Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
docs:guide-developer:procd-init-scripts [2020/09/22 15:01] – [Specifying triggers] Warn about procd_add_reload_interface_trigger() usage rmileckidocs:guide-developer:procd-init-scripts [2024/02/10 16:23] (current) – [Defining service instances] systemcrash
Line 1: Line 1:
-======procd init script parameters ======+====== procd init scripts ======
  
 A procd init script is similiar to an old init script, but with a few differences: A procd init script is similiar to an old init script, but with a few differences:
  
-  * procd expects services to **run in the foreground** +  * procd expects services to start as if they were to  **run in the foreground**, but of course procd runs them the background.
   * Different shebang line: ''#!/bin/sh /etc/rc.common''   * Different shebang line: ''#!/bin/sh /etc/rc.common''
 +  * procd expects that shell variable (not environment variable) ''initscript'' is set to the path of the script that invoked it
   * Explicitly use procd ''USE_PROCD=1''   * Explicitly use procd ''USE_PROCD=1''
  
 Example: Example:
-<code>+<code bash>
 #!/bin/sh /etc/rc.common #!/bin/sh /etc/rc.common
  
Line 29: Line 30:
 ===== Defining service instances ===== ===== Defining service instances =====
  
-For every controlled service ''procd'' stores its current state. State is set in the ''start_service()'' (see next section to see when it's called). Once the new state doesn't match the old one, ''procd'' restarts the service.+The purpose of ''start_service()'' (see next section to see when it's called) is to define instance(s) with:
  
-State is a set of parameters and their values. Some values are set directly in the ''start_service()'' (like ''command'') while some are determined by ''procd'' (like ''file'' and file hash). There are two helpers for setting parameters:+  - Command to execute to start service 
 +  - Information on what to observe for changes (e.g. files or devices) - optional 
 +  - Settings that ''procd'' should use (e.g. auto respawning, logging stdout, user to use) - optional 
 + 
 +The above information is stored by ''procd'' as service instance state. On every relevant system change (e.g. config change), ''start_service()'' is called by designed triggers. If it generates any different state (e.g. command will change) than the previous one, ''procd'' will detect it and restart the service. 
 + 
 +Defining service instance details is handled by setting parameters. Some values are set directly in the ''start_service()'' (like ''command'') while some are determined by ''procd'' (like ''file'' and file hash). There are two helpers for setting parameters:
   - ''procd_set_param()''   - ''procd_set_param()''
   - ''procd_append_param()''   - ''procd_append_param()''
  
-Below example lists supported parameters and describes them. For implementation details see the [[commit>?p=openwrt/openwrt.git;a=blob;f=package/system/procd/files/procd.sh|procd.sh]].+The below example lists supported parameters and describes them. For implementation details see the [[commit>?p=openwrt/openwrt.git;a=blob;f=package/system/procd/files/procd.sh|procd.sh]].
  
-<code>+<code bash>
 start_service() { start_service() {
          procd_open_instance [instance_name]          procd_open_instance [instance_name]
Line 44: Line 51:
  
          # respawn automatically if something died, be careful if you have an alternative process supervisor          # respawn automatically if something died, be careful if you have an alternative process supervisor
-         # if process dies sooner than respawn_threshold, it is considered crashed and after 5 retries the service is stopped+         # if process exits sooner than respawn_threshold, it is considered crashed and after 5 retries the service is stopped 
 +         # if process finishes later than respawn_threshold, it is restarted unconditionally, regardless of error code 
 +         # notice that this is literal respawning of the process, not in a respawn-on-failure sense
          procd_set_param respawn ${respawn_threshold:-3600} ${respawn_timeout:-5} ${respawn_retry:-5}          procd_set_param respawn ${respawn_threshold:-3600} ${respawn_timeout:-5} ${respawn_retry:-5}
  
