X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=automate-default.sh;h=3e65fcf0cd477edc26f52a7512d584720882ce79;hb=6c38b5cef7bf12fa4ade23ae1c54b3491797fb1d;hp=b5508c13ffacb32f1eb29e6d0c5840478e91a375;hpb=0fabfc8dbe8f1f2c0d12397e1bc8c6ed686fb5ed;p=monitor.git diff --git a/automate-default.sh b/automate-default.sh index b5508c1..3e65fcf 100755 --- a/automate-default.sh +++ b/automate-default.sh @@ -1,7 +1,9 @@ #!/bin/bash +export PATH=$PATH:/usr/local/sbin:/usr/local/bin:/sbin:/bin:/usr/sbin:/usr/bin + # NOTE: Must be an absolute path to guarantee it is read. -INSTALLPATH=/usr/share/monitor-server/ +INSTALLPATH=/usr/share/monitor/ # Generate an 'sh' style file full of variables in monitor.conf $INSTALLPATH/shconfig.py > $INSTALLPATH/monitorconfig.sh source $INSTALLPATH/monitorconfig.sh @@ -54,74 +56,30 @@ if [ -z "$AGENT" ] ; then # if no agent is running, set it up. ssh-agent > ${MONITOR_SCRIPT_ROOT}/agent.sh source ${MONITOR_SCRIPT_ROOT}/agent.sh + ssh-add /etc/planetlab/myops_ssh_key.rsa ssh-add /etc/planetlab/debug_ssh_key.rsa ssh-add /etc/planetlab/root_ssh_key.rsa fi #TODO: should add a call to ssh-add -l to check if the keys are loaded or not. source ${MONITOR_SCRIPT_ROOT}/agent.sh +#${MONITOR_SCRIPT_ROOT}/checksync.py $DATE || : +${MONITOR_SCRIPT_ROOT}/syncwithplc.py $DATE || : +service plc restart monitor -echo "Performing Findbad Nodes" +echo "Performing FindAll Nodes" ######################### # 1. FINDBAD NODES -rm -f ${MONITOR_DATA_ROOT}/production.findbad2.pkl -${MONITOR_SCRIPT_ROOT}/findbad.py --increment --cachenodes --debug=0 --dbname="findbad2" $DATE || : -cp ${MONITOR_DATA_ROOT}/production.findbad2.pkl ${MONITOR_DATA_ROOT}/production.findbad.pkl +${MONITOR_SCRIPT_ROOT}/findall.py $DATE || : ps ax | grep BatchMode | grep -v grep | awk '{print $1}' | xargs -r kill || : - -echo "Performing Findbad PCUs" -######################### -# 2. FINDBAD PCUS -rm -f ${MONITOR_DATA_ROOT}/production.findbadpcus2.pkl -${MONITOR_SCRIPT_ROOT}/findbadpcu.py --increment --refresh --debug=0 --dbname=findbadpcus2 $DATE || : -cp ${MONITOR_DATA_ROOT}/production.findbadpcus2.pkl ${MONITOR_DATA_ROOT}/production.findbadpcus.pkl # clean up stray 'locfg' processes that hang around inappropriately... ps ax | grep locfg | grep -v grep | awk '{print $1}' | xargs -r kill || : -#echo "Generating web data" -# badcsv.txt -#${MONITOR_SCRIPT_ROOT}/printbadcsv.py | grep -v loading | tr -d ' ' > badcsv.txt -#cp badcsv.txt /plc/data/var/www/html/monitor/ -#${MONITOR_SCRIPT_ROOT}/showlatlon.py | head -9 | awk 'BEGIN {print ""} { print ""} END{print "
", $0, "
"}' | sed -e 's\|\\g' > /plc/data/var/www/html/monitor/regions.html - -echo "Performing uptime changes for sites, nodes, and pcus" -######################## -# 3. record last-changed for sites, nodes and pcus. -${MONITOR_SCRIPT_ROOT}/sitebad.py --increment || : -${MONITOR_SCRIPT_ROOT}/nodebad.py --increment || : -${MONITOR_SCRIPT_ROOT}/pcubad.py --increment || : - -echo "Converting pkl files to phpserial" -######################### -# 4. convert pkl to php serialize format. -${MONITOR_SCRIPT_ROOT}/pkl2php.py -i findbadpcus2 -o findbadpcus -for f in act_all plcdb_hn2lb ; do - if [ -f ${MONITOR_DATA_ROOT}/production.$f.pkl ]; then - ${MONITOR_SCRIPT_ROOT}/pkl2php.py -i $f -o $f - else - echo "Warning: ${MONITOR_DATA_ROOT}/production.$f.pkl does not exist." - fi -done -${MONITOR_SCRIPT_ROOT}/pkl2php.py -i findbad -o findbadnodes -#${MONITOR_SCRIPT_ROOT}/pkl2php.py -i ad_dbTickets -o ad_dbTickets -#${MONITOR_SCRIPT_ROOT}/pkl2php.py -i idTickets -o idTickets - -echo "Archiving pkl files" -######################### -# Archive pkl files. -for f in findbad act_all findbadpcus l_plcnodes site_persistflags node_persistflags pcu_persistflags ; do - if [ -f ${MONITOR_DATA_ROOT}/production.$f.pkl ] ; then - cp ${MONITOR_DATA_ROOT}/production.$f.pkl ${MONITOR_ARCHIVE_ROOT}/`date +%F-%H:%M`.production.$f.pkl - else - echo "Warning: It failed to archive ${MONITOR_DATA_ROOT}/production.$f.pkl" - fi -done -echo "Running grouprins on all dbg nodes" -############################ -# 5. Check if there are any nodes in dbg state. Clean up afterward. -${MONITOR_SCRIPT_ROOT}/grouprins.py --mail=1 --reboot --nodeselect 'state=DOWN&&boot_state=(boot|rins|dbg|diag)' --stopselect "state=BOOT" || : -${MONITOR_SCRIPT_ROOT}/grouprins.py --mail=1 --reboot --nodeselect 'state=DEBUG&&boot_state=(rins|dbg|boot)' --stopselect 'state=BOOT' || : +${MONITOR_SCRIPT_ROOT}/policy.py $DATE || : +${MONITOR_SCRIPT_ROOT}/statistics/add-record.py || : +curl -s 'http://summer.cs.princeton.edu/status/tabulator.cgi?table=table_nodeview&formatcsv' > /var/lib/monitor/comon/$DATE.comon.csv || : cp ${MONITOR_SCRIPT_ROOT}/monitor.log ${MONITOR_ARCHIVE_ROOT}/`date +%F-%H:%M`.monitor.log +service plc restart monitor || : rm -f $MONITOR_PID