Introduce `sdcommands` option and `shutdown.default` INSTCMD to all drivers #2686

jimklimov · 2024-11-19T21:44:32Z

Closes: #2670

Started from discussion in the issue linked above, about drivers hard-coding some operations as their late-shutdown implementation -- which may not be the best fit for a particular supported device model.

This PR aims to allow one to configure (a generally comma-separated list of) sdcommands option with arbitrary INSTCMD values supported by the device-specific part of driver code (if any). In testing e.g. beeper commands can be used instead of powering off the load.

It also aliases the call to upsdrv_shutdown() as a shutdown.default for good measure (similar to driver.killpower but without a failsafe flag, and not necessarily exiting the driver).

Shutdown implementations of some drivers got reshuffled into newly supported standard instant commands like shutdown.return or shutdown.stayoff - sometimes guessing the name (testing wanted). For others, the method code remains unique (e.g. when return/stayoff is chosen based on OL/OB status).

Signed-off-by: Jim Klimov <[email protected]>

…t all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…rom upsdrv_shutdown() to upscmd("shutdown.stayoff") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…etworkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…er) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…ser) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…etworkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…rv_shutdown() to upscmd("shutdown.return") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…r); revise warning message [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…sdrv_shutdown() to upscmd("shutdown.return"); newly support INSTCMD at all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…m upsdrv_shutdown() to upscmd("shutdown.return") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…m upsdrv_shutdown() to upscmd("shutdown.return"); newly support INSTCMD at all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…er) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…etworkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…r) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…rv_shutdown() to upscmd("shutdown.return") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…upsdrv_shutdown() to upscmd("shutdown.return"); newly support INSTCMD at all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

Signed-off-by: Jim Klimov <[email protected]>

…zed by user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…rom upsdrv_shutdown() to upscmd("shutdown.stayoff") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…er) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…y user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…() so drivers might implement their own "shutdown.default" at will [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…wn_commands() [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

Signed-off-by: Jim Klimov <[email protected]>

…lt" implementations again [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…tinfo() where we register INSTCMD handlers [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

…fallback() so drivers might implement their own "shutdown.default" at will [networkupstools#2670]" This reverts commit 25250b0: too messy in logs where driver.instcmd() think they are last in stack. Signed-off-by: Jim Klimov <[email protected]>

…" as upsdrv_shutdown() in main loop too [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

AppVeyorBot · 2024-12-01T01:13:28Z

✅ Build nut 2.8.2.2454-master completed (commit c44c15765c by @jimklimov)

jimklimov · 2024-12-02T11:01:23Z

Per #2670 (comment) real-life testing succeeded with usbhid-ups and 3 different APC UPS:

SUA750RMI1U (older model);
SMT1500RMI1U and SMT1500RMI2U (newer models).

jimklimov · 2024-12-02T11:46:51Z

Also checked with make check-NIT-sandbox-devel &, using upscmd against the dummy driver (tool logs and driver logs landing in same terminal):

:; ./clients/upscmd -u admin -p "$TESTPASS_ADMIN" -w "UPS1@localhost:$NUT_PORT" shutdown.default
Instant command: [email protected] did shutdown.default on UPS1 (tracking ID: 2DDC5036-F665-4D29-BE4C-0374D091D263)
instcmd: unknown command [shutdown.default] [(null)]
shutdown not supported
tracking ID: 2DDC5036-F665-4D29-BE4C-0374D091D263       result: SUCCESS
SUCCESS

Note the drivers do not exit now due to INSTCMD alone (because if (handling_upsdrv_shutdown > 0) set_exit_flag(EF_EXIT_FAILURE);):

:; ./clients/upscmd -u admin -p "$TESTPASS_ADMIN" -l "UPS1@localhost:$NUT_PORT"
Instant commands supported on UPS [UPS1]:

driver.killpower - Description unavailable
driver.reload - Description unavailable
driver.reload-or-error - Description unavailable
driver.reload-or-exit - Description unavailable
load.off - Description unavailable
shutdown.default - Description unavailable

It does however set the flag and exit when we "kill power" using drivername -k (which uses a different INSTCMD driver.killpower that toggles handling_upsdrv_shutdown=1 and calls same upsdrv_shutdown() ultimately):

