My server is stuck in "rebooting server" for 16 hours


#1

My server go unsuable yesterday (only ping was still working), so I restarted it.
But now it is stuck with “rebooting server” message since yesterday afternoon.
Nothing shows up in console, it seems that nothing happens.
I opened a ticket (XQCM) and got an answer “I will forward to team concerned and come back to you as soon as possible.”
And since then, no feedback from support team.
What can I do? It’s really critical.


#2

Hello,

I can see your ticket is currently being handled by the concerned teams, and they will provide you an update as soon as possible directly in the ticket.


#3

Hello,

This seems to be the way Scaleway operates.

One of our VM was crashed last night with I/O errors with in HW showing in Scaleways console and VM was offline. VM was rebooted from same console.

VM has now been 17 hours in pending state ‘stopping’.

UNABLE TO START VM!!!

We even tried take snapshot from VM disk and got message ‘Error: Can’t snapshot, associated server is in a transient state’.

Situation is that, there is nothing that us as a customer can do to fix this. After escalating ticket got message that Scaleway helpdesk cannot do anything. Ticket is handeled by other tech guys. Still waiting any information from their side.

This is why You shouldn’t run anything important with Scaleway. Seems that Scaleway isn’t production grade service provider!

I must say that I’m very disappointed and hopefully at least some data can be recovered. Fingers crossed.


#4

UPDATE: Ticket has been open for 24 hours and VM is still down. Outage has now been more than 24 hours.

Seems that there is no one working at Scaleway on weekends. I have paid support, but that seems to be just an empty promisees.


#5

Just got message from support “Unfortunetly, we did not yet get any feedbacks from the team in charge and we probably won"t get any before tomorrow.”. I’m now confessed that there is now service during weekends.

Ticket has been open for more than 24 hours and services down even more. That’s it for promised 99.9% SLA.


#6

the paid support plans are just a joke. There is honestly no difference to free support. You do not get faster nor better answers. Basically just a scam to get more money from you.

And the waiting times are quite normal, feel lucky if you get an answer.


#7

It’s already Monday and VM is still down. No new messages from support. This is ridiculous!


#8

nha mate, this is just scaleway. Been like this for at least 3 years.


#9

Same process for me.

Last night these two servers became completely inaccessible, yet scaleway reported them as fine:

Clicked reboot on them, and they have been rebooting for 30 minutes now. Cannot make snapshots or detach the volumes due to the “transient state” error.

Submitted support tickets. I’m on the dev support plan. Hoping for resolution.


The only thing I could see that was out of the ordinary, is that the boot mode was in an invalid state on the UI. This seems a Scaleway bug rather than something I’ve misconfigured.

The page that displays boot mode for the servers has it set to “local boot”, yet when you mouse over it, it says “local boot not available”.

If you then change it to “bootscript”, then “use local boot” can no longer be selected (what it was previously reported as)

Seems a UI bug.


My other thought on this, is that both servers are VC1S servers. Perhaps they dropped support for them and failed to transition people to the START1-S servers.

However, if that is the case, then they could at least shut down the box so that we can detach the volume and transition it ourself.


All in all, I don’t know why there is no cancel, or hard reboot, or try again option.


Update. One of the servers (discuss.bevry.me) came back up after 40 or so minutes.

Also got this email reply back.

There is a network outage impacting part of our infrastructure.

Issue is currently being investigated and we will share additional details through our status available here :

https://status.scaleway.com/incident/260

We truly apologise for an inconvenience and remain at your disposal should you have any further question.

I ask for your patience.

When I checked status.scaleway.com at the time of reporting, it said all systems were operational. Guess it took a while.


Will see about migrating to one of the START1 servers once the situation stabilises.


#10

moving to a start1 will automatically happen as soon as you reboot. so you either are already in the progress of transfering or you already transferred to start1. This is what scaleway communicated as they announced the start1, but I couldn’t find the thread in a quick search right now.

However, if you want to migrate and have a working network, reliable services & actually a properly working server with reboot & backup and more power for almost the same price, have a look at hetzner. https://www.hetzner.com/cloud


#11

My server also down more than 6 hours.No one help me yet.Very very bad support.


#12

ah and btw, paying support is a absolutely scam. no difference in support at all.


#13

If you shutdown and archive a VC1S then restart it you will be on the new START1-S hardware (cat /proc/cpuinfo confirms) as mentioned above but it still shows as VC1S and you get billed at the lower VC1S price. I just switched over my last VC1S and all went well. If you restore a VC1S snapshot to a new server for any reason then you get billed at the higher rate as you can’t restore it to a new VC1S.