RAVNIX offline?

full_bars · April 7

Here are some MTR results from my two servers. A few things worth noting:
Quad9 trombones out to Equinix in Ashburn rather than peering locally in Atlanta, adding ~27ms of unnecessary latency. Not a FiberLight/Telia issue, just Quad9's Atlanta presence being weak.

Georgia Tech routes through SOX.net (Southern Crossroads), Atlanta's regional research/education exchange. Base latency is 2-3ms, traffic stays fully local. High loss% on some intermediate hops is ICMP rate limiting, not real loss.

Server 1
=== MTR to 1.1.1.1 ===
Start: 2026-04-07T14:30:26
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 2.0% 50 0.7 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 1.8 2.5 1.6 22.8 3.1
3.|-- 67.204.31.68 0.0% 50 2.2 2.0 1.8 2.5 0.1
4.|-- atl-b4-link.ip.twelve99.n 68.0% 50 2.2 2.9 2.1 11.0 2.2
5.|-- atl-bb1-link.ip.twelve99. 0.0% 50 2.3 2.3 2.0 3.3 0.2
6.|-- atl-b5-link.ip.twelve99.n 0.0% 50 2.1 2.0 1.7 3.4 0.2
7.|-- cloudflare-ic-381672.ip.t 0.0% 50 2.0 2.0 1.8 2.3 0.1
8.|-- 108.162.235.127 0.0% 50 2.8 4.9 2.3 20.8 4.7
9.|-- one.one.one.one 0.0% 50 2.2 2.2 2.0 2.5 0.1

=== MTR to 8.8.8.8 ===
Start: 2026-04-07T14:31:33
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 16.0% 50 0.5 0.5 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.3 3.3 1.6 33.2 5.9
3.|-- 67.204.31.68 0.0% 50 3.8 2.0 1.7 3.8 0.3
4.|-- atl-b4-link.ip.twelve99.n 90.0% 50 2.1 2.1 1.9 2.6 0.2
5.|-- 192.178.68.246 0.0% 50 14.5 14.5 14.0 18.4 0.8
6.|-- 172.253.71.211 0.0% 50 2.5 2.2 2.0 2.8 0.1
7.|-- 142.251.241.187 0.0% 50 13.8 13.6 13.4 13.9 0.1
8.|-- dns.google 0.0% 50 2.2 2.0 1.8 2.3 0.1

=== MTR to 9.9.9.9 ===
Start: 2026-04-07T14:32:39
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 14.0% 50 0.6 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.0 3.8 1.6 51.9 8.5
3.|-- 67.204.31.64 0.0% 50 22.1 22.9 21.7 54.1 4.9
4.|-- 67.204.31.53 0.0% 50 22.3 22.2 22.0 22.5 0.1
5.|-- 67.204.31.50 0.0% 50 22.2 22.2 22.0 22.4 0.1
6.|-- equinix-ashburn.woodynet. 0.0% 50 32.9 29.8 29.2 32.9 0.7
7.|-- dns9.quad9.net 0.0% 50 30.1 29.2 28.8 32.1 0.5

=== MTR to 130.207.244.244 (Georgia Tech) ===
Start: 2026-04-07T14:33:45
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.5 0.5 0.3 0.9 0.1
2.|-- 64.186.132.65 0.0% 50 3.0 2.7 1.6 25.6 4.0
3.|-- 67.204.31.68 0.0% 50 1.9 1.9 1.6 2.3 0.2
4.|-- 206.71.12.44 0.0% 50 5.0 4.2 2.8 5.3 0.7
5.|-- richgw2-to-trcps.sox.net 2.0% 50 2.0 2.1 1.9 2.4 0.1
6.|-- 172.18.11.17 70.0% 50 2.2 2.4 2.2 2.6 0.1
7.|-- 172.18.6.113 2.0% 50 2.4 2.5 2.2 6.1 0.6
8.|-- 172.18.6.83 82.0% 50 3.4 3.2 2.7 4.4 0.5
9.|-- 10.10.59.149 0.0% 50 3.0 3.0 2.7 3.6 0.2
10.|-- 10.136.72.1 0.0% 50 3.4 3.4 2.9 4.6 0.4
11.|-- ??? 100.0 50 0.0 0.0 0.0 0.0 0.0