          procd_set_param env SOME_VARIABLE=funtimes  # pass environment variables to your process          procd_set_param env SOME_VARIABLE=funtimes  # pass environment variables to your process
          procd_set_param limits core="unlimited"  # If you need to set ulimit for your process          procd_set_param limits core="unlimited"  # If you need to set ulimit for your process
-         procd_set_param file /var/etc/your_service.conf # /etc/init.d/your_service reload will restart the daemon if these files have changed +         procd_set_param file /var/etc/your_service.conf # /etc/init.d/your_service reload will restart the daemon when these files have changed 
-         procd_set_param netdev dev # likewise, except if dev's ifindex changes. +         procd_set_param netdev dev # likewise, but for when dev's ifindex changes. 
-         procd_set_param data name=value ... # likewise, except if this data changes.+         procd_set_param data name=value ... # likewise, but for when this data changes.
          procd_set_param stdout 1 # forward stdout of the command to logd          procd_set_param stdout 1 # forward stdout of the command to logd
          procd_set_param stderr 1 # same for stderr          procd_set_param stderr 1 # same for stderr
          procd_set_param user nobody # run service as user nobody          procd_set_param user nobody # run service as user nobody
          procd_set_param pidfile /var/run/somefile.pid # write a pid file on instance start and remove it on stop          procd_set_param pidfile /var/run/somefile.pid # write a pid file on instance start and remove it on stop
 +         procd_set_param term_timeout 60 # wait before sending SIGKILL
          procd_close_instance          procd_close_instance
 } }
Line 62: Line 72:
 ==== Stopping services ==== ==== Stopping services ====
  
-''stop_service()'' is only needed when you need special things to stop your service. ''stop_service()'' is called //after// procd killed the service.+''stop_service()'' is only needed when you need special things to stop your service. ''stop_service()'' is called //before// procd killed the service
 + 
 +If you want to add a check //after// procd has sent the terminate signal (e.g. wait for the process to be really gone), you can define an extra function ''service_stopped()''.
  
 ==== Init scripts during compilation ==== ==== Init scripts during compilation ====
  
-{{:meta:icons:tango:48px-dialog-warning.svg.png?32}}WARNING{{:meta:icons:tango:48px-dialog-warning.svg.png?32}}: initscripts **will run** while building OpenWRT images (when installing packages in what will become a ROM image) in the **host system** (right now, for actions "//enable//" and "//disable//").  **They must not fail, or have undesired side-effects in that situation.**  When being run by the build system, environment variable **${IPKG_INSTROOT}** will be set to the working directory being used.  On the "target system", that environment variable will be empty/unset.  Refer to "/lib/functions.sh" and also to "/etc/rc.common" in package "base-files" for the nasty details.+{{:meta:icons:tango:48px-dialog-warning.svg.png?32}}WARNING{{:meta:icons:tango:48px-dialog-warning.svg.png?32}}: initscripts **will run** while building OpenWrt images (when installing packages in what will become a ROM image) in the **host system** (right now, for actions "//enable//" and "//disable//").  **They must not fail, or have undesired side-effects in that situation.**  When being run by the build system, environment variable **${IPKG_INSTROOT}** will be set to the working directory being used.  On the "target system", that environment variable will be empty/unset.  Refer to "/lib/functions.sh" and also to "/etc/rc.common" in package "base-files" for the nasty details.
  
 ===== Specifying triggers ===== ===== Specifying triggers =====
Line 72: Line 84:
 While ''start_service()'' takes care of setting service instances states and submitting them to the ''procd'' (for a potential service restart), it has to be explicitly called to do so. In most cases it should happen on some related change. While ''start_service()'' takes care of setting service instances states and submitting them to the ''procd'' (for a potential service restart), it has to be explicitly called to do so. In most cases it should happen on some related change.
  
-That's where ''service_triggers()'' (that allows specifying triggers) comes in handy. Most system important changes result in generating events that ''service_triggers()'' can use for triggering various actions. There are multiple ''procd_add_*_trigger()'' helpers for that purpose.+That's where ''service_triggers()'' comes in handy and allows specifying triggers. Most system important changes result in generating events that ''service_triggers()'' can use for triggering various actions. There are multiple ''procd_add_*_trigger()'' helpers for that purpose.
  
-Every configurable service has to specify what changes should result in its reconfiguration. They should be defined in the `service_triggers()using available helpers. When related ''procd'' ''service'' event occurs it will result in executing  which by default executes ''start_service()''.+Every configurable service has to specify what system changes should result in its reconfiguration. Those events should be defined in the ''service_triggers()'' using available helpers. When related ''procd'' ''service'' event occurs it will result in executing ''/etc/init.d/<foo> reload''.
  
