How to , and other stuff about linux, photo, php … Another blog just like others on internet

December 10, 2018

Your browser sent a request that this server could not understand

Filed under: Linux — Tags: , , , — admin @ 4:51 pm

Well today I am facing another problem.

Got this answer on a apache behind a proxy .

400 Bad Request
Bad Request
Your browser sent a request that this server could not understand.
Size of a request header field exceeds server limit.

So .. it appear the header that is received by webserver is bigger then normal .

Solution is to increase it .

LimitRequestFieldSize 32760
LimitRequestLine 32760

Please note that is loaded from first VirtualHost.

One solution is to tested with this

curl -v -H “CustomHeader: `printf ‘1%.0s’ {1..n}`” localhost

Where n can be replace with a value of for example 20000

 

April 25, 2018

apparmor=”DENIED” operation=”open” profile=”/usr/sbin/mysqld”

Filed under: Linux — Tags: , , — admin @ 12:07 pm

Today try to move the path of mysql to new one , because on ec2 I was left without space.

Well I have stopped the mysqld copy files modity the my.cnf with new datadir but when try to start .. well didn’t started .. After a little dmesg search see that on ubuntu is apparmor .

Try to make alias into file /etc/mysql/mysql.conf.d/mysqld.cnf  but didn’t resolve the problem

Find out that you have to edit also /etc/apparmor.d/usr.sbin.mysqld

and add

/db/data/mysql r

/db/data/mysql** rwk

After this mysqld started with new path.

December 8, 2017

delete large directory containing thousands of files

Filed under: Linux — Tags: , , , , , — admin @ 11:22 am

Well if you try to remove a directory with multiple files rm will take a long time also some load on that server .

One solution is to rsync a empty directory .

mkdir empty_dir
rsync -a –delete empty_dir/ yourdirectory/

Also a perl can be made but the rsync is easy for me to remeber .

Some more information about this you can find here

https://serverfault.com/questions/183821/rm-on-a-directory-with-millions-of-files/328305#328305

 

August 30, 2017

httpd with multiple ssl and password for every key automatically

Filed under: Linux — Tags: , , , , — admin @ 1:19 pm

Well to use this we must use the SSLPassPhraseDialog functionality of httpd – apache .

Basically into ssl.conf we put this line or modify the existing one into

SSLPassPhraseDialog exec:/path/script

Where script is read/execute only by root.

And the script is

 

#!/bin/perl
$server = $ARGV[0];
#print $server;
if ($server eq ‘www.example.com:443’ || $server eq ‘www.example2.com:443’ ) {
print ‘password one’;
} elsif ($server eq ‘example3.com’) {
print ‘Password two’;
}

 

September 22, 2015

centos /dev/md127 problem after reboot

Filed under: Linux — Tags: , , , , — admin @ 1:06 pm

Hello
Well today I want to explain about my raid experience.
I have to setup a hybrid server on hetzner. I setup the ssd on raid1 using installimage and those other hard drive I have create a raid from linux and mounted them on /raid1 directory .
However after reboot, my /dev/md4 disappear and a /dev/md127 appear.
To create the raid I have use this
fdisk /dev/sdc and /dev/sdd . Make a primary partition and set it up as fd software raid partition .

mdadm –create /dev/md4 –level=1 –raid-devices=2 /dev/sdc1 /dev/sdd1
mkfs.ext4 /dev/md4
mkdir /raid1
mount /dev/md4 /raid1

And insert this into fstab file
/dev/md4 /raid1 ext4 noatime,rw 0 0

So how to fix this ?
First I try to input some information into /etc/mdadm.conf but without luck . It appear that the linux read this late and the centos/ubuntu initalize this from initrd . So in order to fix this some steps have to be done .
After reboot try to fill the information into mdadm.conf with
ARRAY /dev/md/4 UUID=b3c33fe5:3b078681:e2776e37:4f9fd991
The UUID I have taken from

mdadm –detail /dev/md4

First unmount and stop the raid

mdadm –stop /dev/md127
After this assemble it again with your desire
mdadm –assemble /dev/md4 /dev/sdc1 /dev/sdd1

After this I have copy the
/boot/initramfs-2.6.32-573.3.1.el6.x86_64.img to a backup file ( initramfs-2.6.32-573.3.1.el6.x86_64.img-back ) in case something go wrong.

After this run
dracut –force

If is centos 5 or less you have to copy initrd file and to recreate use
mkinitrd -f -v /boot/initrd-$(uname -r).img $(uname -r)

On ubuntu you have to run
sudo update-initramfs -u

And restart the server.

Older Posts »

Powered by WordPress