I'm much more inclined to create a script to perform the simple steps to
compile, than to spend a couple of hours (at least, that's what it takes
me) to package an RPM that will not be necessarily correct for all
systems. I do, however, agree that a service start/stop script is
useful. That said, I add a couple of lines in the /etc/rc.d/rc.local
file to:
cat the contents of ~ldmd.pid
Remove that file
perform, as ldm, an ldmadmin start
trap an error
remake the queue if necessary
start ldm
That could be modified or cleaned up, but it works on all my LDM systems...
gerry
Art Wildman wrote:
Hi folks,
1. Looking for an LDM RHEL-4+ RPM binary or rpm.spec files? If not, I
may be interested in building one for i386 (advice welcome).
2. Does anyone have a ldmd init script for RHEL that can support some of
the advanced RH Linux features listed below (chkconfig, service, PID
files & locks).
3. Are there any other gotchas (like syslog) running LDM at Boot as a
service?
I was following the linux-HA Cluster thread & noticed there are some LDM
Docs referencing clusters, but most are geared toward Hi-Performance
Clusters, not Hi-Availability (HA) Clusters. Some configs & principles
(IP Virtual Server) may translate though. For HA clusters, LDM can be
started & stopped via init-scripts in /etc/rc.d/init.d or
/etc/ha.d/haresources as Richard noted.
ConfiguringHeartbeat - Linux HA
http://www.linux-ha.org/ConfiguringHeartbeat
LDM Clusters
http://www.unidata.ucar.edu/software/ldm/ldm-6.6.4/basics/cluster.html
LDM at Boot - Init Scripts (this should be included in the
ldm-binary.rpm specific to RHELvX)
http://www.unidata.ucar.edu/software/ldm/ldm-6.6.4/basics/configuring.html#boot
http://www.unidata.ucar.edu/software/ldm/ldm-6.6.4/ldmConfigInstallList.html
ftp://ftp.unidata.ucar.edu/pub/ldm/scripts/linux/ldmd
Comparing it to a simple init script like /etc/rc.d/init.d/atd & we may
be able to improve it some & use RHEL service commands, locks & pstools
(pgrep, pkill...) to manage the service like:
# chkconfig --level 345 ldmd on
# service ldmd restart
# service ldmd status
# head /etc/rc.d/init.d/atd | grep chkconfig
# chkconfig: 345 95 5
Note: the chkconfig: line determines runlevels & the priority of kill
and start scripts (man chkconfig)
# service crond status
crond (pid 1117) is running...
# cat /var/run/crond.pid
1117
# ls -l /var/lock/subsys/crond
-rw-r--r-- 1 root root 0 Jan 5 17:26
/var/lock/subsys/crond
This may help shutdown & restart the ldmd safely during failover in an
HA cluster.
-Art@JAX
--
Gerry Creager -- gerry.creager@xxxxxxxx
Texas Mesonet -- AATLT, Texas A&M University
Cell: 979.229.5301 Office: 979.458.4020 FAX 979.862.3983
MAIL: AATLT, 3139 TAMU
Physical: 1700 Research Parkway, Suite 160,
College Station, TX 77843-3139
