https://bitsilla.com/wiki/ 2026-04-05T19:56:13+00:00 https://bitsilla.com/wiki/ https://bitsilla.com/wiki/lib/exe/fetch.php?media=wiki:dokuwiki.svg text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:cassandra_fix_upgrade_ubuntu&rev=1539079978&do=diff Upgrading Cassandra on Ubuntu 14.04 and fixing IllegalStateException Note that this recipe may also be used to reinstall cassandra! Upgrading Cassandra via apt-get to version 3.0.0 (from 2.x) gives the following barf: ERROR 08:40:42 Exception encountered during startup java.lang.IllegalStateException: One row required, 2 found at org.apache.cassandra.cql3.UntypedResultSet$FromResultSet.one(UntypedResultSet.java:84) ~[apache-cassandra-3.0.0.jar:3.0.0] at org.apache.cassandra.schema.LegacyS… text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:cassandra_install&rev=1539079978&do=diff Install Cassandra on Ubuntu 14.04 Installing via apt-get is always a good idea, as you keep getting goodies with a simple sudo apt-get upgrade. Unfortunately, Cassandra is not in Ubuntu's repository, and my Data Science students require it (for their assignment, nonetheless). Here's a recipe to make it work: text/html 2018-10-16T17:01:03+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:create_python_environment_for_study_unit_xxx&rev=1539709263&do=diff Create a Python Environment for Study-Unit XXX1234 This is relatively easy (especially if you are using Ubuntu). * Install the latest version of Anaconda. * Download the environment file from VLE for the study-unit you are attending. This is a file with extension text/html 2021-10-13T11:03:41+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:export_an_anaconda_environment_to_a_file&rev=1634123021&do=diff Export an Anaconda Environment to a File If you want others to have the exact same Python environment as yours truly, export your Anaconda environment with: conda env export | grep -v "^prefix" > somefile.yml You then just need to pass on the file ( text/html 2021-03-23T15:58:56+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:git_create_tag_from_branch&rev=1616515136&do=diff Create a Tag from a Branch using Git So you have realized your whole assignment repo. is incorrectly structured with dead-end commits in branches you will never get back to. The solution across many timelines is to use tags and not branches. So how do we rectify the situation? text/html 2019-05-23T08:18:59+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:jupyterhub_run&rev=1558599539&do=diff Run JupyterHub on Titan First you need a working/running jupyterhub installation (via conda, ideally). Then once done, you need to run JupyterHub, in a tmux session, with: jupyterhub --ip=10.249.45.45 --port=18888 --JupyterHub.spawner_class=sudospawner.SudoSpawner --Spawner.notebook_dir='/home/%U/notebooks' --JupyterHub.ssl_key=/home/jebej02/.jupyterhub/jupyterhub.key --JupyterHub.ssl_cert=/home/jebej02/.jupyterhub/jupyterhub.crt text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:linux_gnu_parallel&rev=1539079978&do=diff Use GNU parallel (to compress large files) GNU parallel is a great utility to take advantage of multiple cores. You can download the latest source from the bottom of this page. The installation is pretty straightforward also. Unzip, ./configure, make, and sudo make install text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:linux_insert_header_in_all_files&rev=1539079978&do=diff Insert the same header (file) in a number of files The trick here is to use cat and pipe to a tmp file which you then rename to something sensible. for f in chr*txt; do echo $f; cat header.txt $f > tmp; mv tmp $(basename $f .txt).csv; done Note that the above assumes the existence of the header in a file ( text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:linux_motd&rev=1539079978&do=diff Update MOTD Banner Assuming you want to add some fancy graphic at the top of the banner. Generate one using TAAG. * Edit the file /etc/update-motd.d/00-header * Using printf add the banner title. Remember to add newlines \n) at the end of each printf (good old C style), e.g. text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:linux_split_file_first_col&rev=1539079978&do=diff Split a file based on the first column of that file This is useful, but unfortunately uses awk. Note that you should remove the header from the file (if this exists). awk '{ st = index($0,","); print $0 > (substr($0,st+1) ".txt") }' snps_matrix_noheader.csv text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:molgenis_install&rev=1539079978&do=diff Installing Molgenis on Ubuntu Note that the below procedure would not have been possible without the help of David van Enckevort. The document highlights the installation of the Molgenis web application. If you would like to develop or contribute (code) to the project, you are better off forking the text/html 2021-03-18T15:40:25+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:pymol_install_on_windows&rev=1616082025&do=diff Install PyMOL on Windows If you are on Ubuntu, installing PyMOL is as easy as sudo apt-get install pymol. Unfortunately, almost all students in my medicinal biochemistry course use Windows as their OS (why?!). This page contains a “followable” text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:pymol_select_by_res_from_obj&rev=1539079978&do=diff PyMOL: Select all residues within n Ångströms of an object Here is your hard to find golden