:; ./drivers/dummy-ups -a UPS1 -DDDDDD -k
   0.000002     [D5:61105] send_to_all: SETINFO driver.state "init.starting"
Network UPS Tools 2.8.2.1303.84-1387-g019529e68 (development iteration after 2.8.2) - Device simulation and repeater driver 0.20
   0.000075     [D5:61105] do_upsconf_args: confupsname=(null), var=maxretry, val=3
   0.000101     [D5:61105] do_upsconf_args: call do_global_args()
   0.000107     [D3:61105] do_global_args: var='maxretry' val='3'
   0.000112     [D5:61105] do_upsconf_args: confupsname=(null), var=driverpath, val=/home/jim/nut/drivers
   0.000118     [D5:61105] do_upsconf_args: call do_global_args()
   0.000121     [D3:61105] do_global_args: var='driverpath' val='/home/jim/nut/drivers'
   0.000124     [D5:61105] do_upsconf_args: confupsname=dummy, var=driver, val=dummy-ups
   0.000128     [D5:61105] do_upsconf_args: confupsname=dummy, var=desc, val=Crash Dummy
   0.000131     [D5:61105] do_upsconf_args: confupsname=dummy, var=port, val=dummy.seq
   0.000151     [D5:61105] do_upsconf_args: confupsname=UPS1, var=driver, val=dummy-ups
   0.000155     [D5:61105] do_upsconf_args: call main_arg()
   0.000177     [D3:61105] main_arg: var='driver' val='dummy-ups'
   0.000198     [D5:61105] do_upsconf_args: not a main_arg()
   0.000218     [D5:61105] do_upsconf_args: this is a 'driver' setting, may we proceed?
   0.000224     [D6:61105] testval_reloadable: var=driver, oldval=dummy-ups, newval=dummy-ups, reloadable=0, reload_flag=0
   0.000226     [D6:61105] testval_reloadable: verdict for (re)loading var=driver value: -1
   0.000247     [D5:61105] do_upsconf_args: 'driver' setting already applied with this value
   0.000255     [D5:61105] do_upsconf_args: confupsname=UPS1, var=desc, val=Example event sequence
   0.000257     [D5:61105] do_upsconf_args: call main_arg()
   0.000262     [D3:61105] main_arg: var='desc' val='Example event sequence'
   0.000265     [D5:61105] do_upsconf_args: confupsname=UPS1, var=port, val=evolution500.seq
   0.000267     [D5:61105] do_upsconf_args: call main_arg()
   0.000269     [D3:61105] main_arg: var='port' val='evolution500.seq'
   0.000272     [D6:61105] testinfo_reloadable: var=port, infoname=driver.parameter.port, newval=evolution500.seq, reloadable=0, reload_flag=0
   0.000290     [D6:61105] testinfo_reloadable: verdict for (re)loading var=port value: 1
   0.000298     [D5:61105] send_to_all: SETINFO driver.parameter.port "evolution500.seq"
   0.000318     [D5:61105] do_upsconf_args: confupsname=UPS2, var=driver, val=dummy-ups
   0.000340     [D5:61105] do_upsconf_args: confupsname=UPS2, var=desc, val=Example ePDU data dump
   0.000345     [D5:61105] do_upsconf_args: confupsname=UPS2, var=port, val=epdu-managed.dev
   0.000376     [D5:61105] do_upsconf_args: confupsname=UPS2, var=mode, val=dummy-once
   0.000385     [D1] Network UPS Tools version 2.8.2.1303.84-1387-g019529e68 (development iteration after 2.8.2) built with gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 and configured with flags: --enable-Wcolor --enable-warnings --enable-Werror --enable-keep_nut_report_feature --with-all=auto --with-cgi=auto --with-serial=auto --with-dev=auto --with-doc=skip --with-nut_monitor=auto --with-pynut=auto --disable-force-nut-version-header --enable-check-NIT --enable-maintainer-mode --disable-silent-rules --with-debuginfo=auto
   0.000391     [D1:61105] debug level is '6'
   0.000394     [D5:61105] send_to_all: SETINFO driver.debug "6"
   0.000414     [D5:61105] send_to_all: SETFLAGS driver.debug RW NUMBER
   0.000496     [D1:61105] Can not become_user(nobody): not root initially, remaining UID=1000 GID=1000
   0.000534     [D5:61105] upsdrvquery_write: write to driver socket: NOBROADCAST

   0.000560     [D5:61105] upsdrvquery_write: write to driver socket: PING

   0.000583     [D5:61105] upsdrvquery_prepare: waiting for a while to flush server messages
sock_connect: enabling asynchronous mode (auto)
   0.162337     [D5:61105] upsdrvquery_read_timeout: received 5 bytes from driver socket: PONG

