Lost all work from this morning

Hi, very new to Gitpod. It looks promising but I’m experiencing a glitch.

I lost my internet connection while on a zoom call and when I returned to gitpod I received a message saying something like, “Sorry, we could not save your workspace. Please contact technical support if you need it.”

I should be able to reconstruct my work before getting support on this community forum but can you tell me what happened? I looked in the documentation for an autosave feature but it’s not clear if it is on by default or if it’s even an option anymore because it’s now built in.

A few more questions:

  1. What does the share option do on the workspaces dashboard? EDIT: while waiting for my db to import, I read the docs on this feature. However, shouldn’t the indicator that a workspace is shared by prominently on the dashboard and not tucked away in a hidden contextual menu?
  2. Is this the method to communicate with tech support for urgent matters? There is a form on the contact us page but if I did want that workspace back how do I do that without waiting until Monday morning? Do I paste the workspace ID here? Is that safe?
  3. Is there a page describing how to get technical support for people on paid plans?

EDIT: It happened again:

Edit: I performed another test this morning (morning, June 7/2021). After uploading the database, I closed the workspace purposefully before a disconnect occured (out of my control) and a subsequent time out because I was on a zoom call.

Same message about the backup failing and once again I lost the workspace. The database I had just imported is 9GB (it takes over an hour). It seems possible that the backup is failing due to the size of the database and the disconnect issue isn’t really the problem.

Any insight would be appreciated. The workspace is chocolate-bear-9qb2tq1z.

Cheers,
André

1 Like

Hi @aangelinsf ,

very new to Gitpod. It looks promising but I’m experiencing a glitch

welcome aboard - and sorry for that glitch! let’s dive in to see what we can do.

However, shouldn’t the indicator that a workspace is shared by prominently on the dashboard and not tucked away in a hidden contextual menu?

Yes, absolutely. We discussed several options lately, and will forward your feedback.

Is this the method to communicate with tech support for urgent matters?

The way to go is email to support@gitpod.io with as many concrete details (ids) as possible. We generally discourage pasting workspace IDs publicy, but more as general secuirty measure than because of a concrete concern. We’re currently growing our organization, but only have support on weekdays. This is the same for paid plans.

I looked in the documentation for an autosave feature

A workspace is stopped after a certain time (usually 30mins, may be extended for some plans) of inactivity (as triggered by connectivity issues9. When that happens a backup is made. This workspace (and it’s state) is persisted for at least for 14 days - or forever if you either pin it or it contains a git diff.

But this backup has an upper limit: 30 GBs atm. It sounds like you might have triggered that. In these cases, the backup will fail. We definitely need to communicate this boundary more clearly.

I will check your workspace now to verify this is what happened.

Indeed: the workspaces was slightly bigger than the allowed 30GBs.

Hi, Gero. Thank you for that and for passing along my suggestion. I hope you take the following in the spirit of wanting you to have a great service…Sunday was frustrating for several reasons.

Some additional feedback regarding support, I looked but didn’t see that guidance for getting support (using the email and what to include). And often there is a support link prominently on the page but there isn’t one now. I did eventually look at the bottom for a Contact Is but I think it should be easier to find up in top menu when logged in—with guidance on what you need to provide support on the page.

Pantheon makes it prominent.

I understand there needs to be a limit. I did read some prior questions in the community that mentioned a new 30GB limit but with 9GB of data it seemed to me that I was safe. Perhaps there should be a warning?

Also, is there a way to see how big a workspace has gotten? docker system df -v ?

One more suggestion. Please consider making it a bit faster to open a workspace with one click. Perhaps make it a drop-down button menu i.e. clicking the “Open” button opens the workspace but clicking a down arrow to the right of the button shows the additional options.

@geropl I just performed a test and the 8GB db snapshot ballooned to 24GB in MariaDB. With the unzipped original file of 8 090 376 121, I’m definitely over:

REPOSITORY                        TAG                     IMAGE ID            CREATED             SIZE                SHARED SIZE         UNIQUE SIZE         CONTAINERS
drud/ddev-webserver               v1.17.4-ww2-dev-built   4501b8b119b2        2 hours ago         1.067GB             1.067GB             11.32kB             1
drud/ddev-dbserver-mariadb-10.3   v1.17.3-ww2-dev-built   b4bc0497f4bf        2 hours ago         653.8MB             653.8MB             9.592kB             1
drud/ddev-ssh-agent               v1.17.0-built           48bb819088bd        2 hours ago         111.9MB             111.9MB             6.768kB             1
busybox                           latest                  69593048aa3a        23 hours ago        1.236MB             0B                  1.236MB             0
phpmyadmin                        5                       df1995f1ce3d        4 days ago          473.6MB             0B                  473.6MB             1
drud/ddev-webserver               v1.17.4                 bb6a7602af8a        4 days ago          1.067GB             1.067GB             0B                  0
drud/ddev-dbserver-mariadb-10.3   v1.17.3                 9e467f9d6dbc        5 days ago          653.8MB             653.8MB             0B                  0
drud/ddev-router                  v1.17.3                 e7b0c51fc332        5 days ago          243.2MB             0B                  243.2MB             1
drud/ddev-ssh-agent               v1.17.0                 849e4bb443a5        2 months ago        111.9MB             111.9MB             0B   

VOLUME NAME                 LINKS               SIZE
ddev-ssh-agent_dot_ssh      1                   0B
ddev-ssh-agent_socket_dir   2                   0B
ww2-dev-mariadb             1                   24.18GB
ddev-global-cache           3                   4.237kB

It’s odd that I don’t see the original 9GB in the above accounting. Hmm.

In any case, I’ve deleted the snapshot; only the data in the db should remain. I’ve also stopped the workspace, which I presume triggers a backup. Let’s see what happens.