nugget: select somename, byres (polymer within 6 of someobject) text/html 2021-03-25T09:42:55+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rclone_copy_shared_folder_locally&rev=1616665375&do=diff Use RClone to Copy a "Shared With Me" Google Drive Folder Locally After some (a lot) of digging up I found the following works: rclone copy --drive-shared-with-me gdrive-um:"RNAseq Robert Pituitary Project" ./rformosa_rnaseq_pituitary/ text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_add_interface&rev=1539079978&do=diff Add an interface to a Rocks Cluster compute node Right after installation the compute nodes have no access to the outside network/world. This is of course problematic (especially if you have a licence server on the network). There are two ways how to fix this. First, either route traffic to the outside network via the frontend. Good luck with that, I have tried everything twice with no success and asked the experts text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_add_user&rev=1539079978&do=diff Add a user to the Rocks cluster As root: * For a normal user: useradd -g users jp * For a super user: useradd -G users,wheel jp * Set the password with passwd jp * We now need to rocks sync users to sync the user with all of the cluster. text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_change_private_network_domain_name&rev=1539079978&do=diff Change the domain name of the private network of a Rocks cluster Hopefully you won't have to do this. I did because the domain of the private network .local which is also the domain name of the public network (e.g. hercules.jp.local). You will have text/html 2019-07-16T16:03:52+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_configure_the_second_ethernet_card_on_compute_nodes&rev=1563293032&do=diff Configure the second ethernet card on compute nodes so that the rest of the intranet (and internet) is available (to compute nodes) By default Rocks does not configure your second ethernet card (most likely called em2). You can easily verify this by issuing text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_create_package_rpm&rev=1539079978&do=diff How to create an RPM package to install on Rocks compute nodes Install the all the files (or the program) in a directory and create a package: $ rocks create package /opt/psycopg2-2.6 psycopg2 This creates an RPM file in the directory, which you can then add to the compute nodes as explained text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_distribute_file&rev=1539079978&do=diff Distribute a file across a Rocks cluster (to all compute nodes) As root (or using sudo) * Edit /var/411/Files.mk (add your files at the bottom by concatenating to the FILES array) * Execute make -C /var/411. During execution you should see your files being listed. text/html 2019-07-04T12:28:12+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_ftp_a_file_from_head_to_compute_nodes_on_installation&rev=1562243292&do=diff FTP a file from head to compute nodes during compute installation This is usually done when you need some custom setup on your compute nodes. For example we use this to disseminate the same anaconda packages (and environments) across all compute nodes. text/html 2018-10-09T10:12:58+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_install_package_rpm&rev=1539079978&do=diff Installing an RPM on the Rocks compute nodes at kickstart On the frontend (as root or using sudo) install the package using yum install: $ sudo yum install postgresql-libs Sometimes you need to enable a particular repo to find the right package: text/html 2019-07-10T07:32:49+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:rocks_reinstall_compute_node&rev=1562743969&do=diff Install a Rocks Cluster compute node from scratch ssh in the compute node (as root) you want to reinstall and simply run: /boot/kickstart/cluster-kickstart-pxe The above will force a PXE boot and a reinstall from the frontend. text/html 2025-09-23T12:51:20+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:vm_install_lubuntu&rev=1758631880&do=diff How do I install and run Lubuntu as a Virtual Machine? Kindly follow these instructions to setup your local machine (e.g. laptop). ​ * Download and install the latest VirtualBox for your architecture * Download the Linux image for Lubuntu 24.03.1. Lubuntu is a lightweight and modern distribution of Ubuntu. We are going to run Lubuntu on your Windows or Mac machine (how cool is that?!) text/html 2023-08-13T17:23:26+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:vpn_install_arch&rev=1691947406&do=diff Install UM's VPN on Arch Linux Thank your god, this is not as hard as installing Eduroam on Arch Linux. First, you need to download all the necessary files (the certificate, your user's certificate and your user's key from the UM website; IT Support >> Manage IT Account text/html 2023-05-08T19:26:55+00:00 Anonymous (anonymous@undisclosed.example.com) https://bitsilla.com/wiki/doku.php?id=howto:wordpress_reset_theme&rev=1683574015&do=diff Reset a Theme in Wordpress (like it was a fresh install) When you install a theme and edit it in Wordpress, this tries to be clever and remember all your customizations so if you reinstall the theme you will have them all there. But sometimes when you fresh install a theme all you want is the original theme (which you have probably hosed with your carefree editing). Before running the commands