Signal server down

[ryannathans]

I have:

• [ x] searched open and closed issues for duplicates
• [ x] read https://github.com/WhisperSystems/Signal-Android/wiki/Submitting-useful-bug-reports

---

Bug description

Nobody I speak to can send messages on signal, all Android users in Australia.

Steps to reproduce

Try to send message, throws an error.

Screenshots

Device info

Multiple different Android phones, different Android versions.

Link to debug log

https://hastebin.com/umakuzuyan

[MartinX3]

Up in germany, thanks tomthe developers. :)

[3rg1s]

They are just a small team, shouldn’t take more to bring the service up!

Edit: Up in my Country :)

[ghost]

Doesn't they use AWS servers to manage everything? Keeping it on rPI would be silly and completely dependent on the poor bandwidth(no offence but not everyone has 1GB / s) of the people who host it.

[petcap]

You can add this to your hosts file as a temporary fix: 52.22.204.209 textsecure-service.whispersystems.org

[MartinX3]

Scale vertical instead of horizontal would be the magic keyword here. Combined with a smart load balancer like any DNS The more raspis are connected, the stronger is the network against failures.

Am 5. März 2018 11:01:17 MEZ schrieb angaunba notifications@github.com:

Doesn't they use AWS servers to manage everything? Keeping it on rPI would be silly and completely dependent on the poor bandwidth(no offence but not everyone has 1GB / s) of the people who host it.

-- You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: https://github.com/signalapp/Signal-Android/issues/7460#issuecomment-370368410

-- Mit freundlichen Grüßen

Martin Dünkelmann

[KarolTrzeszczkowski]

Up in Poland :)

[ar329]

Up in Canada. Yay!!

[GAS85]

@Cerberus0 Still no info...

[537]

Up in Bahrain!

[versable]

@NayefAlharbi, you might want to redact your phone number

[MartinX3]

We need an info page on the homepage about the service status. https://github.com/signalapp/Signal-Android/issues/3835 https://github.com/signalapp/Signal-Android/issues/7461

[vanitasvitae]

https://signal.org/blog/the-ecosystem-is-moving/ :smile:

[ghost]

So no information about what really happened? Something more than "brief outage that has been resolved" https://github.com/signalapp/Signal-Desktop/issues/2106#issuecomment-370502588

[mawed]

Just for documentation:

https://mobile.twitter.com/signalapp/status/970751021445144576

[ghost]

Service was disrupted for few hours and I just don't believe it was by "huge user traffic". If you had a bug or someone fucked up something then no problem no one is perfect but you should explain. As a foundation you should be one hundred percent transparent with your users and this explanation is just "complete and utter garbage" and "bullshit" as would Linus say.

[ryannathans]

Looks like someone got DoSed and doesn't want to admit it ;)

Servers can probably be scaled to anyone who wishes to volunteer a node without "federating". Server version doesn't match? Don't let the node into the DNS load balancing pool. Throw in auto updating and make the binaries public too for new nodes and authenticity sake. Heck, you could just use some generic containerised approach.

Either that or find another way to circumvent DoS attacks..

[sigenc]

yeah i too don't believe this. It looks like ddos, we need to have the ability to run nodes that have remote attestation

[rainerzufall]

Let's all stop discussing this here on github. Let's do the discussion in the community forum: https://community.signalusers.org

Additionally Signal just answered on twitter:

The outage was not caused by a DDoS or any other type of attack. (https://twitter.com/signalapp/status/970781128331214849?s=20)