MongoDB Volume Restore
I had a MongoDB database service in the US West region. I decided to move it to Singapore. During the migration errors occurred and I decided to move back. But after migrating back almost all my data is lost. I can see in the dashboard that it is there, but I can't retrieve it in any way. I even tried uninstalling the service and screwing that volume back to it, but now there is nothing at all. Please help me as soon as possible. It would be nice to rollback to the beginning of September 28 and download a dump of that database
26 Replies
Project ID:
ce31b5f2-77b4-444c-9940-d6ef2bd33cc0Your mongo instance definitely has stuff in it
There's 114GB of stuff there
It looks like youv'e removed all yourservices as well?
New reply sent from Help Station thread:
Yes, we removed MongoDB services, but we didn't delete the volume. Data was lost during the transfer to another regionYou're seeing this because this thread has been automatically linked to the Help Station thread.
How soon will action be taken? We are losing our reputation and revenue because of your migration error. We are talking about tens of thousands of dollars and we would not want to be idle for so long. We are willing to go to a higher rate or something else to get you to restore the database
Hi, as I mentioned before, your data is there
Whatever you've done, it's on the application level
Can you tell me what error you're running into? What data is "missing"
Because you have ~200GB worth of stuff in here
We have a completely empty database. We manually looked at the volume and there is no data in it. Please rollback all changes to the project until September 28th
In the users collection (database name: test) we last saw 22K users, but the actual number of users is 14M+. Just roll it back 24 hours
Looking into what happened first
We were migrating our MongoDB database from US West to Singapore and during the migration we encountered a problem that broke all the data
Sorry, and now it's back in us west?
How did you manage that? We lock failed migrations for a reason here...
Did you make a new service and delete your old service? This Mongo instance had it's first deployment 15 hours ago
Now we're back in the US West. We have been removing MongoDB services and creating new ones by baiting them with old volume
Oh man
Can you just rollback the project as you have done before with other users?
We do not do this for other users
Railway does not manage backups for you
But we can see that our data is there, but where?
That's what we're looking into
At the time of the transfer, your dashboard showed that we were supposedly running out of space, even though we clearly had it
Also, if you want a higher level of service, we're definitely going to have to talk about that after
Having us respond and then you torch us on social media because you didn't respond here doesn't work for us
https://x.com/301Community/status/1840350767980450219
301 Community (@301Community) on X
So, regarding what's going on: our hosting provider @Railway has broken our database and is currently not responding to our support requests. We hope that action will be taken soon and look forward to the support of our community. Once restored, we will be sure to give everyone a
Twitter
We will remove it if you really try to help us. We had no other options and your support was silent. Our app has been idle for over a day
We responded here and you did not
https://discord.com/channels/713503345364697088/1289695241716371579/1289738325510455327
We will talk about this after
after what?
Okay, we deleted the post.
After we fix your problem
We've found your volume and we're looking to roll it out. We will redeploy your instance and are working on getting it back up with the data
Please do not deploy it while we do this
Can you check if your data is restored? We think we tracked back down your original volume
Database Connection
We are unable to connect to the database over the public network.
Can you connect using this command?
mongosh "mongodb://mongo:**@junction.proxy.rlwy.net:55268"
Gotchya looking into it
One more time please
I can confirm through our data UI that we see something
I want to make sure it is the correct data
i really love you bro
how can I ever thank you?
You can pay us what people normally pay us to wake us up in the middle of the night :kekw:
Starts at a 25k/mo minimum committed compute spend
(We will also fix that issue with the volume migration that messed you up, and for that, we're sorry)
Hindsight 20/20, but how could you NOT schedule db backups? Also, I don't get the point of that (deleted) tweet, it screams attack vector & too lazy to do db backups