Tip

I was doing a lot of changes to my old posts, switched to HTTPS, etc. Sometimes it was useful to change some particular text in all my old posts at a time, but there is no such feature in WordPress. But WordPress runs on MySQL and I could use SQL query to update such posts. Make backup - it’s not required but strongly advised 😃 Now use this query as template to replace in place whatever you need: UPDATE wp_posts SET post_content = REPLACE(post_content, " ", ""); You should see something like: ...

From few days I have access to UPC’s www.horizon.tv  external link platform - until now it was useless on Linux. But there is Pipelight that will use Wine to emulate Silverlight on Linux and it’s working pretty well - you’re just few commands away from achieving that: # stop browser killall firefox # remove old version if you have it sudo apt-get remove pipelight Now configure repos and install packages: ...

I was looking at backup task running on my desk and saw that it’s spending a lot of time on ~/.local/share/zeitgeist directory. I checked and it had 4.6GB: du -sh ~/.local/share/zeitgeist/* 118M activity.sqlite 44M activity.sqlite.bck 32K activity.sqlite-shm 4,4G activity.sqlite-wal 311M fts.index WTF? Fortunately I found here  external link that I could easily delete some of this: zeitgeist-daemon --quit Now check that it’s not running: ...

Sometime you need to make quick and dirty image backup of VM running on XenServer and this post is about such case 😃 List machines: xl list Name ID Mem VCPUs State Time(s) Domain-0 0 4066 8 r----- 3526567.3 webfront1.example.com 1 4096 4 r----- 3186487.2 webfront2.example.com 2 2048 2 -b---- 920408.2 Now you may export one: xe vm-export vm=webfront1.example.com filename=/srv/backup/webfront.xva Export succeeded You may also use uuid for that - list machines with xe vm-list (best with less) and then: ...

Sometimes deployment process or other heavy task may cause some Nagios checks to rise below normal levels and bother admin1. If this is expected and you want to add downtime on host/service during this task you may use this script: #!/bin/bash function die { echo $1; exit 1; } if [[ $# -eq 0 ]] ; then die "Give hostname and time in minutes as parameter!" fi if [[ $# -eq 1 ]] ; then MINUTES=15 else MINUTES=$2 fi HOST=$1 NAGURL=http://nagios.example.com/nagios/cgi-bin/cmd.cgi USER=nagiosuser PASS=nagiospassword SERVICENAME=someservice COMMENT="Deploying new code" export MINUTES echo "Scheduling downtime on $HOST for $MINUTES minutes..." # The following is urlencoded already STARTDATE=`date "+%d-%m-%Y %H:%M:%S"` # This gives us the date/time X minutes from now ENDDATE=`date "+%d-%m-%Y %H:%M:%S" -d "$MINUTES min"` curl --silent --show-error \ --data cmd_typ=56 \ --data cmd_mod=2 \ --data host=$HOST \ --data-urlencode "service=$SERVICENAME" \ --data-urlencode "com_data=$COMMENT" \ --data trigger=0 \ --data-urlencode "start_time=$STARTDATE" \ --data-urlencode "end_time=$ENDDATE" \ --data fixed=1 \ --data hours=2 \ --data minutes=0 \ --data btnSubmit=Commit \ --insecure \ $NAGURL -u "$USER:$PASS"| grep -q "Your command request was successfully submitted to Nagios for processing." || die "Failed to con tact nagios"; echo Scheduled downtime on nagios from $STARTDATE to $ENDDATE Threat this script as template with some tips: ...

From the first moment I heard about Let’s Encrypt I liked it and wanted to use it as fast as possible. But the more I read how they want to implement it, the more I dislike it. Current project with automatic configuration is not what I want to use at all. I have many very complicated configs and I do not trust such tools enough to use them. I like UNIX’s single purpose principle, tools should do one thing and do it well - nothing more. ...

Lately I had a lot of brute force attacks on my WordPress blog. I used basic auth to /wp-admin part in nginx configuration to block this and as a better solution I wan’t to block source IPs at all on firewall. To do this, place this filter code in /etc/fail2ban/filter.d/wp-login.conf: # WordPress brute force wp-login.php filter: # # Block IPs trying to authenticate in WordPress blog # # Matches e.g. # 178.218.54.109 - - [31/Dec/2015:10:39:34 +0100] "POST /wp-login.php HTTP/1.1" 401 188 "-" "Mozilla/5.0 (Windows NT 6.0; rv:34.0) Gecko/20100101 Firefox/34.0" # [Definition] failregex = ^<HOST> .* "POST /wp-login.php ignoreregex = Then edit your /etc/fail2ban/jail.local and add: ...

I have some Ansible roles to configure my vps, Raspberry Pi, etc. I like to test them before I broke something on my real, not clustered machines - I use Vagrant for that. But with it I had one problem - in playbooks I define hosts as groups of severs ex. web for my vps: Example Ansible playbook - hosts: web gather_facts: True sudo: True ... But testing machine wasn’t in this group and when I run vagrant I could only see: ...

Normally you want dynamic content to be fresh and not cacheable. But sometimes it may be useful to cache it, like when you have website behind reverse proxy1. To do this try something like this: Add new Apache conf file <filesmatch "\.(php|cgi|pl)$"> Header unset Pragma Header unset Expires Header set Cache-Control "max-age=3600, public" </filesmatch> http://www.askapache.com/htaccess/speed-up-your-site-with-caching-and-cache-control.html  external link  ↩︎ ...

It will happen from time to time, that you’re on alien machine and have to brutally update things in db without knowing credentials. Example is for root (quite secure candidate to change because it shouldn’t be used in app 😃 ) but will work for any user. shutdown db service mysql stop create text file with command like this (update user accordingly) ex. in /tmp/pwchange.txt SET PASSWORD FOR "root"@"localhost" = PASSWORD("HereYourNewPassword"); start mysqld with --init-file param mysqld_safe --init-file=/tmp/pwchange.txt sometimes you may require to point configuration file ex. --defaults-file=/etc/mysql/my.cnf ...