good practice for Galaxy admin and updates ?
Hi Galaxy Community, I manage a Galaxy local instance on a standalone station since 6 years. Regarding the numerous troubleshoots I have encountered (and the very long time I spent) during these years with Galaxy updates trying to minimize impact for my users, trying in the same time to keep safe our customization of this instance. I'm wondering about the best solution to manage my instance and to test my own dev and also Galaxy updates before pushing them on my production server. So I want to test a new way and I'd appreciate community feedback on this idea or let me know if other practices could be more appropriate in my situation. So my basic idea is: First install a new Galaxy from scratch, not by cloning the stable branch of Galaxy but by forking it. And use the master branch of this fork as production server. Then to manage an update of the Galaxy stable, my idea is to make a test branch on my fork, then pull Galaxy update on this test branch and run this test instance as a dev server in another station to be sure that update will not override our customization or any crucial tools/functionality for my users. Finally when I'll be sure this update is safe, I'll merge the test branch with the master branch of my forked Galaxy. Is anyone already acting in this way? Is it a good (the best) way to manage local instance of Galaxy? Have you other good practices? Thanks a lot for your help. Julie
Hi Julie We do something similar, I have described it here: https://galaxyproject.org/admin/ten-simple-steps-galaxy-as-a-service/#rule-1... Regards, Hans-Rudolf On 09/03/2018 01:19 PM, julie dubois wrote:
Hi Galaxy Community,
I manage a Galaxy local instance on a standalone station since 6 years. Regarding the numerous troubleshoots I have encountered (and the very long time I spent) during these years with Galaxy updates trying to minimize impact for my users, trying in the same time to keep safe our customization of this instance. I'm wondering about the best solution to manage my instance and to test my own dev and also Galaxy updates before pushing them on my production server.
So I want to test a new way and I'd appreciate community feedback on this idea or let me know if other practices could be more appropriate in my situation. So my basic idea is:
First install a new Galaxy from scratch, not by cloning the stable branch of Galaxy but by forking it. And use the master branch of this fork as production server. Then to manage an update of the Galaxy stable, my idea is to make a test branch on my fork, then pull Galaxy update on this test branch and run this test instance as a dev server in another station to be sure that update will not override our customization or any crucial tools/functionality for my users. Finally when I'll be sure this update is safe, I'll merge the test branch with the master branch of my forked Galaxy.
Is anyone already acting in this way? Is it a good (the best) way to manage local instance of Galaxy? Have you other good practices?
Thanks a lot for your help.
Julie ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/
Hi Thanks for your answer, so I will test this solution. Julie Le lun. 3 sept. 2018 à 13:53, Hans-Rudolf Hotz <hrh@fmi.ch> a écrit :
Hi Julie
We do something similar, I have described it here:
https://galaxyproject.org/admin/ten-simple-steps-galaxy-as-a-service/#rule-1...
Regards, Hans-Rudolf
On 09/03/2018 01:19 PM, julie dubois wrote:
Hi Galaxy Community,
I manage a Galaxy local instance on a standalone station since 6 years. Regarding the numerous troubleshoots I have encountered (and the very long time I spent) during these years with Galaxy updates trying to minimize impact for my users, trying in the same time to keep safe our customization of this instance. I'm wondering about the best solution to manage my instance and to test my own dev and also Galaxy updates before pushing them on my production server.
So I want to test a new way and I'd appreciate community feedback on this idea or let me know if other practices could be more appropriate in my situation. So my basic idea is:
First install a new Galaxy from scratch, not by cloning the stable branch of Galaxy but by forking it. And use the master branch of this fork as production server. Then to manage an update of the Galaxy stable, my idea is to make a test branch on my fork, then pull Galaxy update on this test branch and run this test instance as a dev server in another station to be sure that update will not override our customization or any crucial tools/functionality for my users. Finally when I'll be sure this update is safe, I'll merge the test branch with the master branch of my forked Galaxy.
Is anyone already acting in this way? Is it a good (the best) way to manage local instance of Galaxy? Have you other good practices?
Thanks a lot for your help.
Julie ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/
Hi Julie, This is (almost) exactly what we do. The only difference I see is due to modifications we have made to Galaxy itself we do not use the Galaxy master as our production server. Instead we have our own separate production and development branches. To upgrade we: 1. Pull the lastest master from upstream into the master branch 2. Merge upstream/master into our development branch 3. Test and fix until we are happy with everything. 4. Merge the development branch into our production branch. HTH, Keith
On Sep 3, 2018, at 7:19 AM, julie dubois <dubjulie@gmail.com> wrote:
Hi Galaxy Community,
I manage a Galaxy local instance on a standalone station since 6 years. Regarding the numerous troubleshoots I have encountered (and the very long time I spent) during these years with Galaxy updates trying to minimize impact for my users, trying in the same time to keep safe our customization of this instance. I'm wondering about the best solution to manage my instance and to test my own dev and also Galaxy updates before pushing them on my production server.
So I want to test a new way and I'd appreciate community feedback on this idea or let me know if other practices could be more appropriate in my situation. So my basic idea is:
First install a new Galaxy from scratch, not by cloning the stable branch of Galaxy but by forking it. And use the master branch of this fork as production server. Then to manage an update of the Galaxy stable, my idea is to make a test branch on my fork, then pull Galaxy update on this test branch and run this test instance as a dev server in another station to be sure that update will not override our customization or any crucial tools/functionality for my users. Finally when I'll be sure this update is safe, I'll merge the test branch with the master branch of my forked Galaxy.
Is anyone already acting in this way? Is it a good (the best) way to manage local instance of Galaxy? Have you other good practices?
Thanks a lot for your help.
Julie ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/
---------------------- Keith Suderman Research Associate Department of Computer Science Vassar College, Poughkeepsie NY suderman@cs.vassar.edu
participants (3)
-
Hans-Rudolf Hotz
-
julie dubois
-
Keith Suderman