As many of you noticed, the Zenodo upgrade last Friday didn’t go as smoothly as hoped. Afterwards Zenodo was sluggish and file uploads were painfully slow, some features didn’t quite work as expected. We’ve been working flat-out behind the scenes to steer Zenodo out of these stormy waters. We’d like to keep you updated on the issues we had, what we solved so far, and what we are still working on.

The upgrade involved orchestration of three components; Zenodo’s refactored software platform, the data, and the technical infrastructure:

1. We’d developed the refactored software platform in collaboration with partners around the world, 6 of which already had production instances running on the new code (*), giving confidence, but we none the less tested extensively on the full feature set used on Zenodo. We’d been smoothing off rough edges, and were prepared to rapidly address more that you might discover after release.
2. As custodians of your data, for months we meticulously exercised the process of migration to the new system to ensure every bit made it reliably, and also have multiple backups as safety and for verification.
3. We’d revamped the technical infrastructure to better serve the continued scaling demanded by Zenodo’s ceaseless growth, by using components heavily used in our other front-line services, which none the less we also stress/performance tested in the preparations.

The main issue after the upgrade was immediately evident that file upload/download was painfully slow (1GB file taking 1 hour instead of 20 seconds). We worked the weekend to finally discover that circumventing the front-facing load balancer alleviated the problem, and on removing it we restored Zenodo’s expected data performance (the root cause is still being investigated by the infrastructure experts).

We also had indexing issues with records which meant they didn't show up despite the data having been migrated, and despite having had successful test migration runs. We have almost finished to resolve this issue and expect the main issues to have been solved by Monday.

We want to thank you for all the incredible support and understanding that we have been receiving even under such difficulties. We’ve not ignored the functionality hiccups you’ve been reporting, we already solved some in parallel to the performance debugging. Please rest assured that we continue to work relentlessly to address all your support requests and feedback, and to address all outstanding issues in a timely manner to help you publish/access your content.

We apologise for all the inconvenience it has caused. We hope once we're over the hiccups the new Zenodo will serve you and your communities well!

Final lesson learnt: Friday 13th[**] might not have been a good day for a major release after all.

(*)

• Northwestern Uni: https://prism.northwestern.edu/
• NYU: https://ultraviolet.library.nyu.edu/
• GEO Knowledge Hub: https://gkhub.earthobservations.org/
• Caltech Library: https://data.caltech.edu/
• TUGraz: https://repository.tugraz.at/
• TUWien: https://researchdata.tuwien.ac.at/

(**) We chose a Friday as Zenodo has considerable less traffic during weekends and thus as risk mitigation measure we would impact less users if there were problems (as there unfortunately turned out to be).

CERN, OpenAIRE, and the InvenioRDM open source community are excited to announce that Zenodo has moved onto our next generation underlying technical platform, InvenioRDM!

Over the past year, we've been working intensely on preparing to move Zenodo on top of a refreshed underlying technical platform, InvenioRDM. Zenodo's simple user experience and high scalability stay the same, but the underlying engine has been substantially upgraded. In addition InvenioRDM comes with a suite of new features and improvements that have been high on many of our users' wishlist.

What's new?

We've significantly expanded Zenodo's collaborative features in many different areas:

• Communities: Our community feature has been upgraded with support for multiple curators, members, reviews, curation, and branding, so e.g. multiple curators can now edit records in their community.
• Sharing: You can now share records for confidential peer review, enable access requests, or simply create a preview link for your colleagues.
• Deposit: Our upload form has received many usability improvements, e.g. being able to select the file which should be previewed by default. In addition we've strengthened it through connections to the open science PID infrsatructure, e.g. you can now auto-complete creators from ORCiD and affiliations from ROR, and link to custom funders/awards.
• Extras: We've also made significant improvements to web accessibility, enabled institutional login via the OpenAIRE AAI, improved usability, and added a download all button for files among other things.

You can find a comprehensive overview of new and changed behaviors in our documentation:

• What's new?
• What's changed?

Checkout the user guide for a full overview on how to use the many new features:

• User guide

Feedback

Don't hesitate to reach out to us on support in case you find any issues on the refreshed Zenodo.

InvenioRDM - an open source community

The engine behind the new Zenodo, InvenioRDM, is available for institutions worldwide and can be customized to your institutional or domain-specific needs, staying close to Zenodo's simple and seamless user experience.

We're proud to have built InvenioRDM together with 27 other partner organisations across the globe as an open source community around the simple vision of providing a great user experience to researchers with the ability to handle large amounts of data. Next, we're working towards making InvenioRDM a fully collaborative platform that empowers its users to share and preserve research.

Credit

The new Zenodo platform was made possible through a strong open collaboration.

We're grateful to our project funders who have enabled us to build and operate Zenodo for the long tail of science over the past 10 years:

• European Union
  • OpenAIRE, OpenAIREplus, OpenAIRE2020, OpenAIRE-Connect, OpenAIRE-Advance, OpenAIRE-Nexus, BICIKL, CS3MESH4EOSC, FAIRCORE4EOSC, HORIZON-ZEN
• US National Institutes of Health
  • GREI - Generalist Repository Ecosystem Initiative
• Arcadia Fund
  • Biodiversity Literature Repository 1 and 2
• Alfred P. Sloan Foundation
  • Asclepias, Hub for accessible data publishing.
• CERN Knowledge Transfer Fund
  • InvenioRDM - an open-source research data management platform
• Donors to CERN & Society Foundation

We're grateful to all the InvenioRDM partners who shared our vision and helped build an amazing next-generation platform. Thanks to:

• Brookhaven National Laboratory
• Caltech
• CERN
• CNUDST
• Cottage Labs
• Data Futures
• EkoConnect
• EU Joint Research Centre (JRC)
• FZ Juelich
• Group on Earth Observations (GEO)
• Front Matter
• HZDR
• Japan National Institutes of Informatics
• KTH
• New York University
• Northwestern University
• OpenAIRE
• TIND
• TU Graz
• TU Wien
• TUBITAK
• Universität Bamberg
• Universität Hamburg
• Universität Freiburg
• Universität Tübingen
• WACREN
• WWU Münster
• ZHB Luzern

We'd like to extend a big thank you to our users and project partners for testing, feedback, and patience, in particular:

• California Digital Library
• Christopher Erdmann
• Dryad
• Northwestern University
• OpenAIRE
• Plazi
• ZHB Luzern

The CERN team who worked on the implementation and migration:

• Alexandros Ioannidis
• Anika Churilova
• Antonio Vivace
• Dimitris Frangiadakis
• Fatimah Zulfiqar
• Francois Decourcelle
• George Lignos
• German Cancio Melia
• Javier Romero Castro
• Jean-Yves Le Meur
• Jenny Bonsak
• Jose Benito Gonzalez Lopez
• Karolina Prezerwa
• Lars Holm Nielsen
• Manuel Alejandro De Oliveira Da Costa
• Martin Lettry
• Nicola Tarocco
• Pablo Garcia Marcos
• Pablo Panero
• Pablo Saiz
• Panna Georgina Liptak
• Rodrigo Oliveira Almeida
• Tim Smith
• Yash Lamda
• Zacharias Zacharodimos

Including many other highly experienced CERN colleagues from the IT Department who helped with all things infrastructure (admin, compute, databases, network, openshift, outreach, search, storage).

The InvenioRDM open source community who contributed with design, development, testing, requirements, outreach, documentation, and expertise:

• Alessia Bardi, OpenAIRE/National Research Council of Italy
• Alexander Bardel, TU Graz
• Anders Friis-Christensen, European Commission JRC
• Audun Bjørkøy, TIND
• Austin Sharp, Northwestern University
• Bessem AAmira, CNUDST
• Camelia Ignat, European Commission JRC
• Chokri Ben Romdhane, CNUDST
• Christian Erlinger, ZHB
• Christina Zeller, University of Bamberg
• Christoph Ladurner, TU Graz
• Claudio Atzori, OpenAIRE/National Research Council of Italy
• Dan Granville, Data Futures GmbH
• David Eckhard, TU Graz
• David Pape, HZDR
• Eric Newman, Northwestern University
• Esteban Gabancho, Universität Hamburg
• Gretchen Neidhardt, Northwestern University
• Guido Juckeland, HZDR
• Guillaume Viger, Northwestern University
• Hagar Lowenthal, European Commission JRC
• Hakan Bayındır, TÜBİTAK ULAKBİM
• Hermann Schranzhofer, TU Graz
• Ilire Hasani-Mavriqi, TU Graz
• Jan-Olof Wiefel, Universität Münster
• Jesse Asamoa, WACREN
• Jonathan Bauer, University of Freiburg
• Jonathan LeBonzec, TIND
• Joshua Elder, Northwestern University
• Kai Wörner, Universität Hamburg
• Karen Gutzman, Northwestern University
• Karl Krägelin, Universität Münster
• Katerina Iatropoulou, OpenAIRE/Athena Research and Innovation Centre
• Kathrin Heim, ZHB
• Kristi Holmes, Northwestern University
• Lars Holm Nielsen, CERN
• Mahonry Estrada, Northwestern University
• Maik Fiedler, HZDR
• Martin Fenner, Frontmatter
• Matt Carson, Northwestern University
• Maximilian Johannes Moser, TU Wien
• Michael Dütting, Universität Münster
• Michael Groh, University of Bamberg
• Mike Hucka, Caltech
• Mojib Wali, TU Graz
• Nikolaos Alexandrakis, European Commission JRC
• Oliver Knodel, HZDR
• Paolo Manghi, OpenAIRE/National Research Council of Italy
• Peter Cornwell, Data Futures GmbH
• Philipp Gualdi, TU Graz
• Robert Doiel, Caltech
• Rosa Lönneborg, KTH
• Salaheddine Ben Ali, CNUDST
• Sam Arbid, KTH
• Sara Gonzales, Northwestern University
• Sarah Wiechers, Universität Münster
• Sefakor Ankora, WACREN
• Sotirios Tsepelakis, TU Wien
• Thomas Gruber, HZDR
• Tom Morrell, Caltech
• Werner Greßhoff, Universität Münster
• Wibke Fellermann, Universität Münster

See the original announcement.

We unfortunately have to postpone the planned migration on September 29th 6-8 UTC. The new migration date is October 13 from 06:00-08:00 UTC. We apologise for any inconvenience it may cause.

During a final trial migration run last week we have discovered some issues that we've been trying to address in time to be ready to start the final migration process today. Unfortunately, we've not be able to fully address the issues and as a precautionary measure we're postponing the migration.

Zenodo has over the past 10 years grown in both scale and size, and thus migrating the full system to a new technical platform is a large undertaking. In order to minimize service disruption for our users, we're taking an incremental approach. This approach works by migrating a snapshot of the existing system, and afterwards incrementally apply changes in the existing system to the new system. This method enables us to keep the downtime of Zenodo to about 1-2 hours instead of the ~3 days which full migration takes due to Zenodo's scale and size. The incremental approach however is also more technically challenging, and it is through our extensive quality assurance checks that we discovered issues that we need further time to investigate.

If you're eager to learn more abou the new platform, then you can find a comprehensive overview over what's new and what's changed on the following links:

• What's new?
• What's changed?