Add simple wrapper for monit to print which process caused monit to trigger
restart.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Distinguish between start-stop-daemon return value 0, 1 and the rest for start.
Check start-stop-daemon return value for stop.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Use start-stop-daemon to start/stop syslogd daemon.
Distinguish between start-stop-daemon return value 0, 1 and the rest.
Fix syslog's pid filename (was /var/log/rsyslog.pid now is
/var/run/rsyslogd.pid).

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Use start-stop-daemon to start/stop snmpd daemon.
Change name of init script from snmp to snmpd. Also rename link from S80snmp to
S80snmpd. Rename and update snmpd config for monit.

Distinguish between start-stop-daemon return value 0, 1 and the rest.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Fix copy-paste error.
Distinguish between start-stop-daemon return value 0, 1 and the rest.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Improve getting scb version in boot scripts

Use wrs_version -s in /wr/sbin/startup-mb.sh to get scb version.
Previous method to distinguish scb versions did not work in all cases:
mtdinfo -a | grep -A 1 dataflash | grep 264

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

This parameter will be used by /wr/sbin/startup-mb.sh script to pass scb
version to lm32 loader. "-t" parameter cannot be used here because lm32 loader
expect version number to be without dots. It is shorter, better, faster, more
elegant to put extra parameter "-s" instead of using e.g. sed to parse "-t"
output.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Print version at startup of HAL and RTUd.
Replace "+" with "--dirty" for version printouts

Print "-dirty" for:
--at91boot
--startup guide
--switch's tools

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

It is good to have in logs when and what version of rtud started.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

It is good to have in logs when and what version of hal started.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Sometimes we want to print message with high priority without indication that
it is error. This is workaround for this problem.
For example report program's version.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Unfortunately this is non compatible change, but hopefully noone use SNMP
so far. Due to changes in ppsi.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

bugfix for commit:
c552ec92

 serspace/wrsw_rtud: Don't inform about waiting for HAL for 5 seconds

I forgot "!" at the beginning of condition

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

This commit also fixes the tools that access ppsi's shmem.

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Alessandro Rubini authored 9 years ago and Adam Wujek committed 9 years ago

This allows, among other things, to see big jumps happening (if we are
slave and the master jumped in time due to user interaction: this is "only"
useful during developement, i.e. it *very* useful during development).

Signed-off-by: Alessandro Rubini <rubini@gnudd.com>

Make rtud and rtu_stat to be able to start in parallel with hal and ppsi.
(wait for hal in rtud, wait for hal and rtud in stu_stat)

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Since rtud and rtu_stat wait internally for hal and rtud shmem to be available,
there is no need to check that in startup script. For the same reason we can
remove "sleep 4" before calling rtu_stat.

Also there is no need to check whether ppsi binary is available. Such situation
will be reported in different ways.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

rtu_stat has to wait for hal and rtud to be ready.
It waits for hal's shmem and rtud's minipc connection. There is no need to
wait for rtud's shmem since it is initialized before ipc server is available.

Additionally few minor changes in printouts.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

It may take few seconds to start hal, rtud shouln't complain about it during
5 seconds.
Additionally change fprintf into pr_error.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Sometimes it happens at boot that rts_get_state fails. It may be due to race
condition with ppsi or something else. Anyway try again when rts_get_state
fails.
Additionally print message when got timeout from tmo_expired.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

--Move assert_init macro to libwr/wrs-msg.h
--Add printout to assert_init macro to know which function fails.
--Remove assert_init macro duplication from rsw_hal/hal_main.c
--Remove #include <libwr/shw_io.h> from libwr/shw_io.h (no need to include file
from itself)

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Also support HAL and PPSI restarts in wr_mon.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Make monit to start from inittab instead of rcS script.
Add config files for:
--dropbear
--lighttpd

We want that monit don't propagate its state over restart. Move monit.stat to
/tmp.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Additionally change startup script to use start-stop-daemon
Note: start-stop-daemon does not create pid file

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

When monit is started from inittab instead of script /etc/init.d/monit.sh
it will be respawned after its crash.
Support stopping/starting monit by /etc/init.d/monit.sh
To stop monit, STOP signal is sent to it. By this, init will not respawn monit,
because it is still present in process' list.
Add disabling monit in dot-config by option CONFIG_MONIT_DISABLE.
When monit is disabled by dot-config, monit.sh will sleep forever, to
satisfy init and prevent respawning
Update dot-config items in user-manual.

Include only *.conf instead of all files from /etc/monit.d/.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Add explicit error messages when problems with downloaded dot-config file.
Messages added in /etc/init.d/dot-config

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

parameters show up as a result of copy-paste error

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Move:
/etc/init.d/[SK]??<name> to /etc/init.d/<name>
Then create symlinks in
/etc/init.d/rcS/[SK]??* to all files that previously were /etc/init.d/[SK]??*
With exception of monit since monit will start soon from init.

Rename all occurences of changed files.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Date was never restored due to wrong filename in /etc/init.d/S70wr_date
Use saved_date instead of update_date

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Update required because otherwise setting of auxclk hangs the switch.

Signed-off-by: Adam Wujek <adam.wujek@cern.ch>

Add wrs_auxclk to start at boot

NOTE: This branch requires new Gateware! wrs_auxclk will hang switch when using
      gateware from last release of switch.