=== MTR to 206.71.12.44 (SOX.net) ===
Start: 2026-04-07T14:34:54
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 4.0% 50 0.4 0.5 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 1.8 3.3 1.6 29.1 5.5
3.|-- 67.204.31.68 0.0% 50 1.8 1.9 1.7 2.4 0.1
4.|-- 206.71.12.44 0.0% 50 3.0 4.1 2.8 5.6 0.7

=== MTR to 74.50.11.1 (gateway) ===
Start: 2026-04-07T14:35:59
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.4 0.4 0.3 0.8 0.1

=== MTR to 64.186.132.65 (FiberLight) ===
Start: 2026-04-07T14:36:55
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.4 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.3 3.6 1.7 32.7 5.9

Server 2
=== MTR to 1.1.1.1 ===
Start: 2026-04-07T14:30:29
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 4.0% 50 0.6 0.4 0.3 1.0 0.1
2.|-- 64.186.132.65 0.0% 50 27.3 7.8 1.5 38.0 11.8
3.|-- 67.204.31.68 0.0% 50 1.8 2.0 1.7 5.1 0.5
4.|-- atl-b4-link.ip.twelve99.n 74.0% 50 2.1 2.2 2.0 2.6 0.2
5.|-- atl-bb1-link.ip.twelve99. 6.0% 50 2.2 2.2 2.0 3.6 0.2
6.|-- atl-b5-link.ip.twelve99.n 0.0% 50 2.1 2.0 1.8 3.3 0.3
7.|-- cloudflare-ic-381672.ip.t 0.0% 50 2.0 1.9 1.7 2.3 0.1
8.|-- 108.162.235.133 0.0% 50 2.2 4.1 2.1 17.5 3.4
9.|-- one.one.one.one 0.0% 50 2.2 2.2 1.9 2.5 0.1

=== MTR to 8.8.8.8 ===
Start: 2026-04-07T14:31:37
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 24.0% 50 0.3 0.4 0.3 0.8 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 3.6 1.6 48.0 7.7
3.|-- 67.204.31.68 0.0% 50 1.9 2.0 1.7 2.5 0.2
4.|-- atl-b4-link.ip.twelve99.n 64.0% 50 2.4 2.3 2.0 4.0 0.4
5.|-- 192.178.68.246 0.0% 50 14.3 14.5 14.0 17.5 0.7
6.|-- 172.253.71.211 0.0% 50 13.5 13.6 13.4 14.1 0.1
7.|-- 142.251.55.119 0.0% 50 13.6 13.6 13.4 14.5 0.2
8.|-- dns.google 0.0% 50 2.0 2.1 1.9 2.8 0.1

=== MTR to 9.9.9.9 ===
Start: 2026-04-07T14:32:42
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 6.0% 50 0.5 0.4 0.3 0.6 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 3.1 1.6 40.1 6.0
3.|-- 67.204.31.64 0.0% 50 21.9 24.5 21.7 87.2 11.5
4.|-- 67.204.31.53 0.0% 50 21.9 22.1 21.9 22.8 0.2
5.|-- 67.204.31.50 0.0% 50 22.0 22.1 21.9 22.4 0.1
6.|-- equinix-ashburn.woodynet. 0.0% 50 29.5 29.6 29.3 31.0 0.4
7.|-- dns9.quad9.net 0.0% 50 29.0 29.0 28.8 29.4 0.1

=== MTR to 130.207.244.244 (Georgia Tech) ===
Start: 2026-04-07T14:33:49
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 4.0% 50 0.6 0.4 0.3 0.8 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 2.1 1.6 8.1 1.3
3.|-- 67.204.31.68 0.0% 50 2.2 1.9 1.7 2.4 0.1
4.|-- 206.71.12.44 0.0% 50 3.3 4.5 2.5 6.7 0.9
5.|-- richgw2-to-trcps.sox.net 2.0% 50 2.2 2.0 1.9 2.2 0.1
6.|-- 172.18.11.17 64.0% 50 2.4 2.3 2.2 2.6 0.1
7.|-- 172.18.6.113 2.0% 50 2.5 2.4 2.1 3.4 0.2
8.|-- 172.18.6.83 84.0% 50 2.8 3.0 2.8 3.2 0.1
9.|-- 10.10.59.149 0.0% 50 49.7 5.1 2.6 54.6 9.8
10.|-- 10.136.72.1 2.0% 50 4.7 4.0 2.8 16.2 2.2
11.|-- ??? 100.0 50 0.0 0.0 0.0 0.0 0.0