   0.162382     [D5:61105] upsdrvquery_prepare: got expected PONG
   0.162390     [D5:61105] upsdrvquery_prepare: ready for tracked commands
   0.162427     [D5:61105] upsdrvquery_write: write to driver socket: SET driver.flag.allow_killpower 1 TRACKING 2DDC5036-F665-4D29-BE4C-0374D091D263

   0.162471     [D5:61105] upsdrvquery_request: will wait up to 15.000000 sec for response to SET driver.flag.allow_killpower 1

   0.162626     [D5:61105] upsdrvquery_read_timeout: received 48 bytes from driver socket: TRACKING 2DDC5036-F665-4D29-BE4C-0374D091D263 0

   0.162668     [D5:61105] upsdrvquery_request: parsed out command status: 0
   0.162676     [D5:61105] upsdrvquery_close: closing driver socket, try to say goodbye
   0.162680     [D5:61105] upsdrvquery_write: write to driver socket: LOGOUT

   0.162742     [D5:61105] upsdrvquery_close: okay
   1.163005     [D5:61105] upsdrvquery_write: write to driver socket: NOBROADCAST

   1.163055     [D5:61105] upsdrvquery_write: write to driver socket: PING

   1.163061     [D5:61105] upsdrvquery_prepare: waiting for a while to flush server messages
sock_connect: enabling asynchronous mode (auto)
   2.163279     [D5:61105] upsdrvquery_read_timeout: received 5 bytes from driver socket: PONG

   2.163293     [D5:61105] upsdrvquery_prepare: got expected PONG
   2.163322     [D5:61105] upsdrvquery_prepare: ready for tracked commands
   2.163330     [D5:61105] upsdrvquery_write: write to driver socket: INSTCMD driver.killpower TRACKING 98C489FD-821C-4132-9D93-4F6590B55E63

   2.163394     [D1:61105] upsdrvquery_request: will wait indefinitely for response to INSTCMD driver.killpower

Requesting UPS [UPS1] to power off, as/if handled by its driver by default (may exit), due to socket protocol request
   2.163430     select with socket: Invalid argument
   2.163451     [D5:61105] upsdrvquery_close: closing driver socket, try to say goodbye
instcmd: unknown command [shutdown.default] [(null)]
   2.163455     [D5:61105] upsdrvquery_write: write to driver socket: LOGOUT

UPS [UPS1]: shutdown request(s) failed
   2.163467     [D5:61105] upsdrvquery_close: okay
Signal -1: exiting
Can't connect to UPS [UPS1] (dummy-ups-UPS1): No such file or directory
   3.163624     [D1:61105] Socket dialog with the other driver instance: Invalid argument
   3.163681     [D5:61105] send_to_all: SETINFO device.type "ups"
   3.163688     [D5:61105] send_to_all: SETINFO driver.state "init.device"
   3.163720     [D2:61105] Dummy (simulation) mode with a sequence file name pattern (looping infinitely)
   3.163748     [D1:61105] Dummy (simulation) mode looping infinitely
   3.163777     [D5:61105] send_to_all: SETINFO driver.parameter.mode "dummy-loop"
   3.163819     [D2:61105] Located evolution500.seq for device simulation data: /home/jim/nut/tests/NIT/tmp/etc/evolution500.seq
   3.163849     [D5:61105] send_to_all: SETINFO driver.state "init.quiet"
   3.163875     Initiating UPS [UPS1] shutdown
   3.163885     [D1:61105] loop_shutdown_commands: call do_loop_shutdown_commands() with driver-default sdcommands
   3.163909     [D1:61105] Starting do_loop_shutdown_commands(shutdown.default), call depth 1...
   3.163914     [D1:61105] This driver does not implement INSTCMD support
   3.163939     [D1:61105] Handle 'shutdown.default' directly, ignore other `sdcommands` (if any): shutdown.default
   3.163944     shutdown not supported
   3.163948     [D1:61105] set_exit_flag: raising exit flag due to signal -1
   3.164000     [D1:61105] Ending do_loop_shutdown_commands(shutdown.default), call depth 1: return-code 0
   3.164026     UPS [UPS1]: shutdown request was successful with '(null)'
   3.164032     [D1:61105] set_exit_flag: raising exit flag due to programmatic abort: EXIT_FAILURE
   3.164037     [D5:61105] send_to_all: SETINFO driver.state "cleanup.upsdrv"
   3.164041     [D5:61105] send_to_all: SETINFO driver.state "cleanup.exit"
   3.164047     [D1:61105] upsnotify: notify about state NOTIFY_STATE_STOPPING with libsystemd: was requested, but not running as a service unit now, will not spam more about it
   3.164050     [D1:61105] On systems without service units, consider `export NUT_QUIET_INIT_UPSNOTIFY=true`
   3.164054     [D1:61105] upsnotify: failed to notify about state NOTIFY_STATE_STOPPING: no notification tech defined, will not spam more about it
   3.164057     [D1:61105] On systems without service units, consider `export NUT_QUIET_INIT_UPSNOTIFY=true`


