No.
The test output is still messed. I figured it out, it receives 0 0 instead of 0 because nvram get runs twice.
The system log changed a bit - there's an occasional 0 in there somehow, the rest is rc=1.
#!/bin/sh
# copy original nvram executable to /tmp
cp /bin/nvram /tmp/_nvram
# create nvram wrapper that calls original nvram executable in /tmp
cat << 'EOF' > /tmp/nvram
#!/bin/sh
#set -x # comment/uncomment to disable/enable debug mode
run_cmd () {
local to
local start
local child
to=$1; shift
$@ & local child=$! start=0
usleep 1000
while [ "$(kill -0 $child >/dev/null 2>&1; printf "%s" "$?")" = "0" ]; do
start=$((start+1))
if [ $start -ge $to ]; then
kill -s 9 $child 2>/dev/null
return 1
fi
done
return 0
}
# make the new function accessible
run_cmd 1 /tmp/_nvram $@
rc=$?
#echo "$rc" >> /tmp/nvram.log
if [ "$rc" != "0" ]; then
#echo "$rc" >> /tmp/nvram.log
unset rc
usleep 1000
logger -t "nvram-override" "Couldn't execute nvram $@, exit status: $rc (124=timeout)"
run_cmd 1 /tmp/_nvram $@
rc=$?
logger -t "nvram-override" "Retried executing nvram $@, exit status: $rc"
fi
exit $rc
EOF
chmod +x /tmp/nvram
# replace nvram in /usr/sbin w/ nvram wrapper in /tmp
mount -o bind /tmp/nvram /bin/nvramI edited the if statement, make sure you are using the one where the if statement equals zero and not 1.
#!/bin/sh
# copy original nvram executable to /tmp
cp /bin/nvram /tmp/_nvram
# create nvram wrapper that calls original nvram executable in /tmp
cat << 'EOF' > /tmp/nvram
#!/bin/sh
#set -x # comment/uncomment to disable/enable debug mode
run_cmd () {
local to
local start
local child
to=$1; shift
$@ & local child=$! start=0
while [ "$(kill -0 $child >/dev/null 2>&1; printf "%s" "$?")" = "0" ]; do
usleep 1000
start=$((start+1))
if [ $start -ge $to ]; then
kill -s 9 $child 2>/dev/null
return 1
fi
done
return 0
}
# make the new function accessible
run_cmd 1 /tmp/_nvram $@
rc=$?
#echo "$rc" >> /tmp/nvram.log
if [ "$rc" != "0" ]; then
#echo "$rc" >> /tmp/nvram.log
unset rc
usleep 1000
logger -t "nvram-override" "Couldn't execute nvram $@, exit status: $rc (124=timeout)"
run_cmd 1 /tmp/_nvram $@
rc=$?
logger -t "nvram-override" "Retried executing nvram $@, exit status: $rc"
fi
exit $rc
EOF
chmod +x /tmp/nvram
# replace nvram in /usr/sbin w/ nvram wrapper in /tmp
mount -o bind /tmp/nvram /bin/nvramTry again, this time I edited where the placement of usleep was.
We need to optimize usleep in that placement then. Ideally how many ms does it take a nvram to run?Maybe you got the right order this time.
The loop shows the expected output, it's just moving slowly.
I'll let it run, take a break and report later.
usleep 1000 include sleep time of 1 millisecond. That's 15 sec for 15,000 nvram commands - negligible.unset rc after the first logger message, now it shows no value.
If you wait a little bit I got some optimizations that may make it better. A couple of performance mods and what not. Just hard to do it from my cellphone at work.
You're doing this from a cellphone?
The problem with loops are that they need their own form of rate limiting, usually a lock file of sorts. Otherwise there is no way to tell if you have stumbled onto a racing condition between competing processes spawned from the same loop. Hence one of the needs for optimization.
Correct, I got something in mind already. I have been brewing it in my mind for the past couple of hours.
This one should prove to be pretty interesting...
#!/bin/sh
# copy original nvram executable to /tmp
cp /bin/nvram /tmp/_nvram
# create nvram wrapper that calls original nvram executable in /tmp
cat << 'EOF' > /tmp/nvram
#!/bin/sh
#set -x # comment/uncomment to disable/enable debug mode
INTERVAL="100"
MAXCOUNT="3"
run_cmd () {
local to
local start
local child
# here as the number of tries required increases our usleep increases in case the sleep is not long enough.
to="$1"
to="$((to*INTERVAL))"; shift
$@ & local child="$!" start=0
usleep "$to"
while { [ "$(kill -0 $child >/dev/null 2>&1; printf "%s" "$?")" = "0" ] && [ "$start" -le "$to" ]; }; do
# to account for killing too soon, as the number of tries required increases our count requirement increases before we attempt to kill the process.
usleep 1
start="$((start+1))"
if [ $start -gt $to ]; then
kill -s 9 $child 2>/dev/null
return 1
fi
done
return 0
}
# make the new function accessible, on the first run we want to exit right away if successful.
i="1"
if { run_cmd "$i" /tmp/_nvram "$@"; }; then
exit 0
fi
# here we add an interval check and allow up to 3 retries.
while [ "$i" -le "$MAXCOUNT" ]; do
logger -t "nvram-override" "Waiting for NVRAM (retry attempt ${i}/${MAXCOUNT})"
usleep $INTERVAL
if { run_cmd "$i" /tmp/_nvram "$@"; }; then
exit 0
fi
i="$((i+1))"
done
logger -t "nvram-override" "NVRAM remained locked too long; continuing anyway."
exit 1
EOF
chmod +x /tmp/nvram
# replace nvram in /usr/sbin w/ nvram wrapper in /tmp
mount -o bind /tmp/nvram /bin/nvram
Well this one should speed things with your nvram right along because we will not get stuck waiting forever with this one before we attempt a retry.Oh, boy! Here we go again.
Btw, my router executes dozens of nvram operations per hour. I'll check later what are those.
No wonder it takes a day or two to get one stuck.
So you didn't see any of theOh, boy! Here we go again.
Btw, my router executes dozens of nvram operations per hour. I'll check later what are those.
No wonder it takes a day or two to get one stuck.
This last one ran like a blaze but didn't pick anything. Completely empty output, no messages in the log.
Welcome To SNBForums
SNBForums is a community for anyone who wants to learn about or discuss the latest in wireless routers, network storage and the ins and outs of building and maintaining a small network.
If you'd like to post a question, simply register and have at it!
While you're at it, please check out SmallNetBuilder for product reviews and our famous Router Charts, Ranker and plenty more!
