HOWTO: fix your wordpress install after a system update – main page shows only php code

Background: we had a very big disconnection and the local WP installation didn’t come back after all the other systems came back. I have access to the website (so the apache2 service is running) but the main site as well as the admin site show only php text code. I tried rebooting the computer (a virtual machine) without luck. A very simple solution this one has. As usual, the solution is in StackOverflow.

sudo apt install php7.0 libapache2-mod-php7.0 
php7.0-mysql php7.0-curl php7.0-json

I restart then my apache2 and my website comes back. Fiu! Have a nice weeked, or see you tomorrow for the Mars discussion post…

A jupyter notebook in a docker: Jupyter Docker Stacks

Again I’m with python. I want to run as package independent as possible, so I want to run the external python apps in a docker. If we have a jupyter notebook, we can easily run it this way if we have it in a “test” folder. Note that the “test” folde needs to be readable for everyone (777) or you’ll get this error :

tornado.web.HTTPError: HTTP 403: Forbidden

Anyway, this is what I get when I run my docker mapping the test folder:

# > docker run -p 8888:8888 -v "${PWD}"/test:/home/jovyan/work 
jupyter/scipy-notebook
Entered start.sh with args: jupyter lab
Executing the command: jupyter lab
[I DATE ServerApp] jupyter_server_terminals 
| extension was successfully linked.
[I DATE ServerApp] jupyterlab 
| extension was successfully linked.
[W DATE NotebookApp] 'ip' has moved from NotebookApp 
to ServerApp. This config will be passed to ServerApp. 
Be sure to update your config before our next release.
[W DATE NotebookApp] 'ip' has moved from NotebookApp 
to ServerApp. This config will be passed to ServerApp. 
Be sure to update your config before our next release.
[I DATE ServerApp] nbclassic 
| extension was successfully linked.
[I DATe ServerApp] Writing Jupyter server cookie secret 
to /home/jovyan/.local/share/jupyter/runtime/jupyter_cookie_secret
[I DATE ServerApp] notebook_shim 
| extension was successfully linked.
[I DATE ServerApp] notebook_shim 
| extension was successfully loaded.
[I DATE ServerApp] jupyter_server_terminals 
| extension was successfully loaded.
[I DATE LabApp] JupyterLab extension loaded from
/opt/conda/lib/python3.10/site-packages/jupyterlab
[I DATE LabApp] JupyterLab application directory 
is /opt/conda/share/jupyter/lab
[I DATE ServerApp] jupyterlab 
| extension was successfully loaded.
[I DATE ServerApp] nbclassic 
| extension was successfully loaded.
[I DATE ServerApp] Serving notebooks 
from local directory: /home/jovyan
[I DATE ServerApp] Jupyter Server 2.1.0 is running at:
[I DATE ServerApp] http://XXX:8888/lab?token=XXX
[I DATE ServerApp] or http://127.0.0.1:8888/lab?token=XXX
[I DATE ServerApp] Use Control-C to stop this server 
and shut down all kernels (twice to skip confirmation).
[C DATE ServerApp] 

To access the server, open this file in a browser:
file:///home/jovyan/.local/share/jupyter/runtime/jpserver-7-open.html
Or copy and paste one of these URLs:
http://XXX:8888/lab?token=XXX
or http://127.0.0.1:8888/lab?token=XXX

The Jupyter Docker stack documentation is available here. Happy docking!

A python dash web multiplot example reading CSV data

I previously used plotty to get a python plot on a website. It’s time to elaborate. Now we want to have two plots with data from a CSV file. It’s not as complicated as it sounds. Allow me to show and explain.

import dash
from dash import dcc, html
from dash.dependencies import Input, Output
import plotly.express as px
from plotly.subplots import make_subplots
import plotly.graph_objects as go
import pandas as pd

app = dash.Dash(__name__)

df = pd.read_csv('data3colums.csv')
##columns are named here and you need to use these names
df.columns=['Time','data','Size(T)']
fig = px.line(df, x="Time",y="data",markers=True)
fig2 = px.line(df, x="Time",y="Size(T)",markers=True)
app.layout = html.Div(children=[
     # All elements from the top of the page
     html.Div([
     html.H1(children='Time series one'),
     html.Div(children='''Title one'''),
         dcc.Graph(id='graph1',figure=fig), 
     ]),
     # New Div for the new 'row' of the page
     html.Div([
     html.H1(children='Time Series two'),
     html.Div(children='''Title two'''),
         dcc.Graph(id='graph2',figure=fig2), 
     ]),
])
if __name__ == '__main__':
app.run_server(debug=True,host='IP-OF-MY-MACHINE')

If you want to know more about how to plot data from CSV here you have the plotly documetation. It can be useful also to know about plotting time series, since at least in my CSV I have time tags. And here you have the stackoverflow post about this topic. Because we can’t forget about stackoverflow.

