pob
/
Proxmox_Scripts
mirror of https://github.com/tteck/Proxmox
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update monitor-all.sh

Browse Source 
Sorry before my last request.
I had completely overlooked the syntax.
I have now corrected the syntax and checked the script on a test instance. I hope I have not overlooked anything.

I have extended the script and fixed some bugs. The most important one:

    The detection of whether "onboot" should be started was defective.
    The LCX or VM containers are restarted a maximum of 3 times.
    The detection of the IP address had a bug, the variable was not reset, so if none was found, the last determined one was used.
    I have improved the log output.
    Additionally I added an email output in case of restarts. Email is disabled by default.
pull/2080/head
FrodoVDR 2 years ago committed by GitHub
parent 6c643e39d1
commit 0f2ff36d14
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 87 additions and 38 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Unified View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 125
      misc/monitor-all.sh

125
misc/monitor-all.sh
Unescape View File

@ -26,67 +26,116 @@ while true; do
done done
echo '#!/usr/bin/env bash echo '#!/usr/bin/env bash
# Maximum number of restarts
maxrestartcount=3
# Emails for restarts
mailaddress='\''admin@domain.tld'\''
mail="false"
# Search string for IP address, here more precise selection, IP addresses can be excluded.
searchip='\''192\.168\.|10\.'\''
# Read excluded instances from command line arguments # Read excluded instances from command line arguments
excluded_instances=("$@") excluded_instances=("$@")
echo "Excluded instances: ${excluded_instances[@]}" echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): Excluded instances: ${excluded_instances[@]}"
while true; do while true; do
for instance in $(pct list | awk '\''{if(NR>1) print $1}'\''; qm list | awk '\''{if(NR>1) print $1}'\''); do for vmid in $(pct list | awk '\''{if(NR>1) print $1}'\''; qm list | awk '\''{if(NR>1) print $1}'\'')
# Skip excluded instances do
if [[ " ${excluded_instances[@]} " =~ " ${instance} " ]]; then IP=
echo "Skipping $instance because it is excluded" skip="false"
continue if [ -f /tmp/$vmid.count ] ; then
count=$(cat /tmp/$vmid.count)
else
count=0
fi fi
# Determine the type of the instance (container or virtual machine) # Determine the type of the instance (container or virtual machine)
if pct status $instance >/dev/null 2>&1; then if pct status $vmid >/dev/null 2>&1; then
# It is a container # It is a container
config_cmd="pct config" config_cmd="pct config"
IP=$(pct exec $instance ip a s dev eth0 | awk '\''/inet / {print $2}'\'' | cut -d/ -f1) test=$(pct status $vmid | grep -q "status: running")
if [ $? -eq 0 ] ; then
IP=$(pct exec $vmid ip a s dev eth0 | awk '\''/inet / {print $2}'\'' | cut -d/ -f1)
fi
else else
# It is a virtual machine # It is a virtual machine
config_cmd="qm config" config_cmd="qm config"
IP=$(qm guest cmd $instance network-get-interfaces | egrep -o "([0-9]{1,3}\.){3}[0-9]{1,3}" | grep -E "192\.|10\." | head -n 1) test=$(qm status $vmid | grep -q "status: running")
if [ $? -eq 0 ] ; then
IP=$(qm guest cmd $vmid network-get-interfaces | egrep -o "([0-9]{1,3}\.){3}[0-9]{1,3}" | grep -E "$searchip" | head -n 1)
fi
fi fi
name=$($config_cmd $vmid | grep name: | awk '\''{ print $2 }'\'')
# Skip instances based on onboot and templates # Skip instances based on onboot and templates
onboot=$($config_cmd $instance | grep onboot | grep -q "onboot: 0" && echo "true" || echo "false") test=$($config_cmd $vmid | grep "onboot" | awk '\''{ print $2 }'\'')
