Defunct cfg_server Zombies

[SomeWhereOverTheRainBow]

Here is probably the best and safest option.

#!/bin/sh

i=1

pid_loc="$(basename "$0")"
procs="cfg_server"
if [ "$(pidof "$pid_loc" | wc -w)" -gt 1 ]; then exit; fi
while [ "$(pidof "$procs" | wc -w)" -le 3 ]; do sleep 1; done
for pid in "$(pidof "$procs")"; do
   if [ "$i" -gt "3" ]; then
      #echo $i
      #echo $pid
      /bin/kill -s 9 $pid
      /usr/bin/logger "Running /jffs/sbin/zomg_cfg_server"
   fi
   i=$((i + 1))
done

 

[Martinski]

it does I tested it. You can test it with simple echo inside the terminal. when dealing with numbers that is Pidof returns numbers and we are telling it to execute a legit command.

Hmmm, OK. I'll take your word for it.

However, when I ran the following simple test in my Ubuntu terminal:

#!/bin/bash

procs="$(pidof init)"
while [ "$(printf "%s" "$($procs)" | wc -w)" -le 1 ] ; do echo "IDs=[$procs]" ; break ; done

This is the result:

<MY_PROMPT>$ ./TESTING.sh
./TESTING.sh: line 4: 9: command not found
IDs=[9 8 1]

IOW, with the syntax $($procs) the conditional statement ends up trying to execute the 1st PID value of "9" as a command, which is obviously incorrect and not what it's intended.

 

[SomeWhereOverTheRainBow]

Hmmm, OK. I'll take your word for it.

However, when I ran the following simple test in my Ubuntu terminal:

#!/bin/bash

procs="$(pidof init)"
while [ "$(printf "%s" "$($procs)" | wc -w)" -le 1 ] ; do echo "IDs=[$procs]" ; break ; done

This is the result:

<MY_PROMPT>$ ./TESTING.sh
./TESTING.sh: line 4: 9: command not found
IDs=[9 8 1]

IOW, with the syntax $($procs) the conditional statement ends up trying to execute the 1st PID value of "9" as a command, which is obviously incorrect and not what it's intended.

Like I said having "pidof someprogram'" is what would be executed and the returned values will be read. Which is intended.

Obviously you can't execute a list of values though. But you can execute pidof in which the numbers get read back.

 

[SomeWhereOverTheRainBow]

Hmmm, OK. I'll take your word for it.

However, when I ran the following simple test in my Ubuntu terminal:

#!/bin/bash

procs="$(pidof init)"
while [ "$(printf "%s" "$($procs)" | wc -w)" -le 1 ] ; do echo "IDs=[$procs]" ; break ; done

This is the result:

<MY_PROMPT>$ ./TESTING.sh
./TESTING.sh: line 4: 9: command not found
IDs=[9 8 1]

IOW, with the syntax $($procs) the conditional statement ends up trying to execute the 1st PID value of "9" as a command, which is obviously incorrect and not what it's intended.

Look at my original post on the matter https://www.snbforums.com/threads/defunct-cfg_server-zombies.77650/page-2#post-750101

#!/bin/sh

i=1

pid_loc="pidof $(basename "$0")"
procs="pidof cfg_server"

if [ "$(printf "%s" "$($pid_loc)" | wc -w)" -gt 1 ]; then exit; fi
while [ "$(printf "%s" "$($procs)" | wc -w)" -le 3 ]; do sleep 1; done

for pid in $($procs); do
   if [ "$i" -gt "3" ]; then
      #echo $i
      #echo $pid
      /bin/kill -s 9 $pid
      /usr/bin/logger "Running /jffs/sbin/zomg_cfg_server"
   fi
   i=$((i + 1))
done

syntax is key.

pid_loc="pidof $(basename "$0")"

procs="pidof cfg_server"

 

[SomeWhereOverTheRainBow]

Your

procs="$(pidof init)"

should be written as

procs="pidof init"

 

[SomeWhereOverTheRainBow]

Hmmm, OK. I'll take your word for it.

However, when I ran the following simple test in my Ubuntu terminal:

#!/bin/bash

procs="$(pidof init)"
while [ "$(printf "%s" "$($procs)" | wc -w)" -le 1 ] ; do echo "IDs=[$procs]" ; break ; done

This is the result:

<MY_PROMPT>$ ./TESTING.sh
./TESTING.sh: line 4: 9: command not found
IDs=[9 8 1]

IOW, with the syntax $($procs) the conditional statement ends up trying to execute the 1st PID value of "9" as a command, which is obviously incorrect and not what it's intended.

just for giggles

RT-AX88U-C7C0:/tmp/home/root# it="pidof init"
RT-AX88U-C7C0:/tmp/home/root# echo "$($it)"
1

for the encore

RT-AX88U-C7C0:/tmp/home/root# pidof init
1

 

[SomeWhereOverTheRainBow]

Here is the logic. We are executing the command using $( ) instead of expanding the variable before it is time to call.