EDIT: this post has been backdated to “yesterday” because I forgot to publish it, so another post may come later. Hopefully…this will mean I had time for my things 😉

Error: cannot find -lmkl_intel_lp64 while compiling on Ubuntu 20.04

Ubuntu 20 is escalating positions as the heir of CentOS 7, and I started experiencing compatibility issues. I was compiling relion with the next options

cmake -DAMDFFTW=ON -DCUDA_ARCH=86 -DCUDA=ON -DCudaTexture=ON 
-DFORCE_OWN_FLTK=ON -DCMAKE_CXX_COMPILER=g++ -DCMAKE_C_COMPILER=gcc
-DMPI_C_COMPILER=/usr/bin/mpicc 
-DMPI_C_LIBRARIES=/usr/lib/x86_64-linux-gnu/openmpi/lib/ 
-DMPI_C_INCLUDE_PATH=/usr/lib/x86_64-linux-gnu/openmpi/include/ 
-DCUDA_ARCH=61 -DCUDA=ON -DCudaTexture=ON -DMKLFFT=ON 
-DFORCE_OWN_FLTK=ON -DGUI=ON 
-DCMAKE_INSTALL_PREFIX=/XXX/relion_local/ 
-D CMAKE_BUILD_TYPE=Release ..

when I found this error

[ 59%] Linking CXX executable ../../bin/relion_tomo_tomo_ctf
/usr/bin/ld: cannot find -lmkl_intel_lp64
/usr/bin/ld: cannot find -lmkl_sequential
/usr/bin/ld: cannot find -lmkl_core
collect2: error: ld returned 1 exit status
make[2]: *** [src/apps/CMakeFiles/tomo_ctf.dir/build.make:102: 
bin/relion_tomo_tomo_ctf] Error 1
make[1]: *** [CMakeFiles/Makefile2:352: 
src/apps/CMakeFiles/tomo_ctf.dir/all] Error 2
make: *** [Makefile:130: all] Error 2

I’m using the default package install locations, no modules or anything fancy. So it looks like some libraries are missing. Of course I try to find them with ldconfig -p | grep ‘mkl’ but they are obviously not there. Let’s install them:

# apt-get install libmkl*intel* libmkl*se* libmkl*core*
Reading package lists... Done
Building dependency tree 
Reading state information... Done
Note, selecting 'libmkl-intel-thread' for glob 'libmkl*intel*'
Note, selecting 'libmkl-blacs-intelmpi-ilp64' for glob 'libmkl*intel*'
Note, selecting 'libmkl-intel-ilp64' for glob 'libmkl*intel*'
Note, selecting 'libmkl-blacs-intelmpi-lp64' 
for glob 'libmkl*intel*'
Note, selecting 'libmkl-intel-lp64' for glob 'libmkl*intel*'
Note, selecting 'libmkl-sequential' for glob 'libmkl*se*'
Note, selecting 'libmkl-core' for glob 'libmkl*core*'
Note, selecting 'libmkl-cdft-core' for glob 'libmkl*core*'
The following additional packages will be installed:
libmkl-def libmkl-locale libmkl-vml-def
The following NEW packages will be installed:
libmkl-blacs-intelmpi-ilp64 libmkl-blacs-intelmpi-lp64 
libmkl-cdft-core libmkl-core libmkl-def 
libmkl-intel-ilp64 libmkl-intel-lp64 
libmkl-intel-thread libmkl-locale 
libmkl-sequential libmkl-vml-def
0 upgraded, 11 newly installed, 0 to remove and 19 not upgraded.
Need to get 38,4 MB of archives.
After this operation, 204 MB of additional disk space will be used.
Do you want to continue? [Y/n] y

You may need to run apt-get install mklibs in addition. Of course it depends on what you have been doing with your system so far, since this will call some python libraries also. After getting those packages, make and make install run without errors. Time to ask the user to test it…