=== MTR to 206.71.12.44 (SOX.net) ===
Start: 2026-04-07T14:34:55
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 2.0% 50 0.4 0.4 0.3 0.8 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 3.6 1.5 45.9 7.8
3.|-- 67.204.31.68 0.0% 50 1.6 1.9 1.6 2.5 0.2
4.|-- 206.71.12.44 0.0% 50 5.2 3.9 2.0 5.7 0.9

=== MTR to 74.50.11.1 (gateway) ===
Start: 2026-04-07T14:36:00
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.6 0.4 0.3 0.7 0.1

=== MTR to 64.186.132.65 (FiberLight) ===
Start: 2026-04-07T14:36:56
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.4 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.1 3.8 1.7 38.7 6.7

wii747 · April 7

Email just received

Hey everyone,

As you are probably already aware, there was an outage lasting up to approximately 7 days, depending on your node.

We feel that we owe everyone a clear explanation of what happened, what we are doing to prevent this from happening again, and what compensation will be offered.

Timeline of Events

Originally planned

We originally had a migration planned in February to take place in early to mid March.
The plan was to provide about a week of notice and schedule the outage in advance.
However, we had a hard deadline of March 31st that unfortunately could not be changed.
March 30

Due to unforeseen circumstances, the upstream provider and the new datacenter were not available until March 30.
Because of this, we were forced to move forward with the migration immediately, without the testing and preparation time we had originally planned.
Unfortunately, due to issues with our on-site team, we were unable to send an email notice before the network was removed.
Expected outage vs. actual start

The outage was originally expected to last around 6 hours.
However, the de-racking process took 3 hours longer than expected.
After arrival in Atlanta

Once we arrived at the new datacenter in Atlanta, the same upstream provider that caused the original delay deployed the wrong fiber line to us.
Because of that, we had to urgently source and ship the correct fiber to the new datacenter, but by that point it was already around 1:00 AM.
Attempts to obtain the needed part

We placed 4 separate orders for the required part:

2 Amazon orders
1 warehouse pickup order for the next morning
1 direct warehouse delivery using a private courier
Unfortunately:

Both Amazon orders were delayed and did not arrive until 2 days later
The warehouse pickup order was canceled for an unknown reason
The private courier did not arrive until midnight, at which point our on-site team was no longer present
April 2

On the morning of 4/2, we finally had the needed part and rushed to get it installed.
However, our upstream provider had still not provided our BGP session, which meant we were unable to announce our IP space.
A bit later, we were able to bring our websites and some dedicated server customers online through a tunnel.
However, most of the network was still unstable, and we still could not finish bringing all nodes online.
April 2 Late Afternoon

The new upstream provider had skipped the BGP portion of the deployment, so the issue had to be sent back to their deployment team.
Our BGP session was not brought up until 4/4.
April 3 through April 6

After that, we began gradually restoring services.
All dedicated server customers were back online by 4/5.
Most VPS customers were back online by 4/6.
At this point, all customers are now back online.
IP Changes

Unfortunately, during the move, our prefix 38.247.3.0/24 was removed. As a result, all customers who were using that prefix have been moved to other available prefixes.

SLA Credits

SLA credits will be issued upon opening a ticket, based on your current plan and remaining time.

Please open a ticket for details. We also ask that customers avoid opening urgent tickets unless absolutely necessary, as if more than 50 tickets are opened in the same category, additional tickets may be automatically dropped.****

RAVNIX · April 7

@full_bars said:
Here are some MTR results from my two servers. A few things worth noting:
Quad9 trombones out to Equinix in Ashburn rather than peering locally in Atlanta, adding ~27ms of unnecessary latency. Not a FiberLight/Telia issue, just Quad9's Atlanta presence being weak.