template=$($config_cmd $instance | grep template | grep -q "template:" && echo "true" || echo "false") if [ "$test" == "1" ] ; then
onboot="true"
if [ "$onboot" == "true" ]; then else
echo "Skipping $instance because it is set not to boot" onboot="false"
continue fi
elif [ "$template" == "true" ]; then test=$($config_cmd $vmid | grep "template:" | awk '\''{ print $2 }'\'')
echo "Skipping $instance because it is a template" if [ "$test" == "1" ] ; then
continue template="true"
else
template="false"
fi fi
# Ping the instance if [ "$onboot" == "false" ]; then
if ! ping -c 1 $IP >/dev/null 2>&1; then echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): Skipping $vmid $name because it is set not to boot"
# If the instance can not be pinged, stop and start it skip="true"
if pct status $instance >/dev/null 2>&1; then fi
# It is a container if [ "$template" == "true" ]; then
echo "$(date): CT $instance is not responding, restarting..." echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): Skipping $vmid $name because it is a template"
pct stop $instance >/dev/null 2>&1 skip="true"
sleep 5 fi
pct start $instance >/dev/null 2>&1 if [ "$skip" == "false" ] ; then
else # Ping the instance
# It is a virtual machine if ! ping -c 1 $IP >/dev/null 2>&1; then
if qm status $instance | grep -q "status: running"; then if [ $count -le $maxrestartcount ] ; then
echo "$(date): VM $instance is not responding, restarting..." count=$((count + 1))
qm stop $instance >/dev/null 2>&1 # If the instance can not be pinged, stop and start it
sleep 5 if pct status $vmid >/dev/null 2>&1; then
# It is a container
echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): CT $vmid $name is not responding, restarting..."
if [ "$mail" == "true" ] ;then
echo "CT $vmid $name is not responding, restarting" | mail -s "$(date +'\''%Y-%m-%d %H:%M:%S'\''): $(hostname) - $name" $mailaddress
fi
pct stop $vmid >/dev/null 2>&1
sleep 5
pct start $vmid >/dev/null 2>&1
else
# It is a virtual machine
test=$(qm status $vmid | grep -q "status: running")
if [ $? -eq 0 ] ; then
echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): VM $vmid $name is not responding, restarting..."
if [ "$mail" == "true" ] ;then
echo "VM $vmid $name is not responding, restarting" | mail -s "$(date +'\''%Y-%m-%d %H:%M:%S'\''): $(hostname) - $name" $mailaddress
fi
qm stop $vmid >/dev/null 2>&1
sleep 5
else
echo "$(date +'%Y-%m-%d %H:%M:%S'): VM $vmid $name is not running, starting..."
fi
qm start $vmid >/dev/null 2>&1
echo "$count" > /tmp/$vmid.count
fi
else else
echo "$(date): VM $instance is not running, starting..." echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): VM $vmid $name max restart count $count reached"
if [ "$mail" == "true" ] ;then
echo "VM $vmid $name max restart count $count reached" | mail -s "$(date +'\''%Y-%m-%d %H:%M:%S'\''): $(hostname) - $name" $mailaddress
fi
fi fi
qm start $instance >/dev/null 2>&1 else
echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): CT/VM $vmid $name with ip $IP is pingable..."
echo "0" > /tmp/$vmid.count
fi fi
fi fi
done done
# Wait for 5 minutes. (Edit to your needs) # Wait for 5 minutes. (Edit to your needs)
echo "$(date): Pausing for 5 minutes..." echo "$(date +'\''%Y-%m-%d %H:%M:%S'\''): Pausing for 5 minutes..."
sleep 300 sleep 300
done >/var/log/ping-instances.log 2>&1' >/usr/local/bin/ping-instances.sh done >/var/log/ping-instances.log 2>&1' >/usr/local/bin/ping-instances.sh
touch /var/log/ping-instances.log touch /var/log/ping-instances.log

Write Preview
Loading…
Cancel
Save