:; ./clients/upscmd -u admin -p "$TESTPASS_ADMIN" -l "UPS1@localhost:$NUT_PORT"
Error: Driver not connected

Got a few wrinkles to investigate though, with instcmd: unknown command [shutdown.default] [(null)] and UPS [UPS1]: shutdown request was successful with '(null)' - maybe the tested system was from an older iteration though

…handler" [networkupstools#2686] Signed-off-by: Jim Klimov <[email protected]>

Signed-off-by: Jim Klimov <[email protected]>

…a debug message about its limited usefullness Signed-off-by: Jim Klimov <[email protected]>

Signed-off-by: Jim Klimov <[email protected]>

jimklimov · 2024-12-02T17:03:49Z

Sandboxing helped find some issues with upsdrvquery, fixed now (longer timestamp in log is from the running driver daemon, shorter is from drivername -k):

...
   0.000584     [D1:173231] Can not become_user(nobody): not root initially, remaining UID=1000 GID=1000
   0.000621     [D5:173231] upsdrvquery_write: write to driver socket: NOBROADCAST

   0.000647     [D5:173231] upsdrvquery_write: write to driver socket: PING

   0.000669     [D5:173231] upsdrvquery_prepare: waiting for a while to flush server messages
   0.000674     [D1:173231] upsdrvquery_read_timeout: tv={sec=3, usec=000000}
   8.009857     [D1:173190] entering parse_data_file()
   8.009910     [D1:173190] leaving (paused)...
   8.009924     [D5:173190] send_to_all: SETINFO driver.state "quiet"
   8.009985     [D6:173190] send_to_all: write 29 bytes to socket 5 succeeded (ret=29): SETINFO driver.state "quiet"

   8.010033     sock_connect: enabling asynchronous mode (auto)
   8.010044     [D3:173190] sock_connect: new connection on fd 6
   8.010081     [D6:173190] sock_arg: Driver on /home/jim/nut/tests/NIT/tmp/run/dummy-ups-UPS1 is now handling NOBROADCAST with 1 args
   8.010087     [D1:173190] sock_arg: socket 6 requested NOBROADCAST mode
   8.010109     [D6:173190] sock_arg: Driver on /home/jim/nut/tests/NIT/tmp/run/dummy-ups-UPS1 is now handling PING with 1 args
   8.010116     [D2:173190] send_to_one: sending PONG
   8.010121     [D5:173190] send_to_one: PONG
   8.010151     [D6:173190] send_to_one: write 5 bytes to socket 6 succeeded (ret=5): PONG

   0.097185     [D5:173231] upsdrvquery_read_timeout: received 5 bytes from driver socket: PONG

   0.097217     [D5:173231] upsdrvquery_prepare: got expected PONG
   0.097243     [D5:173231] upsdrvquery_prepare: ready for tracked commands
   0.097253     [D5:173231] upsdrvquery_write: write to driver socket: SET driver.flag.allow_killpower 1 TRACKING 2DDC5036-F665-4D29-BE4C-0374D091D263

   0.097298     [D5:173231] upsdrvquery_request: will wait up to 15.000000 sec for response to SET driver.flag.allow_killpower 1

   0.097307     [D1:173231] upsdrvquery_read_timeout: tv={sec=15, usec=000000}
   8.010314     [D6:173190] sock_arg: Driver on /home/jim/nut/tests/NIT/tmp/run/dummy-ups-UPS1 is now handling SET with 5 args
   8.010328     [D3:173190] sock_arg: TRACKING = 2DDC5036-F665-4D29-BE4C-0374D091D263
   8.010334     [D2:173190] entering main_setvar(driver.flag.allow_killpower, 1) for [UPS1] on socket 6
   8.010369     [D1:173190] main_setvar: Setting driver.flag.allow_killpower=1
   8.010378     [D5:173190] send_to_all: SETINFO driver.flag.allow_killpower "1"
   8.010402     [D6:173190] send_to_all: write 40 bytes to socket 5 succeeded (ret=40): SETINFO driver.flag.allow_killpower "1"