Python Import Error: can’t import name gcd from fractions

Deprecation is a big issue in python. I’m in need of Molecular Dynamic (MD) simulations tools. The error above comes from one tool I already posted about, called LipIDens, more specifically, it’s a complain thrown away by vermouth. Vermouth (for VERsatile, MOdular, and Universal Tranformation Helper) is also a drink and the python library that powers Martinize2. The vermouth source comes here. It is supposed to be used to apply transformation on molecular structures. Which means I don’t really know what it does! Anyway, my error reads

Installed 
/usr/local/lib/python3.9/site-packages/lipidens-1.0.0-py3.9.egg
Processing dependencies for lipidens==1.0.0
error: networkx 3.0 is installed but 
networkx~=2.0 is required by {'vermouth'}

What to do here? I found the solution and the explanation once more on StackOverflow. It’s very interesting to know that the the grammar for a mathematical library changed after Python 3.5. So then, why on the LipIDens documentation it is recommended to use a python above 3.9? I’m going to leave the answer to this question open (old developer environments with remnants or insufficient tests) and show you my solution. We install a specific python package. I choose pip to install it instead of conda because it goes to my python site-packages, which I personally consider a more elegant solution. Here you have my output:

bash-5.1# pip install networkx==2.5
Collecting networkx==2.5
Downloading networkx-2.5-py3-none-any.whl (1.6 MB)
|XXXXXXX| 1.6 MB 4.3 MB/s 
Requirement already satisfied: decorator>=4.3.0 in /usr/local/lib/python3.9/site-packages (from networkx==2.5) (5.1.1)
Installing collected packages: networkx
Attempting uninstall: networkx
Found existing installation: networkx 2.0
Uninstalling networkx-2.0:
Successfully uninstalled networkx-2.0
Successfully installed networkx-2.5
WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv
bash-5.1# python setup.py install

After the pip install over my local python I run again the LipIDens installer and it works. Another issue is to get meaningful results from the program! BTW, I decided to keep writing my bits thanks to some good feedback that I was missing before… thank you guys. I appreciate it.

WordPress multisite: more than one WP site in one server

I don’t have so much mood to write lately. I guess I’ve been busy thinking or something like that. Anyway, here I am again. (B)logging the last thing I did, that is to transform a single WP install onto a multisite install.I will be following this guide to arrive to the desired multisite configuration.

I start with an installed version (so no docker) on CentOS 7. I deactivate the plugins, then edit /var/www/html/wp-config.php as indicated.

Add this code before the /* That's all, stop editing! Happy blogging. */ line:

/* Multisite */
define('WP_ALLOW_MULTISITE', true);

I save the edits on the wp-config.php, and restart httpd service. Unfortunately I don’t get the Network Setup Menu as in the guide (maybe you get it) but I don’t worry about it and continue. After that I will regret, and I will need to edit various other files to achieve what I want. Here you have the details about the file modifications needed. But basicallyt I have installed phpMyAdmin also. For that I have installed and configured mod_rewrite for apache and fixed a package error problem with php 7.4 install on CentOS 7.9. In brief

yum-config-manager --disable 'remi-php*'
yum-config-manager --enable remi-php74

And of course, I have given permission to access phpmyadmin on this server. But at the end, as usual I manage to get what I want. Or should I say you managed to get what you wanted? 🧐. Because what I want is to write about what I think, not about what I do. And I’m not in the mood 😔. See you around…

HOWTO: Install WordPress in debian 11 (no docker)

Previously I have installed WP on CentOS 7.X, now I have tried in another system, Debian GNU/Linux 11 (bullseye). I have followed this tutorial from Cloud Infrastructure Services basically cut’n’copying until Step 5, replacing :

• user@yourdomain.com by my admin email address,
• yourdomain.com by my domain wptest
• “https://net.cloudinfrastructureservices.co.uk/var/www/html/yourdomain.com” on the /etc/apache2/sites-available/wptest.conf by simply “https://var/www/html/wptest”

In other words, replace the generic entries. I test the apache configuration

# apache2ctl configtest
Syntax OK

Then I go to http://IP_ADDRESS and create the root user, etc. Like it is written here, on the howto install WP on Debian. It was very easy, as it was for CentOS 😎! Note that in one of the tutorial we use an apache server, and in the last one it’s an NGINX Virtual Host. Happy wordpressing! 😁.