 Example: Example:
  
-<code>+<code bash>
 service_triggers() service_triggers()
 { {
         procd_add_reload_trigger "<uci-file-name>" "<second-uci-file>"         procd_add_reload_trigger "<uci-file-name>" "<second-uci-file>"
         procd_add_reload_interface_trigger <interface>         procd_add_reload_interface_trigger <interface>
 +        procd_add_reload_mount_trigger <path> [<path> ...]
 } }
 </code> </code>
Line 89: Line 102:
 | procd_add_reload_trigger           | list of config files | ''config.change'' | Uses ''/etc/init.d/<foo> reload'' as the handler | | procd_add_reload_trigger           | list of config files | ''config.change'' | Uses ''/etc/init.d/<foo> reload'' as the handler |
 | procd_add_reload_interface_trigger | interface name       | ''interface.*''   | Uses ''/etc/init.d/<foo> reload'' as the handler | | procd_add_reload_interface_trigger | interface name       | ''interface.*''   | Uses ''/etc/init.d/<foo> reload'' as the handler |
 +| procd_add_reload_mount_trigger     | paths to watch for   | ''mount.add''     | Uses ''/etc/init.d/<foo> reload'' as the handler |
 +| procd_add_restart_mount_trigger    | paths to watch for   | ''mount.add''     | Uses ''/etc/init.d/<foo> restart'' as the handler |
  
 <WRAP important> <WRAP important>
Line 97: Line 112:
  
 <WRAP important> <WRAP important>
-Subscribing to the ''interface.*'' event and having ''/etc/init.d/<foo> reload'' called won't automatically make ''procd'' restart service.+Adding ''interface.*'' trigger and having ''/etc/init.d/<foo> reload'' called won't automatically make ''procd'' notice any state change and won't make it restart service.
  
 Relevant interface has to be made part of service state using the ''procd_set_param netdev''. Relevant interface has to be made part of service state using the ''procd_set_param netdev''.
 </WRAP> </WRAP>
  
-See use cases of [[https://github.com/openwrt/packages/search?q=procd_add_interface_trigger|procd_add_interface_trigger]] and [[https://github.com/openwrt/packages/search?q=procd_add_reload_trigger|procd_add_reload_trigger]] in the OpenWrt packages repository.+<WRAP important> 
 +Using mount triggers depends on mount notifications emitted by ''blockd''. Hence ''blockd'' needs to be installed and the mount need to be configured in ''/etc/config/fstab''
 + 
 +See also [[:docs:guide-user:storage:fstab]] 
 +</WRAP> 
 + 
 +See use cases of [[https://github.com/openwrt/packages/search?q=procd_add_interface_trigger|procd_add_interface_trigger]][[https://github.com/openwrt/packages/search?q=procd_add_reload_trigger|procd_add_reload_trigger]], [[https://github.com/openwrt/packages/search?q=procd_add_reload_mount_trigger|procd_add_reload_mount_trigger]] in the OpenWrt packages repository.
  
 ==== ucitrack ==== ==== ucitrack ====
  
 In older versions of OpenWrt, a system called "ucitrack" attempted to track UCI config files, and the processes that depended on each of them, and would restart them all as needed.  This too, is replaced with ubus/procd, and expanded to allow notifying services when network interfaces change. In older versions of OpenWrt, a system called "ucitrack" attempted to track UCI config files, and the processes that depended on each of them, and would restart them all as needed.  This too, is replaced with ubus/procd, and expanded to allow notifying services when network interfaces change.
 +
 +===== Manual reload =====
 +
 +Sometimes service state may depend on information that doesn't have any events related. This may happen e.g. with service native configuration files that don't get build using UCI config.
 +
 +In such cases ''procd'' should be told to use relevant config file using ''procd_set_param file /etc/foo.conf''. After every config file modification ''/etc/init.d/foo reload'' should be called manually.
  
 ===== Custom service reload ===== ===== Custom service reload =====
Line 116: Line 143:
 If some service requires a restart when ''reload'' is called, it can be implemented as follows: If some service requires a restart when ''reload'' is called, it can be implemented as follows:
  
-<code>+<code bash>
 reload_service() reload_service()
 { {
Line 125: Line 152:
 </code> </code>
  