   8.010425     [D2:173190] send_to_one: sending TRACKING 2DDC5036-F665-4D29-BE4C-0374D091D263 0
   8.010429     [D5:173190] send_to_one: TRACKING 2DDC5036-F665-4D29-BE4C-0374D091D263 0
   8.010467     [D6:173190] send_to_one: write 48 bytes to socket 6 succeeded (ret=48): TRACKING 2DDC5036-F665-4D29-BE4C-0374D091D263 0

   0.097501     [D5:173231] upsdrvquery_read_timeout: received 48 bytes from driver socket: TRACKING 2DDC5036-F665-4D29-BE4C-0374D091D263 0

   0.097546     [D5:173231] upsdrvquery_request: parsed out command status: 0
   0.097586     [D5:173231] upsdrvquery_close: closing driver socket, try to say goodbye
   0.097626     [D5:173231] upsdrvquery_write: write to driver socket: LOGOUT

   0.097696     [D5:173231] upsdrvquery_close: okay
   8.010698     [D6:173190] sock_arg: Driver on /home/jim/nut/tests/NIT/tmp/run/dummy-ups-UPS1 is now handling LOGOUT with 1 args
   8.010747     [D2:173190] send_to_one: sending OK Goodbye
   8.010776     [D5:173190] send_to_one: OK Goodbye
   8.010805     [D6:173190] send_to_one: write 11 bytes to socket 6 succeeded (ret=11): OK Goodbye

   8.010826     [D2:173190] sock_arg: received LOGOUT on socket 6, will be disconnecting
   1.097967     [D5:173231] upsdrvquery_write: write to driver socket: NOBROADCAST

   9.010971     [D4:173190] sock_arg: LOGOUT processing finished
   9.011006     [D1:173190] sock_read: returning early, socket may be not valid anymore
   1.098017     [D5:173231] upsdrvquery_write: write to driver socket: PING

   9.011011     [D3:173190] sock_disconnect: disconnecting socket 6
   1.098023     [D5:173231] upsdrvquery_prepare: waiting for a while to flush server messages
   1.098027     [D1:173231] upsdrvquery_read_timeout: tv={sec=3, usec=000000}
   9.011027     [D5:173190] sock_disconnect: finishing parsing context
   9.011031     [D5:173190] sock_disconnect: relinking the chain of connections
   9.011063     [D5:173190] sock_disconnect: freeing the conn object
   9.011086     sock_connect: enabling asynchronous mode (auto)
   9.011132     [D3:173190] sock_connect: new connection on fd 6
   9.011172     [D5:173190] send_to_all: SETINFO driver.state "updateinfo"
   9.011245     [D6:173190] send_to_all: write 34 bytes to socket 6 succeeded (ret=34): SETINFO driver.state "updateinfo"

   1.098291     [D5:173231] upsdrvquery_read_timeout: received 34 bytes from driver socket: SETINFO driver.state "updateinfo"

   9.011320     [D6:173190] send_to_all: write 34 bytes to socket 5 succeeded (ret=34): SETINFO driver.state "updateinfo"

   9.011333     [D1:173190] upsdrv_updateinfo...
   1.098342     [D1:173231] upsdrvquery_read_timeout: tv={sec=3, usec=000000}
  10.011486     [D1:173190] entering parse_data_file()
  10.011537     [D1:173190] leaving (paused)...
  10.011549     [D5:173190] send_to_all: SETINFO driver.state "quiet"
  10.011608     [D6:173190] send_to_all: write 29 bytes to socket 6 succeeded (ret=29): SETINFO driver.state "quiet"

   2.098659     [D5:173231] upsdrvquery_read_timeout: received 29 bytes from driver socket: SETINFO driver.state "quiet"

  10.011663     [D6:173190] send_to_all: write 29 bytes to socket 5 succeeded (ret=29): SETINFO driver.state "quiet"

   2.098690     [D1:173231] upsdrvquery_read_timeout: tv={sec=2, usec=999999}
  10.011696     [D6:173190] sock_arg: Driver on /home/jim/nut/tests/NIT/tmp/run/dummy-ups-UPS1 is now handling NOBROADCAST with 1 args
  10.011725     [D1:173190] sock_arg: socket 6 requested NOBROADCAST mode
  10.011734     [D6:173190] sock_arg: Driver on /home/jim/nut/tests/NIT/tmp/run/dummy-ups-UPS1 is now handling PING with 1 args
  10.011763     [D2:173190] send_to_one: sending PONG
  10.011768     [D5:173190] send_to_one: PONG
  10.011793     [D6:173190] send_to_one: write 5 bytes to socket 6 succeeded (ret=5): PONG

