Increase "load average" limit on manual backups

saxico · January 7, 2021, 11:07am

Hi,

Can i increase the load average limit on manual backup and restore tasks?

For me is too dificult keep load average below 2, my servers always are on 3-4.

Thanks for your work!

eris · January 7, 2021, 11:53am

Permanent option: Increase core count on your server …

github.com

hestiacp/hestiacp/blob/5efa3ea8e0f580479c56378de9a12f2d8c61fbe6/bin/v-restore-user#L96-L110


# Checking load average
la=$(cat /proc/loadavg |cut -f 1 -d ' ' |cut -f 1 -d '.')
i=0
while [ "$la" -ge "$BACKUP_LA_LIMIT" ]; do
    echo -e "$(date "+%F %T") Load Average $la"
    sleep 60
    if [ "$i" -ge "15" ]; then
        la_error="LoadAverage $la is above threshold"
        echo "Error: $la_error" |$SENDMAIL -s "$subj" $email $notify
        sed -i "/ $user /d" $HESTIA/data/queue/backup.pipe
        check_result $E_LA "$la_error"
    fi
    la=$(cat /proc/loadavg |cut -f 1 -d ' ' |cut -f 1 -d '.')
    (( ++i))
done

This is the check

saxico · January 11, 2021, 9:21am

But if i change that file it will work for backups? Or only for restoring?

fabiodeoliveira · February 5, 2021, 4:03am

After searching a lot, I found a solution:

Edit the file:
/usr/local/hestia/func/main.sh

I switched this line:
BACKUP_LA_LIMIT=cat /proc/cpuinfo | grep processor | wc -l

For this
BACKUP_LA_LIMIT=30

Using the command v-backup-user username, the backup (zstd) was created in 0.5s.

Whoever has this problem, here is the solution.

g1naveen · May 3, 2022, 10:05pm

Just what I was searching for!
Saved my day…