Georgia Tech routes through SOX.net (Southern Crossroads), Atlanta's regional research/education exchange. Base latency is 2-3ms, traffic stays fully local. High loss% on some intermediate hops is ICMP rate limiting, not real loss.

Server 1
=== MTR to 1.1.1.1 ===
Start: 2026-04-07T14:30:26
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 2.0% 50 0.7 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 1.8 2.5 1.6 22.8 3.1
3.|-- 67.204.31.68 0.0% 50 2.2 2.0 1.8 2.5 0.1
4.|-- atl-b4-link.ip.twelve99.n 68.0% 50 2.2 2.9 2.1 11.0 2.2
5.|-- atl-bb1-link.ip.twelve99. 0.0% 50 2.3 2.3 2.0 3.3 0.2
6.|-- atl-b5-link.ip.twelve99.n 0.0% 50 2.1 2.0 1.7 3.4 0.2
7.|-- cloudflare-ic-381672.ip.t 0.0% 50 2.0 2.0 1.8 2.3 0.1
8.|-- 108.162.235.127 0.0% 50 2.8 4.9 2.3 20.8 4.7
9.|-- one.one.one.one 0.0% 50 2.2 2.2 2.0 2.5 0.1

=== MTR to 8.8.8.8 ===
Start: 2026-04-07T14:31:33
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 16.0% 50 0.5 0.5 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.3 3.3 1.6 33.2 5.9
3.|-- 67.204.31.68 0.0% 50 3.8 2.0 1.7 3.8 0.3
4.|-- atl-b4-link.ip.twelve99.n 90.0% 50 2.1 2.1 1.9 2.6 0.2
5.|-- 192.178.68.246 0.0% 50 14.5 14.5 14.0 18.4 0.8
6.|-- 172.253.71.211 0.0% 50 2.5 2.2 2.0 2.8 0.1
7.|-- 142.251.241.187 0.0% 50 13.8 13.6 13.4 13.9 0.1
8.|-- dns.google 0.0% 50 2.2 2.0 1.8 2.3 0.1

=== MTR to 9.9.9.9 ===
Start: 2026-04-07T14:32:39
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 14.0% 50 0.6 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.0 3.8 1.6 51.9 8.5
3.|-- 67.204.31.64 0.0% 50 22.1 22.9 21.7 54.1 4.9
4.|-- 67.204.31.53 0.0% 50 22.3 22.2 22.0 22.5 0.1
5.|-- 67.204.31.50 0.0% 50 22.2 22.2 22.0 22.4 0.1
6.|-- equinix-ashburn.woodynet. 0.0% 50 32.9 29.8 29.2 32.9 0.7
7.|-- dns9.quad9.net 0.0% 50 30.1 29.2 28.8 32.1 0.5

=== MTR to 130.207.244.244 (Georgia Tech) ===
Start: 2026-04-07T14:33:45
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.5 0.5 0.3 0.9 0.1
2.|-- 64.186.132.65 0.0% 50 3.0 2.7 1.6 25.6 4.0
3.|-- 67.204.31.68 0.0% 50 1.9 1.9 1.6 2.3 0.2
4.|-- 206.71.12.44 0.0% 50 5.0 4.2 2.8 5.3 0.7
5.|-- richgw2-to-trcps.sox.net 2.0% 50 2.0 2.1 1.9 2.4 0.1
6.|-- 172.18.11.17 70.0% 50 2.2 2.4 2.2 2.6 0.1
7.|-- 172.18.6.113 2.0% 50 2.4 2.5 2.2 6.1 0.6
8.|-- 172.18.6.83 82.0% 50 3.4 3.2 2.7 4.4 0.5
9.|-- 10.10.59.149 0.0% 50 3.0 3.0 2.7 3.6 0.2
10.|-- 10.136.72.1 0.0% 50 3.4 3.4 2.9 4.6 0.4
11.|-- ??? 100.0 50 0.0 0.0 0.0 0.0 0.0

=== MTR to 206.71.12.44 (SOX.net) ===
Start: 2026-04-07T14:34:54
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 4.0% 50 0.4 0.5 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 1.8 3.3 1.6 29.1 5.5
3.|-- 67.204.31.68 0.0% 50 1.8 1.9 1.7 2.4 0.1
4.|-- 206.71.12.44 0.0% 50 3.0 4.1 2.8 5.6 0.7