   2.098821     [D5:173231] upsdrvquery_read_timeout: received 5 bytes from driver socket: PONG

   2.098897     [D5:173231] upsdrvquery_prepare: got expected PONG
   2.098923     [D5:173231] upsdrvquery_prepare: ready for tracked commands
   2.098933     [D5:173231] upsdrvquery_write: write to driver socket: INSTCMD driver.killpower TRACKING 98C489FD-821C-4132-9D93-4F6590B55E63

   2.098977     [D1:173231] upsdrvquery_request: will wait indefinitely for response to INSTCMD driver.killpower

  10.011993     [D6:173190] sock_arg: Driver on /home/jim/nut/tests/NIT/tmp/run/dummy-ups-UPS1 is now handling INSTCMD with 4 args
   2.099007     [D1:173231] upsdrvquery_read_timeout: tv={sec=-1, usec=-00001} (unlimited timeout)
  10.012001     [D3:173190] sock_arg: TRACKING = 98C489FD-821C-4132-9D93-4F6590B55E63
  10.012015     [D2:173190] entering main_instcmd(driver.killpower, (null)) for [UPS1] on socket 6
  10.012021     Requesting UPS [UPS1] to power off, as/if handled by its driver by default (may exit), due to socket protocol request
  10.012047     [D5:173190] send_to_all: SETINFO driver.state "fsd.killpower"
  10.012075     [D6:173190] send_to_all: write 37 bytes to socket 5 succeeded (ret=37): SETINFO driver.state "fsd.killpower"

  10.012103     [D1:173190] loop_shutdown_commands: call do_loop_shutdown_commands() with driver-default sdcommands
  10.012108     [D1:173190] Starting do_loop_shutdown_commands(shutdown.default), call depth 1...
  10.012124     shutdown not supported
  10.012149     [D1:173190] set_exit_flag: raising exit flag due to signal -1
  10.012175     [D1:173190] do_loop_shutdown_commands(): command 'shutdown.default' was handled successfully
  10.012201     [D1:173190] Ending do_loop_shutdown_commands(shutdown.default), call depth 1: return-code 0
  10.012226     UPS [UPS1]: shutdown request was successful with 'shutdown.default'
  10.012252     [D1:173190] set_exit_flag: raising exit flag due to programmatic abort: EXIT_FAILURE
  10.012279     [D5:173190] send_to_all: SETINFO driver.state "quiet"
  10.012326     [D6:173190] send_to_all: write 29 bytes to socket 5 succeeded (ret=29): SETINFO driver.state "quiet"

  10.012351     [D2:173190] send_to_one: sending TRACKING 98C489FD-821C-4132-9D93-4F6590B55E63 0
  10.012355     [D5:173190] send_to_one: TRACKING 98C489FD-821C-4132-9D93-4F6590B55E63 0
  10.012391     [D6:173190] send_to_one: write 48 bytes to socket 6 succeeded (ret=48): TRACKING 98C489FD-821C-4132-9D93-4F6590B55E63 0

   2.099414     [D5:173231] upsdrvquery_read_timeout: received 48 bytes from driver socket: TRACKING 98C489FD-821C-4132-9D93-4F6590B55E63 0

  10.012412     Signal -2: exiting
   2.099425     [D5:173231] upsdrvquery_request: parsed out command status: 0
  10.012420     [D1:173190] On systems without service units, consider `export NUT_QUIET_INIT_UPSNOTIFY=true`
   2.099428     [D5:173231] upsdrvquery_close: closing driver socket, try to say goodbye
  10.012458     [D5:173190] send_to_all: SETINFO driver.state "cleanup.upsdrv"
   2.099469     [D5:173231] upsdrvquery_write: write to driver socket: LOGOUT

  10.012483     [D6:173190] send_to_all: write 38 bytes to socket 5 succeeded (ret=38): SETINFO driver.state "cleanup.upsdrv"

   2.099496     [D5:173231] upsdrvquery_close: okay
  10.012492     [D5:173190] send_to_all: SETINFO driver.state "cleanup.exit"
  10.012526     [D6:173190] send_to_all: write 36 bytes to socket 5 succeeded (ret=36): SETINFO driver.state "cleanup.exit"