 

[Martinski]

...
procs="pidof cfg_server"

while [ "$(printf "%s" "$($procs)" | wc -w)" -le 3 ]; do sleep 1; done
...

Ah yes, that absolutely works. No question.

 

[SomeWhereOverTheRainBow]

Ah yes, that absolutely works. No question.

The problem with

procs="$(pidof cfg_server)"

as you pointed out initially is that it is called at the time we establish the variable. so the while loop runs infinitely.

without the $( ) in the initial variable, we don't call it since we don't expand.

putting $( ) only in the while loop means we call each time the loop runs.

 

[SomeWhereOverTheRainBow]

I bet we can even shorten it.

from

...
procs="pidof cfg_server"

while [ "$(printf "%s" "$($procs)" | wc -w)" -le 3 ]; do sleep 1; done
...

to

...
procs="pidof cfg_server"

while [ "$($procs | wc -w)" -le 3 ]; do sleep 1; done
...

 

[SomeWhereOverTheRainBow]

fully shortened would look like this.

#!/bin/sh

i=1

pid_loc="pidof $(basename "$0")"
procs="pidof cfg_server"

if [ "$($pid_loc | wc -w)" -gt 1 ]; then exit; fi
while [ "$($procs | wc -w)" -le 3 ]; do sleep 1; done

for pid in $($procs); do
   if [ "$i" -gt "3" ]; then
      #echo $i
      #echo $pid
      /bin/kill -s 9 $pid
      /usr/bin/logger "Running /jffs/sbin/zomg_cfg_server"
   fi
   i=$((i + 1))
done

 

[evolempt]

i know this is over a year old and I’m so, so sorry, but

I went so far as to read the BusyBox source code and can confidently report that the XT8 (RT-AX95Q) is cursed. Its implementation of `pidof` appears to actually be an implementation of `pidof -s`. It only returns the highest PID.

# ps wT | grep cfg_server | wc -l
1013
# pidof cfg_server
25795

and it is killing me. I’ll save the lengthy tirade on the webui being functionally unusable and just focus on needing to deal with tribble cfg_server processes.

My bash skills are middling at best. Implementing the zomg_cfg_server script has been an absolute ride on an XT8 (every line needed one thing or another changed), but I’m absolutely stymied at not having a list of PIDs for the `for` to iterate over. i’ve probably put 20 hours into my webui issues over the past couple of weeks and honestly it feels like i’ve read 80% of the SNB forums for one reason or another. Now that I have discovered/confirmed greedy cfg_server processes are the actual issue, i’m ready to come plead for pointers.

 

[garycnew]

i know this is over a year old and I’m so, so sorry, but

I went so far as to read the BusyBox source code and can confidently report that the XT8 (RT-AX95Q) is cursed. Its implementation of `pidof` appears to actually be an implementation of `pidof -s`. It only returns the highest PID.

# ps wT | grep cfg_server | wc -l
1013
# pidof cfg_server
25795

and it is killing me. I’ll save the lengthy tirade on the webui being functionally unusable and just focus on needing to deal with tribble cfg_server processes.

My bash skills are middling at best. Implementing the zomg_cfg_server script has been an absolute ride on an XT8 (every line needed one thing or another changed), but I’m absolutely stymied at not having a list of PIDs for the `for` to iterate over. i’ve probably put 20 hours into my webui issues over the past couple of weeks and honestly it feels like i’ve read 80% of the SNB forums for one reason or another. Now that I have discovered/confirmed greedy cfg_server processes are the actual issue, i’m ready to come plead for pointers.

@evolempt

If you have Entware installed, you can opkg install pidof for a more robust version.

Otherwise, you'll have to use ps, grep, and for to iterate through the cfg_server pids.

Kind Regards,


Gary

 

[SomeWhereOverTheRainBow]

i know this is over a year old and I’m so, so sorry, but

I went so far as to read the BusyBox source code and can confidently report that the XT8 (RT-AX95Q) is cursed. Its implementation of `pidof` appears to actually be an implementation of `pidof -s`. It only returns the highest PID.

# ps wT | grep cfg_server | wc -l
1013
# pidof cfg_server
25795

and it is killing me. I’ll save the lengthy tirade on the webui being functionally unusable and just focus on needing to deal with tribble cfg_server processes.

