echo 'hello world'` Starting bolt task run package{{code }}- Log when actions start on nodes- Log the results of from each action on each a node (as json representation) NOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- Log when bolt starts run execution with object type and first argument(command, script, task or file){{code}}Starting bolt command run `echo 'hello world'`Starting bolt task run package {{code}}- Log when actions start on nodes- Log results from each action on a node (as json representation)NOTICE:- log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{code}}Starting bolt command run `echo 'hello world'`Starting bolt task run package{{code}}- Log when actions start on nodes- (previously at debug) Log results from each action on a node (as json representation)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{ code noformat }}Starting bolt command run `echo 'hello world'`Starting bolt task run package{{ code noformat }}- Log when actions start on nodes- (previously at debug) Log results from each action on a node (as json representation)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Alex Dreyer updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Alex Dreyer
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{ noformat}} Starting bolt command run `echo 'hello world'` $COMMAND}} {{ Starting bolt task run package {{noformat }}- Log when actions start on nodes- (previously at debug) Log results from each action on a node (as json representation)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Sprint: Bolt One Point Oh Kanban Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Lucy Wyman assigned an issue to Lucy Wyman Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Lucy Wyman Assignee: Lucy Wyman Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Lucy Wyman assigned an issue to Unassigned Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Lucy Wyman Assignee: Lucy Wyman Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Lucy Wyman assigned an issue to Lucy Wyman Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Lucy Wyman Assignee: Lucy Wyman Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}}- Log when actions start on nodes- (previously at debug) Log results from each action on a node (as json representation) (/) NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions (/) - log connection information (/) INFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}}- Log when actions start on nodes (/) - (previously at debug) Log results from each action on a node (as json representation) (/)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Lucy Wyman commented on BOLT-266 Re: Log run_* actions when called from a plan Alex Dreyer I'm not sure how "Log when bolt starts run execution with object type and first argument(command, script, task or file)" is different from "Log when actions start on nodes". When the action starts it logs Starting bolt task run '/home/lucy/githubs/pe-console-ui/acceptance/files/modules/exec/tasks/init' (https://github.com/puppetlabs/bolt/blob/master/lib/bolt/node/ssh.rb#L281). What's being added? Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions (/) - log connection information (/) INFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}}- Log when actions start on nodes (/) - (previously at debug) Log results from each action on a node (as json representation) (/)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}}- Log when actions start on nodes- (previously at debug) Log results from each action on a node (as json representation) (/)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{ Running task package }}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{ Ran task package on 34 nodes with 3 failures }} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer commented on BOLT-266 Re: Log run_* actions when called from a plan the first one is about logging from the CLI level when bolt is called. The second one should probably drop the full path and just use the task name so it aligns with the CLI level logging and run_task logging. Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}} (/) - Log when actions start on nodes- (previously at debug) Log results from each action on a node (as json representation) (/)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{Running task package}}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{Ran task package on 34 nodes with 3 failures}} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Lucy Wyman assigned an issue to Unassigned Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Lucy Wyman Assignee: Lucy Wyman Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}} (/) - Log when actions start on nodes- (previously at debug) Log results from each action on a node (as json representation) (/) NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{Running task package}}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{Ran task package on 34 nodes with 3 failures}} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}}- Log when actions start on nodes (happens in the {{transport.rb}}? - (previously at debug) Log results from each action on a node (as json representation)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{Running task package}}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{Ran task package on 34 nodes with 3 failures}} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}}- Log when actions start on nodes (happens in the {{transport.rb}}? ) - (previously at debug) Log results from each action on a node (as json representation)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{Running task package}}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{Ran task package on 34 nodes with 3 failures}} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}} (/) - Log when actions start on nodes (happens in the {{transport.rb}}?)- (previously at debug) Log results from each action on a node (as json representation)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{Running task package}}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{Ran task package on 34 nodes with 3 failures}} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title
Lucy Wyman updated an issue
Puppet Task Runner / BOLT-266
Log run_* actions when called from a plan
Change By:
Lucy Wyman
Currently are no log messages at the default log level when a plan is run. As a first iteration in improving this problem we can add some log messages to run_* functions in bolt. It's likely that as we improve the output some of this logging will change but doing it now allows us to get quicker feedback.h2. ConstraintsLoggers will likely work the same for both run_task and run_plan. Since run_task prints a lot of information to stdout we need to avoid duplication. WARN prefacing normal messages may be concerning to users.h2. Current StateIn addition to the stdout stream we have 5 log levels to work with in bolt- DEBUG- INFO- WARN(default)- ERROR- FATAL(used before exiting)DEBUG:- log connection information- log every sub-action taken on a node like changing a task file permissions- log results from each action on a node INFO:- log when an action is started on nodeh2. Proposed changesThis involves adding a new log level notice and making it the default.DEBUG:- log every sub-action taken on a node like changing a task file permissions- log connection informationINFO:- (new) Log when bolt starts run execution with object type and first argument(command, script, task or file){{Starting bolt command run $COMMAND}}{{Starting bolt task run package}} (/) - Log when actions start on nodes (happens in the {{transport.rb}}?)- (previously at debug) Log results from each action on a node (as json representation)NOTICE:- (new) log when a run_* call is starting and what the first argument(command, script, task, or file) is{{Running task package}}- (new) log when a run_* call finishes and summarize how many nodes succeeded or failed{{Ran task package on 34 nodes with 3 failures}} h2. Out of scope:- logging node failures by default. This would result in duplicate output or different logging inside the executor depending on what mode it was running in. It may make sense longer term as part of default output- specifying the warn log level on the commandline. It can be specified in the config file. We could also remove the --debug and --verbose CLI flags and replace them with --loglevel = error|warn|notice|info|debug
Add Comment
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer assigned an issue to Alex Dreyer Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Assignee: Alex Dreyer Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Michael Smith assigned an issue to Unassigned Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Michael Smith Assignee: Alex Dreyer Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Alex Dreyer updated an issue Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Alex Dreyer Fix Version/s: BOLT 0.14.0 Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
Jira (BOLT-266) Log run_* actions when called from a plan
Title: Message Title Zach Reichert assigned an issue to Zach Reichert Puppet Task Runner / BOLT-266 Log run_* actions when called from a plan Change By: Zach Reichert Assignee: Zach Reichert Add Comment This message was sent by Atlassian JIRA (v7.0.2#70111-sha1:88534db) -- You received this message because you are subscribed to the Google Groups "Puppet Bugs" group. To unsubscribe from this group and stop receiving emails from it, send an email to puppet-bugs+unsubscr...@googlegroups.com. To post to this group, send email to puppet-bugs@googlegroups.com. Visit this group at https://groups.google.com/group/puppet-bugs. For more options, visit https://groups.google.com/d/optout.