Virtual machine manager error: no connection driver available for qemu:///system on CentOS 7

I’m trying to find a nice full-sim environment on my dying CentOS 7.X system. It means not a docker neither a LXC solution but a full OS with IP and so on. As similar as the real thing but running on CentOS 7. I have a clean machine, and I remember qemu as the tool that does everything I want. So I install it, start the service, and call the GUI. Like this:

yum install qemu-kvm qemu-img virt-manager libvirt-daemon
systemctl start libvirtd
virt-manager &

The GUI pops up but it gives me the error above. That I solve in one of the ways described in this post. I have updated and enable the service without luck. So I go for missing packages.

yum -y install qemu-kvm qemu-img virt-manager \
libvirt libvirt-python python-virtinst \
libvirt-client virt-install virt-viewer

After that, no need to reboot, I get my GUI and I can start playing with VMs. I will report you my findings, if any 🧐. BONUS: another post about a similar issue.

ModuleNotFoundError: No module named ‘absl’ while using python

This is a very specific error. We have quite a mess-up python setup with multiple versions, network modules and local, a SLURM cluster, and the option to install your own, so it’s quite tricky to track the origin of a module. But it helps if I have two servers with the same kernel and packages, one of them the program runs, but not in the other. I do have logs, but they are not very clear. Message says

ModuleCmd_Load.c(213):ERROR:105: Unable to locate a modulefile for 'python-3.7.3'
Traceback (most recent call last):
File "/XXX/run_docker.py", line 22, in <module>
from absl import app
ModuleNotFoundError: No module named 'absl'

So what is happening here I believe is that the program is trying to load the module, it fails and goes to the local install. As an user I can install the missing absl-py module, but I can’t run it in this case because of the special process. What to do then? All the solutions on stackoverflow were not suitable, since I don’t know which python is getting what. But I have a hint: the program seems to be loading python 3.7.3, that is not the default. So I look for the python modules stored and I found them on the server one. A simple sync

@ server-one ## > rsync -av /usr/local/lib/python3.6/site-packages/ root@server-two:/usr/local/lib/python3.6/site-packages/

and then the “program” works. What is going on? It looks like python, since it didn’t manage to find the modules for the requested python version, took the closest ones available (3.6). But who knows? I’m not a python expert, I’m just passing by 😔.

A LimeSurvey in a docker: make it a quick and dirty

You may need to ask your colleagues about something at one point, I’m sure about it. But I’m sure about also that you don’t have time, neither hardware, to program an entire survey or to install the software in a specific server. Since I’m lazy also, as you know, I looked for a docker solution until I found one that does the job. The basic principles of a lime survey you get on this post. This will work if your docker station has a mysql database – you will be asked to connect to the database. If you don’t have, or you don’t want to use it, we need a compose with a mysql docker.

It is worth to mention all the docker images I found. I hope it saves you googling time! 😁😁. So here you have the crramirez limesurvey image, with a pseudo-docker compose. And here you have the martialblog limesurvey, with a lot of possible customisations. Me, I’m a simple person. I don’t want to play with customisation, I want to download a compose file and get the thing working out of the box. This goal you can achieve with the acspri limesurvey. In short (or TLTR) 😁😁😁: make a folder and copy this into a docker-compose.yml file.

version: '2'
services:
 limesurvey:
  image: acspri/limesurvey
  ports:
   - 8082:80
  environment:
    LIMESURVEY_DB_PASSWORD: example
    LIMESURVEY_ADMIN_USER: admin
    LIMESURVEY_ADMIN_PASSWORD: password
    LIMESURVEY_ADMIN_NAME: Lime Administrator
    LIMESURVEY_ADMIN_EMAIL: lime@lime.lime
  volumes:
    - ./plugins:/var/www/html/plugins
    - ./upload:/var/www/html/upload
    - ./config:/var/www/html/application/config
   mysql:
    image: mariadb
    environment:
      MYSQL_ROOT_PASSWORD: example

As usual, careful with the formatting if you cut’n’paste from above. Once you have this file in a folder in the computer with dockers up and running, just type

docker-compose up -d

And access to:

Frontend (what you send to the people)

http://localhost:8082/

Backend (to create the survey)

http://localhost:8082/index.php/admin

I’m not going to dig more, since now it’s time to write the survey. Don’t forget to check how to configure the default answers and how to export the results. If anyone is interested, that is not always the case 😩😩😩.