  10.012549     [D1:173190] On systems without service units, consider `export NUT_QUIET_INIT_UPSNOTIFY=true`
  10.012744     [D3:173190] sock_disconnect: disconnecting socket 6
  10.012776     [D5:173190] sock_disconnect: finishing parsing context
  10.012801     [D5:173190] sock_disconnect: relinking the chain of connections
  10.012806     [D5:173190] sock_disconnect: freeing the conn object
  10.012811     [D3:173190] sock_disconnect: disconnecting socket 5
  10.012848     [D5:173190] sock_disconnect: finishing parsing context
  10.012870     [D5:173190] sock_disconnect: relinking the chain of connections
  10.012874     [D5:173190] sock_disconnect: freeing the conn object
   3.099617     Request to killpower via running driver returned code 0
   3.099672     [D5:173231] send_to_all: SETINFO driver.state "cleanup.exit"
   3.099706     [D1:173231] upsnotify: notify about state NOTIFY_STATE_STOPPING with libsystemd: was requested, but not running as a service unit now, will not spam more about it
   3.099731     [D1:173231] On systems without service units, consider `export NUT_QUIET_INIT_UPSNOTIFY=true`
   3.099758     [D1:173231] upsnotify: failed to notify about state NOTIFY_STATE_STOPPING: no notification tech defined, will not spam more about it
   3.099786     [D1:173231] On systems without service units, consider `export NUT_QUIET_INIT_UPSNOTIFY=true`

So both copies of the driver program met, talked, and exited (with the daemon handling shutdown the best it could - no-op - and raising the exit flag).

…vate use of timeval={-1,-1} as select(..., NULL) for indefinite wait [networkupstools#1922, networkupstools#2392, networkupstools#2686] Signed-off-by: Jim Klimov <[email protected]>

…vate use of timeval={-1,-1} as select(..., NULL) for indefinite wait [networkupstools#1922, networkupstools#2392, networkupstools#2686, networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

jimklimov added 30 commits November 18, 2024 10:42

Introduce "sdcommands" setting for all drivers [networkupstools#2670]

f6eb991

Signed-off-by: Jim Klimov <[email protected]>

drivers/mge-utalk.c: update to use "sdcommands" [networkupstools#2670]

ccddee5

Signed-off-by: Jim Klimov <[email protected]>

drivers/everups.c: update to use "sdcommands" and have an instcmd() a…

4cf357a

…t all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/adelsystem_cbi.c: update to use "sdcommands" and move logic f…

aa08ae5

…rom upsdrv_shutdown() to upscmd("shutdown.stayoff") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/al175.c: update to use "sdcommands" (if customized by user) [n…

e775b01

…etworkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/apc_modbus.c: update to use "sdcommands" (if customized by us…

1286775

…er) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/apcsmart-old.c: update to use "sdcommands" (if customized by …

f8da375

…user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/apcsmart.c: update to use "sdcommands" (if customized by user) [

b8fb714

networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/apcupsd-ups.c: update to use "sdcommands" (if customized by u…

c18e0aa

…ser) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/asem.c: update to use "sdcommands" (if customized by user) [n…

ac1b463

…etworkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/bcmxcp.c: update to use "sdcommands" (if customized by user) [n…

cd19aa9

…etworkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/belkin.c: update to use "sdcommands" and move logic from upsd…

14a62df

…rv_shutdown() to upscmd("shutdown.return") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/belkinunv.c: update to use "sdcommands" (if customized by use…

2ab7f2c

…r); revise warning message [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/bestfcom.c: update to use "sdcommands" and move logic from up…

b0d683a

…sdrv_shutdown() to upscmd("shutdown.return"); newly support INSTCMD at all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/bestfortress.c: update to use "sdcommands" and move logic fro…

675db77

…m upsdrv_shutdown() to upscmd("shutdown.return") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/bestuferrups.c: update to use "sdcommands" and move logic fro…

5fe1576

…m upsdrv_shutdown() to upscmd("shutdown.return"); newly support INSTCMD at all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/bestups.c: update to use "sdcommands" (if customized by user) […

fd42c1b

…networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/bicker_ser.c: update to use "sdcommands" (if customized by us…

465552c

…er) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/blazer*.c: update to use "sdcommands" (if customized by user) […

4230032

…networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/clone-outlet.c: update to use "sdcommands" (if customized by …

e738dbc

…user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/clone.c: update to use "sdcommands" (if customized by user) [n…

fd6b951

…etworkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/dummy-ups.c: update to use "sdcommands" (if customized by use…

e79cdea

…r) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/etapro.c: update to use "sdcommands" and move logic from upsd…

34b24b6

…rv_shutdown() to upscmd("shutdown.return") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/gamatronic.c: update to use "sdcommands" and move logic from …

80d583d

…upsdrv_shutdown() to upscmd("shutdown.return"); newly support INSTCMD at all [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/gamatronic.c: fix misplaced upslogx(1, ...) by upsdebugx()