My bash skills are middling at best. Implementing the zomg_cfg_server script has been an absolute ride on an XT8 (every line needed one thing or another changed), but I’m absolutely stymied at not having a list of PIDs for the `for` to iterate over. i’ve probably put 20 hours into my webui issues over the past couple of weeks and honestly it feels like i’ve read 80% of the SNB forums for one reason or another. Now that I have discovered/confirmed greedy cfg_server processes are the actual issue, i’m ready to come plead for pointers.

if you can't manage to get a working pidof, you could always do:

#!/bin/sh

i=1

pid_loc="pidof $(basename "$0")"
procs="ps wT | awk '/cfg_server/{if( $0 !~ /awk/)printf "%s ", $1}'"

if [ "$($pid_loc | wc -w)" -gt 1 ]; then exit; fi
while [ "$($procs | wc -w)" -le 3 ]; do sleep 1; done

for pid in $($procs); do
   if [ "$i" -gt "3" ]; then
      #echo $i
      #echo $pid
      /bin/kill -s 9 $pid
      /usr/bin/logger "Running /jffs/sbin/zomg_cfg_server"
   fi
   i=$((i + 1))
done

 

[SomeWhereOverTheRainBow]

i know this is over a year old and I’m so, so sorry, but

I went so far as to read the BusyBox source code and can confidently report that the XT8 (RT-AX95Q) is cursed. Its implementation of `pidof` appears to actually be an implementation of `pidof -s`. It only returns the highest PID.

# ps wT | grep cfg_server | wc -l
1013
# pidof cfg_server
25795

and it is killing me. I’ll save the lengthy tirade on the webui being functionally unusable and just focus on needing to deal with tribble cfg_server processes.

My bash skills are middling at best. Implementing the zomg_cfg_server script has been an absolute ride on an XT8 (every line needed one thing or another changed), but I’m absolutely stymied at not having a list of PIDs for the `for` to iterate over. i’ve probably put 20 hours into my webui issues over the past couple of weeks and honestly it feels like i’ve read 80% of the SNB forums for one reason or another. Now that I have discovered/confirmed greedy cfg_server processes are the actual issue, i’m ready to come plead for pointers.

When you get a chance, run this command. I want to see what the output looks like.

ps wT | awk '/cfg_server/{if( $0 !~ /awk/)printf "%s\n", $0}'

I want to see if we can make our script more clever.

 

[garycnew]

When you get a chance, run this command. I want to see what the output looks like.

ps wT | awk '/cfg_server/{if( $0 !~ /awk/)printf "%s\n", $0}'

I want to see if we can make our script more clever.

@SomeWhereOverTheRainBow

I was thinking something more along the lines of

ps wT | grep cfg_server | grep -v grep | grep -ioE "^\s*([0-9]+)\s+"

Great to see you're still around.

 

[evolempt]

@evolempt

If you have Entware installed, you can opkg install pidof for a more robust version.

Otherwise, you'll have to use ps, grep, and for to iterate through the cfg_server pids.

Kind Regards,


Gary

never even crossed my mind. I generally avoid replacing packages like that but i’m absolutely not a purist about it.

When you get a chance, run this command. I want to see what the output looks like.

ps wT | awk '/cfg_server/{if( $0 !~ /awk/)printf "%s\n", $0}'

I want to see if we can make our script more clever.

truly my lack of sed/awk skills are my greatest professional shame. This gets me the same output as ps wT | grep cfg_server does.

this however ps wT | awk '/cfg_server/{if( $0 !~ /awk/)printf "%s ", $1}' produces the output we would expect out of pidof.

I maybe fell down a rabbit hole today, but I have a functional script for my XT8.

#!/bin/sh


i=1
awk_pidof() { ps wT | awk '/\scfg_server/{if( $0 !~ /awk/)printf "%s ", $1}'; }

if [ "$(ps wT | grep $(basename "$0") | wc -l)" -gt 3 ]; then
  # has to be gt 2 because otherwise grep gets caught in the dragnet
  # gt 3 because… for some reason there are always 2 zomg processes?
  echo "Exiting zombie slayer because it thinks it's a duplicate process"
  exit;
fi

while [ "$(awk_pidof | wc -w)" -le 5 ]; do sleep 1; done

for pid in $(awk_pidof); do
   if [ "$i" -gt "5" ]; then
      # echo $i
      # echo "killing pid $pid"
      /bin/kill -s 9 $pid
      /usr/bin/logger "Zombie slayer killing pid $pid"
   fi
   i=$((i + 1))
done

 

[garycnew]

# gt 3 because… for some reason there are always 2 zomg processes?

@evolempt

It has been my observation that there are always 3 good cfg_server processes and anything greater than 3 cfg_server processes have a tendency to want to eat brains.

My router has been rock solid, since implementing the zomg_cfg_server script.

Glad you have a working solution.

Kind Regards,


Gary

 

[evolempt]

lol this thing is spawning processes almost faster than i can kill them, this cannot be the best way to deal with this problem.

 

[garycnew]

lol this thing is spawning processes almost faster than i can kill them, this cannot be the best way to deal with this problem.

@evolempt

The XT8 might have a higher number of good cfg_server threshold. You might try increasing the greater than to 4, 5, 6, etc, until finding the right good cfg_server balance?

Kind Regards,


Gary