-==== Reloading service setup ====+j==== Reloading service setup ====
  
 Some services may support reloading configuration without a complete restart. It's usually implemented using `SIGHUP` or similar signal. Some services may support reloading configuration without a complete restart. It's usually implemented using `SIGHUP` or similar signal.
Line 149: Line 176:
 procd_send_signal service_name [instance_name] [signal] procd_send_signal service_name [instance_name] [signal]
 </code> </code>
 +
 +<WRAP important>
 +You can also configure reload by signal with ''procd_set_param reload_signal'' service option.
 +</WRAP>
 +===== Service jails =====
 +procd can isolate services using various Linux features typically used for (slim-)containers: //chroot// and //namespaces// (and //limits//, //seccomp//, //capabilities// as well as setting ''PR_SET_NO_NEW_PRIVS'', see [[#service_parameters|Service Parameters]]).
 +
 +^ Function                ^ Arguments            ^ Description                                                        ^
 +| procd_add_jail          | jail name, flags     | Set up service jail (with features according to //flags//         |
 +| procd_add_jail_mount    | read-only paths      | Read-only bind the paths listed to the jail's mount namespace      |
 +| procd_add_jail_mount_rw | read-write paths     | Bind the paths listed to the jail's mount namespace                |
 +
 +^ Flag        ^ Description                                                      ^
 +| log         | Allow jailed service to log to syslog                            |
 +| ubus        | Allow jailed service to access ubus                              |
 +| procfs      | Mount /proc in jail                                              |
 +| sysfs       | Mount /sys in jail                                               |
 +| ronly       | Re-mount jail rootfs read-only                                   |
 +| requirejail | Do not fall back to run without jail in case jail could not be set up |
 +| netns       | Run jailed process in new network namespace                      |
 +| userns      | Run jailed process in new user namespace                         |
 +| cgroupsns   | Run jailed process in new cgroups namespace                      |
 +| console     | Set up console accessible with ''ujail-console''                 |
 +
 +See use cases of [[https://github.com/openwrt/packages/search?q=procd_add_jail|procd_add_jail]].
  
 ===== Debugging ===== ===== Debugging =====
 +
 Set PROCD_DEBUG=1 to see debugging information when starting or stopping a procd init script. Set PROCD_DEBUG=1 to see debugging information when starting or stopping a procd init script.
 Also, ''INIT_TRACE=1 /etc/init.d/mything $action'' Where $action is start/stop etc. Also, ''INIT_TRACE=1 /etc/init.d/mything $action'' Where $action is start/stop etc.
 +
 ===== Examples ===== ===== Examples =====
  