=== MTR to 74.50.11.1 (gateway) ===
Start: 2026-04-07T14:35:59
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.4 0.4 0.3 0.8 0.1

=== MTR to 64.186.132.65 (FiberLight) ===
Start: 2026-04-07T14:36:55
HOST: US-ATL-01 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.4 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.3 3.6 1.7 32.7 5.9

Server 2
=== MTR to 1.1.1.1 ===
Start: 2026-04-07T14:30:29
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 4.0% 50 0.6 0.4 0.3 1.0 0.1
2.|-- 64.186.132.65 0.0% 50 27.3 7.8 1.5 38.0 11.8
3.|-- 67.204.31.68 0.0% 50 1.8 2.0 1.7 5.1 0.5
4.|-- atl-b4-link.ip.twelve99.n 74.0% 50 2.1 2.2 2.0 2.6 0.2
5.|-- atl-bb1-link.ip.twelve99. 6.0% 50 2.2 2.2 2.0 3.6 0.2
6.|-- atl-b5-link.ip.twelve99.n 0.0% 50 2.1 2.0 1.8 3.3 0.3
7.|-- cloudflare-ic-381672.ip.t 0.0% 50 2.0 1.9 1.7 2.3 0.1
8.|-- 108.162.235.133 0.0% 50 2.2 4.1 2.1 17.5 3.4
9.|-- one.one.one.one 0.0% 50 2.2 2.2 1.9 2.5 0.1

=== MTR to 8.8.8.8 ===
Start: 2026-04-07T14:31:37
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 24.0% 50 0.3 0.4 0.3 0.8 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 3.6 1.6 48.0 7.7
3.|-- 67.204.31.68 0.0% 50 1.9 2.0 1.7 2.5 0.2
4.|-- atl-b4-link.ip.twelve99.n 64.0% 50 2.4 2.3 2.0 4.0 0.4
5.|-- 192.178.68.246 0.0% 50 14.3 14.5 14.0 17.5 0.7
6.|-- 172.253.71.211 0.0% 50 13.5 13.6 13.4 14.1 0.1
7.|-- 142.251.55.119 0.0% 50 13.6 13.6 13.4 14.5 0.2
8.|-- dns.google 0.0% 50 2.0 2.1 1.9 2.8 0.1

=== MTR to 9.9.9.9 ===
Start: 2026-04-07T14:32:42
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 6.0% 50 0.5 0.4 0.3 0.6 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 3.1 1.6 40.1 6.0
3.|-- 67.204.31.64 0.0% 50 21.9 24.5 21.7 87.2 11.5
4.|-- 67.204.31.53 0.0% 50 21.9 22.1 21.9 22.8 0.2
5.|-- 67.204.31.50 0.0% 50 22.0 22.1 21.9 22.4 0.1
6.|-- equinix-ashburn.woodynet. 0.0% 50 29.5 29.6 29.3 31.0 0.4
7.|-- dns9.quad9.net 0.0% 50 29.0 29.0 28.8 29.4 0.1

=== MTR to 130.207.244.244 (Georgia Tech) ===
Start: 2026-04-07T14:33:49
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 4.0% 50 0.6 0.4 0.3 0.8 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 2.1 1.6 8.1 1.3
3.|-- 67.204.31.68 0.0% 50 2.2 1.9 1.7 2.4 0.1
4.|-- 206.71.12.44 0.0% 50 3.3 4.5 2.5 6.7 0.9
5.|-- richgw2-to-trcps.sox.net 2.0% 50 2.2 2.0 1.9 2.2 0.1
6.|-- 172.18.11.17 64.0% 50 2.4 2.3 2.2 2.6 0.1
7.|-- 172.18.6.113 2.0% 50 2.5 2.4 2.1 3.4 0.2
8.|-- 172.18.6.83 84.0% 50 2.8 3.0 2.8 3.2 0.1
9.|-- 10.10.59.149 0.0% 50 49.7 5.1 2.6 54.6 9.8
10.|-- 10.136.72.1 2.0% 50 4.7 4.0 2.8 16.2 2.2
11.|-- ??? 100.0 50 0.0 0.0 0.0 0.0 0.0