37af651

Signed-off-by: Jim Klimov <[email protected]>

drivers/generic_gpio_common.c: update to use "sdcommands" (if customi…

889c77e

…zed by user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/generic_modbus.c: update to use "sdcommands" and move logic f…

37d51c4

…rom upsdrv_shutdown() to upscmd("shutdown.stayoff") [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/genericups.c: update to use "sdcommands" (if customized by us…

d16a41a

…er) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/huawei-ups2000.c: update to use "sdcommands" (if customized b…

dea5b93

…y user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/hwmon_ina219.c: update to use "sdcommands" (if customized by …

2377d06

…user) [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

jimklimov added 7 commits November 19, 2024 20:56

drivers/main.{c,h}, drivers/dstate.c: introduce main_instcmd_fallback…

25250b0

…() so drivers might implement their own "shutdown.default" at will [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/main.{c,h}: introduce MAX_SDCOMMANDS_DEPTH for do_loop_shutdo…

17cf2da

…wn_commands() [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/main.c: update some comments [networkupstools#2670]

a94bc83

Signed-off-by: Jim Klimov <[email protected]>

drivers/*.c: simplify upsdrv_shutdown() back, as only "shutdown.defau…

be0a35a

…lt" implementations again [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/main.c: move do_forceshutdown handling to be after upsdrv_ini…

d0b42c4

…tinfo() where we register INSTCMD handlers [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

drivers/main.c: do_loop_shutdown_commands(): handle "shutdown.default…

1d92ded

…" as upsdrv_shutdown() in main loop too [networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

jimklimov added this to the 2.8.3 milestone Nov 19, 2024

Merge branch 'master' into issue-2670

decb643

Merge branch 'master' into issue-2670

edde0c3

jimklimov added 5 commits December 2, 2024 13:14

Merge branch 'master' into issue-2670

a480b29

drivers/dstate.c: revise logging of "Got INSTCMD, but driver lacks a …

2c7d98b

…handler" [networkupstools#2686] Signed-off-by: Jim Klimov <[email protected]>

data/cmdvartab: document CMDDESC shutdown.default [networkupstools#2686]

2072d83

Signed-off-by: Jim Klimov <[email protected]>

clients/ups{rw,cmd}.c: when reporting "OK" without "TRACKING", issue …

aaf254d

…a debug message about its limited usefullness Signed-off-by: Jim Klimov <[email protected]>

clients/ups{rw,cmd}.c: enable use of NUT_DEBUG_LEVEL envvar

35edb4f

Signed-off-by: Jim Klimov <[email protected]>

drivers/upsdrvquery.c, NEWS.adoc: upsdrvquery_read_timeout(): fix pri…

a9b75b1

…vate use of timeval={-1,-1} as select(..., NULL) for indefinite wait [networkupstools#1922, networkupstools#2392, networkupstools#2686, networkupstools#2670] Signed-off-by: Jim Klimov <[email protected]>

jimklimov force-pushed the issue-2670 branch from 7483ccf to a9b75b1 Compare December 2, 2024 17:11

jimklimov added 2 commits December 4, 2024 16:18

Merge branch 'master' into issue-2670

08efbcd

Merge branch 'master' into issue-2670

b804bf8

jimklimov merged commit 0727ce2 into networkupstools:master Dec 7, 2024
30 checks passed

jimklimov deleted the issue-2670 branch December 7, 2024 16:02

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Introduce `sdcommands` option and `shutdown.default` INSTCMD to all drivers #2686

Introduce `sdcommands` option and `shutdown.default` INSTCMD to all drivers #2686

jimklimov commented Nov 19, 2024

AppVeyorBot commented Dec 1, 2024

jimklimov commented Dec 2, 2024

jimklimov commented Dec 2, 2024 •

edited

Loading

jimklimov commented Dec 2, 2024

Introduce sdcommands option and shutdown.default INSTCMD to all drivers #2686

Introduce sdcommands option and shutdown.default INSTCMD to all drivers #2686

Conversation

jimklimov commented Nov 19, 2024

AppVeyorBot commented Dec 1, 2024

jimklimov commented Dec 2, 2024

jimklimov commented Dec 2, 2024 • edited Loading

jimklimov commented Dec 2, 2024

Introduce `sdcommands` option and `shutdown.default` INSTCMD to all drivers #2686

Introduce `sdcommands` option and `shutdown.default` INSTCMD to all drivers #2686

jimklimov commented Dec 2, 2024 •

edited

Loading