Line 168: Line 222:
 | ''env'' | Key-Value-List | Sets a number of environment variables in ''key=value'' notation exported to the spawned process. | | ''env'' | Key-Value-List | Sets a number of environment variables in ''key=value'' notation exported to the spawned process. |
 | ''data'' | Key-Value-List | Sets arbitrary user data in ''key=value'' notation to the ubus service state. This is mainly used to store additional meta data with spawned services, such as mDNS announcements or firewall rules which may be picked up by other services. | | ''data'' | Key-Value-List | Sets arbitrary user data in ''key=value'' notation to the ubus service state. This is mainly used to store additional meta data with spawned services, such as mDNS announcements or firewall rules which may be picked up by other services. |
-| ''limits'' | Key-Value-List | Set ulimit values in ''key=value'' notation for the spawned process. The following limit names are recognized by //procd//: ''as'' (''RLIMIT_AS''), ''core'' (''RLIMIT_CORE''), ''cpu'' (''RLIMIT_CPU''), ''data'' (''RLIMIT_DATA''), ''fsize'' (''RLIMIT_FSIZE''), ''memlock'' (''RLIMIT_MEMLOCK''), ''nofile'' (''RLIMIT_NOFILE''), ''nproc'' (''RLIMIT_NPROC''), ''rss'' (''RLIMIT_RSS''), ''stack'' (''RLIMIT_STACK''), ''nice'' (''RLIMIT_NICE''), ''rtprio'' (''RLIMIT_RTPRIO''), ''msgqueue'' (''RLIMIT_MSGQUEUE''), ''sigpending'' (''RLIMIT_SIGPENDING'') |+| ''limits'' | Key-Value-List | Set ulimit values in ''key=value'' notation for the spawned process. The following limit names are recognized by //procd//: ''as'' (''RLIMIT_AS''), ''core'' (''RLIMIT_CORE''), ''cpu'' (''RLIMIT_CPU''), ''data'' (''RLIMIT_DATA''), ''fsize'' (''RLIMIT_FSIZE''), ''memlock'' (''RLIMIT_MEMLOCK''), ''nofile'' (''RLIMIT_NOFILE''), ''nproc'' (''RLIMIT_NPROC''), ''rss'' (''RLIMIT_RSS''), ''stack'' (''RLIMIT_STACK''), ''nice'' (''RLIMIT_NICE''), ''rtprio'' (''RLIMIT_RTPRIO''), ''msgqueue'' (''RLIMIT_MSGQUEUE''), ''sigpending'' (''RLIMIT_SIGPENDING''**Two numeric values, separated by blank space, are expected for RLIMIT: the first value represents the soft limit and the other the hard limit; e.g.: procd_set_param limits nofile="10000 20000"; the "unlimited" value can be used in cases where "ulimit -{parameter} unlimited" works, for example for the "core" parameter.**|
 | ''command'' | List | Sets the command vector (''argv'') used to execute the process. | | ''command'' | List | Sets the command vector (''argv'') used to execute the process. |
 | ''netdev'' | List | Passes a list of Linux network device names to //procd// to be monitored for changes. Upon starting a service, the interface index of each network device name is resolved and stored as part of //procd//'s in-memory service state. When a service reload request is processed and the interface index of any of the associated network devices changed or if the list itself changed, the running service state is invalidated and //procd// will restart the associated process or deliver a UNIX signal to it, depending on how the service was set up. | | ''netdev'' | List | Passes a list of Linux network device names to //procd// to be monitored for changes. Upon starting a service, the interface index of each network device name is resolved and stored as part of //procd//'s in-memory service state. When a service reload request is processed and the interface index of any of the associated network devices changed or if the list itself changed, the running service state is invalidated and //procd// will restart the associated process or deliver a UNIX signal to it, depending on how the service was set up. |
Line 180: Line 234:
 | ''pidfile'' | String | Instructs //procd// to write the PID of the spawned process into the specified file path. While //procd// itself does not use or require PID files to track spawned processes, this option is useful for sitation where knowledge of the PID is required, e.g. for monitoring or control client software. | | ''pidfile'' | String | Instructs //procd// to write the PID of the spawned process into the specified file path. While //procd// itself does not use or require PID files to track spawned processes, this option is useful for sitation where knowledge of the PID is required, e.g. for monitoring or control client software. |
 | ''user'' | String | Specifies the name of the user to spawn the process as. //procd// will look up the given name in ''/etc/passwd'' and set the effective uid and primary gid of the spawned processaccordingly. If omitted, the process is spawned as ''root'' (uid 0, gid 0) | | ''user'' | String | Specifies the name of the user to spawn the process as. //procd// will look up the given name in ''/etc/passwd'' and set the effective uid and primary gid of the spawned processaccordingly. If omitted, the process is spawned as ''root'' (uid 0, gid 0) |
-| ''seccomp'' | String | Specifies the ''seccomp'' list for //ujail// (''ujail -S''+| ''seccomp'' | String | Specifies file path to read seccomp filter rules from, the file should be JSON formatted like the [[https://github.com/opencontainers/runtime-spec/blob/master/config-linux.md#seccomp|seccomp object of the OCI run-time spec]] 
-| ''capabilities'' | String | This parameter is currently unused. |+| ''capabilities'' | String | Specifies file path to read capability set, the file should be JSON formatted like the [[https://github.com/opencontainers/runtime-spec/blob/master/config.md#linux-process|capabilities object of the OCI run-time spec]]|
 | ''stdout'' | Boolean | If set to ''1'', instruct //procd// to relay the spawned process' stdout to the system log. The stdout will be fed line-wise to ''syslog(3)'' using the basename of the first command argument as identity, ''LOG_INFO'' as priority and ''LOG_DAEMON'' as facility. | | ''stdout'' | Boolean | If set to ''1'', instruct //procd// to relay the spawned process' stdout to the system log. The stdout will be fed line-wise to ''syslog(3)'' using the basename of the first command argument as identity, ''LOG_INFO'' as priority and ''LOG_DAEMON'' as facility. |
 | ''stderr'' | Boolean | If set to ''1'', instruct //procd// to relay the spawned process' stderr to the system log. The stderr will be fed line-wise to ''syslog(3)'' using the basename of the first command argument as identity, ''LOG_ERR'' as priority and ''LOG_DAEMON'' as facility. | | ''stderr'' | Boolean | If set to ''1'', instruct //procd// to relay the spawned process' stderr to the system log. The stderr will be fed line-wise to ''syslog(3)'' using the basename of the first command argument as identity, ''LOG_ERR'' as priority and ''LOG_DAEMON'' as facility. |
 | ''no_new_privs'' | Boolean | Instructs //ujail// to not allow privilege elevation. Sets the //ujail// ''-c'' parameter when true. | | ''no_new_privs'' | Boolean | Instructs //ujail// to not allow privilege elevation. Sets the //ujail// ''-c'' parameter when true. |
  • Last modified: 2020/09/22 15:01
  • by rmilecki