=== MTR to 206.71.12.44 (SOX.net) ===
Start: 2026-04-07T14:34:55
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 2.0% 50 0.4 0.4 0.3 0.8 0.1
2.|-- 64.186.132.65 0.0% 50 1.7 3.6 1.5 45.9 7.8
3.|-- 67.204.31.68 0.0% 50 1.6 1.9 1.6 2.5 0.2
4.|-- 206.71.12.44 0.0% 50 5.2 3.9 2.0 5.7 0.9

=== MTR to 74.50.11.1 (gateway) ===
Start: 2026-04-07T14:36:00
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.6 0.4 0.3 0.7 0.1

=== MTR to 64.186.132.65 (FiberLight) ===
Start: 2026-04-07T14:36:56
HOST: US-ATL-02 Loss% Snt Last Avg Best Wrst StDev
1.|-- _gateway 0.0% 50 0.4 0.4 0.3 0.7 0.1
2.|-- 64.186.132.65 0.0% 50 2.1 3.8 1.7 38.7 6.7

Please open a ticket about the latency to 9.9.9.9 and we'll see if we can do anything about it.

KeyGenMe · April 7

@Ravnix

I received your recent email regarding the outage, where you explained the situation and provided a detailed timeline of events.

Thank you for the transparency, but I have to be honest — a downtime of up to 7 days without any prior notice is simply unacceptable for a production service. This has had a direct negative impact on my projects and operations.

For those who may not be aware, this was related to a datacenter migration that, according to your own statement, was originally planned in advance but ended up being executed under pressure and without proper preparation or customer notification.

What is especially concerning is that this was a planned migration, yet it was carried out without sufficient testing, coordination, or even basic communication with customers.

From a customer perspective, this reflects serious issues in planning, coordination, and overall reliability.

While I will be opening a ticket for SLA credits, I would also like to ask publicly: what additional compensation is being considered, given the scale and duration of this outage?

I sincerely hope that concrete improvements will be made, because situations like this significantly undermine trust in your service.

Saragoldfarb · April 7

My IP 130.12.60.xxx still very much dead.

Saragoldfarb · April 7

Wait... My IP changed...

Saragoldfarb · April 7

Ah, rebooted it. Using new ip it seems to be online now.

RAVNIX · April 7

@KeyGenMe said:
@Ravnix

I received your recent email regarding the outage, where you explained the situation and provided a detailed timeline of events.

Thank you for the transparency, but I have to be honest — a downtime of up to 7 days without any prior notice is simply unacceptable for a production service. This has had a direct negative impact on my projects and operations.

For those who may not be aware, this was related to a datacenter migration that, according to your own statement, was originally planned in advance but ended up being executed under pressure and without proper preparation or customer notification.

What is especially concerning is that this was a planned migration, yet it was carried out without sufficient testing, coordination, or even basic communication with customers.

From a customer perspective, this reflects serious issues in planning, coordination, and overall reliability.

While I will be opening a ticket for SLA credits, I would also like to ask publicly: what additional compensation is being considered, given the scale and duration of this outage?

I sincerely hope that concrete improvements will be made, because situations like this significantly undermine trust in your service.

Open a ticket and mention this message and ill push for more compensation internally

Saragoldfarb · April 7


# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2025-04-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Tue Apr  7 23:51:16 BST 2026

