[opensuse] proliant support pack and systemd
Is anyone running 12.3+systemd on HP proliant? I've installed the PSP, but I can't start the SNMP agents. I get weird messages: 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Health agent (cmahealthd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Standard Equipment agent (cmastdeqd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Host agent (cmahostd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Threshold agent (cmathreshd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start Going through the scripts, it looks like "redirecting to systemctl start" is caused by: systemctl start hp-snmp-agents this calls /opt/hp/hp-snmp-agents/server/etc/cmad : case "$1" in start) $STARTADDON RETVAL=$? if [ $RETVAL -eq 0 ]; then touch $AGNTFILEPATH/$PNAME.pid pidlist=`$PIDOF -o $$ $PNAME` if [ -z "$pidlist" ]; then cmaechon "Starting $NAME ($PNAME): " $PNAME $PFLAGS < /dev/null > /dev/null 2>&1 & ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Why is systemd intercepting this? -- Per Jessen, Zürich (7.2°C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
В Thu, 31 Oct 2013 20:32:41 +0100 Per Jessen <per@computer.org> пишет:
Is anyone running 12.3+systemd on HP proliant? I've installed the PSP, but I can't start the SNMP agents. I get weird messages:
2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Health agent (cmahealthd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Standard Equipment agent (cmastdeqd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Host agent (cmahostd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Threshold agent (cmathreshd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start
Going through the scripts, it looks like "redirecting to systemctl start" is caused by:
systemctl start hp-snmp-agents this calls /opt/hp/hp-snmp-agents/server/etc/cmad :
case "$1" in start) $STARTADDON RETVAL=$? if [ $RETVAL -eq 0 ]; then touch $AGNTFILEPATH/$PNAME.pid pidlist=`$PIDOF -o $$ $PNAME` if [ -z "$pidlist" ]; then cmaechon "Starting $NAME ($PNAME): " $PNAME $PFLAGS < /dev/null > /dev/null 2>&1 & ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Why is systemd intercepting this?
That is standard wrapper for any initscript that is using openSUSE API: bor@opensuse:~> less /etc/rc.status ... # Check if the service is used under systemd but not started with if test -z "$SYSTEMD_NO_WRAP" && /bin/mountpoint -q /sys/fs/cgroup/systemd > /dev/null 2>&1 ; then ... case "$1" in status) SYSTEMD_NO_WRAP=1 "$0" "$1" ;; start|stop|reload|restart|try-restart|force-reload) echo "redirecting to systemctl ${SYSTEMCTL_OPTIONS} $1 ${_rc_base}" 1>&2 ;; So if $PNAME refers to initscript, it will pass call through to systemctl. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Andrey Borzenkov wrote:
В Thu, 31 Oct 2013 20:32:41 +0100 Per Jessen <per@computer.org> пишет:
Is anyone running 12.3+systemd on HP proliant? I've installed the PSP, but I can't start the SNMP agents. I get weird messages:
2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Health agent (cmahealthd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Standard Equipment agent (cmastdeqd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Host agent (cmahostd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: Starting Threshold agent (cmathreshd): ..done 2013-10-31T17:26:45+01:00 rainbow hp-snmp-agents[13542]: redirecting to systemctl start
Going through the scripts, it looks like "redirecting to systemctl start" is caused by:
systemctl start hp-snmp-agents this calls /opt/hp/hp-snmp-agents/server/etc/cmad :
case "$1" in start) $STARTADDON RETVAL=$? if [ $RETVAL -eq 0 ]; then touch $AGNTFILEPATH/$PNAME.pid pidlist=`$PIDOF -o $$ $PNAME` if [ -z "$pidlist" ]; then cmaechon "Starting $NAME ($PNAME): " $PNAME $PFLAGS < /dev/null > /dev/null 2>&1 & ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Why is systemd intercepting this?
That is standard wrapper for any initscript that is using openSUSE API:
bor@opensuse:~> less /etc/rc.status ... # Check if the service is used under systemd but not started with if test -z "$SYSTEMD_NO_WRAP" && /bin/mountpoint -q /sys/fs/cgroup/systemd > /dev/null 2>&1 ; then ... case "$1" in status) SYSTEMD_NO_WRAP=1 "$0" "$1" ;; start|stop|reload|restart|try-restart|force-reload) echo "redirecting to systemctl ${SYSTEMCTL_OPTIONS} $1 ${_rc_base}" 1>&2 ;;
So if $PNAME refers to initscript, it will pass call through to systemctl.
That's the thing, $PNAME is not an initscript, it is the name of a binary in a separate path: /opt/hp/hp-snmp-agents/storage/bin/. $NAME is the textual description, e.g. "Health agent", $PNAME is "cmahealthd". In the log messages, there is no name of what it is being redirected to: "redirecting to systemctl start ????" -- Per Jessen, Zürich (7.4°C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Per Jessen wrote:
Andrey Borzenkov wrote:
That is standard wrapper for any initscript that is using openSUSE API:
bor@opensuse:~> less /etc/rc.status ... # Check if the service is used under systemd but not started with if test -z "$SYSTEMD_NO_WRAP" && /bin/mountpoint -q /sys/fs/cgroup/systemd > /dev/null 2>&1 ; then ... case "$1" in status) SYSTEMD_NO_WRAP=1 "$0" "$1" ;; start|stop|reload|restart|try-restart|force-reload) echo "redirecting to systemctl ${SYSTEMCTL_OPTIONS} $1 ${_rc_base}" 1>&2 ;;
So if $PNAME refers to initscript, it will pass call through to systemctl.
That's the thing, $PNAME is not an initscript, it is the name of a binary in a separate path: /opt/hp/hp-snmp-agents/storage/bin/. $NAME is the textual description, e.g. "Health agent", $PNAME is "cmahealthd".
In the log messages, there is no name of what it is being redirected to: "redirecting to systemctl start ????"
Anyway, you gave me the idea of adding SYSTEMD_NO_WRAP=1 to the initscript, which led to the same problem, i.e. no daemons started, so it's most probably not a systemd issue. Thanks. -- Per Jessen, Zürich (7.4°C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Per Jessen wrote:
Per Jessen wrote:
Andrey Borzenkov wrote:
That is standard wrapper for any initscript that is using openSUSE API:
bor@opensuse:~> less /etc/rc.status ... # Check if the service is used under systemd but not started with if test -z "$SYSTEMD_NO_WRAP" && /bin/mountpoint -q /sys/fs/cgroup/systemd > /dev/null 2>&1 ; then ... case "$1" in status) SYSTEMD_NO_WRAP=1 "$0" "$1" ;; start|stop|reload|restart|try-restart|force-reload) echo "redirecting to systemctl ${SYSTEMCTL_OPTIONS} $1 ${_rc_base}" 1>&2 ;;
So if $PNAME refers to initscript, it will pass call through to systemctl.
That's the thing, $PNAME is not an initscript, it is the name of a binary in a separate path: /opt/hp/hp-snmp-agents/storage/bin/. $NAME is the textual description, e.g. "Health agent", $PNAME is "cmahealthd".
In the log messages, there is no name of what it is being redirected to: "redirecting to systemctl start ????"
Anyway, you gave me the idea of adding SYSTEMD_NO_WRAP=1 to the initscript, which led to the same problem, i.e. no daemons started, so it's most probably not a systemd issue. Thanks.
Just in case someone stumbles over this in the archives: Not a systemd issue, it was libcrypto.so.0.9.8 missing. Sorry about the noise. -- Per Jessen, Zürich (7.7°C) http://www.hostsuisse.com/ - dedicated server rental in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (2)
-
Andrey Borzenkov -
Per Jessen