A major ecloud.global maintenance operation is scheduled today. We are moving all the infrastructure to a new hosting provider. The goal is to offer better scalability, security and performance, and to introduce some E2E encryption features later this year.
As a result, all the ecloud.global infrastructure will be down for several hours today. This includes:
- e.email access from ecloud.global or from a mail client
- ecloud.global portal access
- all /e/ OS synchronization features (pictures, videos, files…)
Note: incoming emails won’t be lost, as mail servers retry for several days before giving up sending an email.
Other websites at /e/ won’t be impacted.
The downtime window will probably be several hours, since several terabytes of data have to be transferred, and some tests performed until everything is back.
The infrastructure will be off starting from: 16:00 CEST July 2nd 2020 (today). Hopefully services will be up again before 20:00 CEST.
We apologize for the inconvenience, but this operation is absolutely needed to increase the quality of our services.
Any update about this maintenance operation will be added to this message.
Update Friday July 3rd 2020: the migration has been completed during the night. Email service should be fully operational now. Ecloud services still have some issues, we hope to be able to fix everything soon.
Update Friday July 3rd 2020, 13:32 CEST all the migration went fine but we are still experiencing some issues under heavy load, so access to ecloud.global is currently disabled by default. We still need to figure out what is causing issues under heavy load. Will update as soon as we have some significant improvements. Sorry again for the inconvenience.
Update Friday July 3rd 2020, 23:31 CEST After the complete migration, we had major issues to have all services back again. The main reason is that we wanted to migrate to Nextcloud Encryption for user data but, despite 4 months of testing, it finally didn’t work. So we had to choose between reverting completely to our older infrastructure, or moving progressively to the new infrastucture, with a quite similar setup as the older one. We chose the second option but it’s still ongoing progress. We hope to be fully operational again during the week end. Stay tuned.
Update Monday July 6th, 10:30 CEST Arnau has updated this thread already during the past three days. Just let me give a quick update: during the rollback to the first server, we notived that the load was huge, and was close to put the server down. Latest investigations show that a few IP generate a very big part of the incoming traffic. Another thing is that probably login token have timeout, so the /e/OS edrive service probably requests new loggin for each user. At least we now have a good view of what prevents us to restart correctly on the short term, and what needs to be done for the new V1.5 implementation. We’re still doing some checks now, and some loading up attempts, then we take the decision to spread the restart load over a few hours or more on this server, or tomorrow we will upgrade to a much bigger server. We are trying to minimize the down time. Sorry for the inconvenience.