Basic System Information:
---------------------------------
Uptime     : 0 days, 0 hours, 1 minutes
Processor  : AMD EPYC 7702 64-Core Processor
CPU cores  : 1 @ 1996.247 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 913.8 MiB
Swap       : 1024.0 MiB
Disk       : 9.6 GiB
Distro     : Ubuntu 24.04.3 LTS
Kernel     : 6.8.0-107-generic
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : Ravnix LLC
ASN        : AS401486 RAVNIX LLC
Host       : Ravnix LLC
Location   : Denver, Colorado (CO)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 112.18 MB/s  (28.0k) | 466.88 MB/s   (7.2k)
Write      | 112.48 MB/s  (28.1k) | 469.34 MB/s   (7.3k)
Total      | 224.66 MB/s  (56.1k) | 936.22 MB/s  (14.6k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 536.04 MB/s   (1.0k) | 507.64 MB/s    (495)
Write      | 564.53 MB/s   (1.1k) | 541.45 MB/s    (528)
Total      | 1.10 GB/s     (2.1k) | 1.04 GB/s     (1.0k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 774 Mbits/sec   | 162 Mbits/sec   | 95.9 ms
Eranium         | Amsterdam, NL (100G)      | 859 Mbits/sec   | 545 Mbits/sec   | 104 ms
Uztelecom       | Tashkent, UZ (10G)        | 620 Mbits/sec   | 256 Mbits/sec   | 201 ms
Leaseweb        | Singapore, SG (10G)       | 591 Mbits/sec   | 137 Mbits/sec   | 247 ms
Clouvider       | Los Angeles, CA, US (10G) | 999 Mbits/sec   | 283 Mbits/sec   | 68.5 ms
Leaseweb        | NYC, NY, US (10G)         | 918 Mbits/sec   | 239 Mbits/sec   | 23.0 ms
Edgoo           | Sao Paulo, BR (1G)        | 631 Mbits/sec   | 218 Mbits/sec   | 123 ms

iperf3 Network Speed Tests (IPv6):                                                                    ---------------------------------                                                                     Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping                -----           | -----                     | ----            | ----            | ----                Clouvider       | London, UK (10G)          | 990 Mbits/sec   | 264 Mbits/sec   | 96.6 ms
Eranium         | Amsterdam, NL (100G)      | 648 Mbits/sec   | 424 Mbits/sec   | 104 ms
Uztelecom       | Tashkent, UZ (10G)        | 856 Mbits/sec   | 343 Mbits/sec   | 201 ms
Leaseweb        | Singapore, SG (10G)       | 658 Mbits/sec   | 263 Mbits/sec   | 242 ms
Clouvider       | Los Angeles, CA, US (10G) | 1.00 Gbits/sec  | 208 Mbits/sec   | 61.8 ms
Leaseweb        | NYC, NY, US (10G)         | 1.10 Gbits/sec  | 326 Mbits/sec   | 22.9 ms
Edgoo           | Sao Paulo, BR (1G)        | 1.01 Gbits/sec  | 296 Mbits/sec   | 123 ms

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 800
Multi Core      | 516
Full Test       | https://browser.geekbench.com/v6/cpu/17519571

YABS completed in 27 min 39 sec

aphex · April 8

If I bought specifically for the niche location in South Carolina, why would I want common Atlanta where I can get better deal elsewhere?

zed · April 8

well now i'm conflicted.
i'm annoyed at how this was handled, and i was already annoyed with ravnix for previous nonsense.
i bought one of their promos because it was SC and that was unusual-ish.

but atl is also a great location for me.

damn.

aphex · April 8

Unfortunate. 38. IP was also much cleaner and less captchas compared to the incredibly dirty reputation IPXO rentals too.

full_bars · April 8

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2025-04-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Tue Apr  7 16:53:24 PDT 2026

Basic System Information:
---------------------------------
Uptime     : 0 days, 2 hours, 46 minutes
Processor  : AMD Ryzen 9 9950X 16-Core Processor
CPU cores  : 1 @ 4291.916 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 1.9 GiB
Swap       : 4.8 GiB
Disk       : 19.3 GiB
Distro     : Ubuntu 24.04.3 LTS
Kernel     : 6.18.1-061801-generic
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : Ravnix LLC
ASN        : AS401486 RAVNIX LLC
Host       : Ravnix LLC
Location   : Denver, Colorado (CO)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 228.39 MB/s  (57.0k) | 2.40 GB/s    (37.6k)
Write      | 228.99 MB/s  (57.2k) | 2.42 GB/s    (37.8k)
Total      | 457.39 MB/s (114.3k) | 4.83 GB/s    (75.5k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 6.20 GB/s    (12.1k) | 4.20 GB/s     (4.1k)
Write      | 6.53 GB/s    (12.7k) | 4.48 GB/s     (4.3k)
Total      | 12.74 GB/s   (24.8k) | 8.68 GB/s     (8.4k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 1.71 Gbits/sec  | 1.19 Gbits/sec  | 95.3 ms
Eranium         | Amsterdam, NL (100G)      | 1.45 Gbits/sec  | 1.48 Gbits/sec  | 109 ms
Uztelecom       | Tashkent, UZ (10G)        | 600 Mbits/sec   | 376 Mbits/sec   | 202 ms
Leaseweb        | Singapore, SG (10G)       | 594 Mbits/sec   | 610 Mbits/sec   | 244 ms
Clouvider       | Los Angeles, CA, US (10G) | 2.28 Gbits/sec  | 1.35 Gbits/sec  | 61.9 ms
Leaseweb        | NYC, NY, US (10G)         | 5.59 Gbits/sec  | 2.91 Gbits/sec  | 26.7 ms
Edgoo           | Sao Paulo, BR (1G)        | 784 Mbits/sec   | 672 Mbits/sec   | 119 ms

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 1.68 Gbits/sec  | 1.34 Gbits/sec  | 95.3 ms
Eranium         | Amsterdam, NL (100G)      | 1.47 Gbits/sec  | 1.51 Gbits/sec  | 109 ms
Uztelecom       | Tashkent, UZ (10G)        | 737 Mbits/sec   | 507 Mbits/sec   | 202 ms
Leaseweb        | Singapore, SG (10G)       | 589 Mbits/sec   | 557 Mbits/sec   | --
Clouvider       | Los Angeles, CA, US (10G) | 2.75 Gbits/sec  | 1.36 Gbits/sec  | 61.8 ms
Leaseweb        | NYC, NY, US (10G)         | 5.69 Gbits/sec  | 2.82 Gbits/sec  | 26.7 ms
Edgoo           | Sao Paulo, BR (1G)        | 1.26 Gbits/sec  | busy            | 119 ms

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 2248
Multi Core      | 1707
Full Test       | https://browser.geekbench.com/v6/cpu/17519977

YABS completed in 14 min 53 sec

tenpera · April 8

@RAVNIX, thanks.

RAVNIX · April 8

@aphex said:
If I bought specifically for the niche location in South Carolina, why would I want common Atlanta where I can get better deal elsewhere?

We understand, if you would like a refund open a ticket.

Unfortunately the proper business decision was to move away from Fairfax. The datacenter was falling apart (they had coolant leaking on our racks) and Cogent refused to handle anything properly and increased power costs with no warning at all (increased by 23%, while already being double normal datacenters).
Being stuck to single-homed cogent was also a heavy downside and no other upstream was willing to even consider getting into the datacenter (even ones with lines going through the town).

Moving to Atlanta allows us to have much better transit and expansion into more premium services rather being limited to a cogent datacenter in a town in the middle of nowhere.

@aphex said:
Unfortunate. 38. IP was also much cleaner and less captchas compared to the incredibly dirty reputation IPXO rentals too.

We could move you to our 130.12.60.0/24 subnet if you open a ticket.

full_bars · April 8

I for one am absolutely thrilled about the move to Atlanta. This is a massive upgrade for me and my use case. Yes it's unfortunate the way it happened but I'm really not upset. Other hosts have absolutely pissed me the fuck off with a complete lack of customer service (cough cough Valornode) but Ravnix is actually trying to do right by their customers and that goes a long way with me.

springai · April 8

@RAVNIX said:

@aphex said:
If I bought specifically for the niche location in South Carolina, why would I want common Atlanta where I can get better deal elsewhere?

We understand, if you would like a refund open a ticket.

Unfortunately the proper business decision was to move away from Fairfax. The datacenter was falling apart (they had coolant leaking on our racks) and Cogent refused to handle anything properly and increased power costs with no warning at all (increased by 23%, while already being double normal datacenters).
Being stuck to single-homed cogent was also a heavy downside and no other upstream was willing to even consider getting into the datacenter (even ones with lines going through the town).

Moving to Atlanta allows us to have much better transit and expansion into more premium services rather being limited to a cogent datacenter in a town in the middle of nowhere.

@aphex said:
Unfortunate. 38. IP was also much cleaner and less captchas compared to the incredibly dirty reputation IPXO rentals too.

We could move you to our 130.12.60.0/24 subnet if you open a ticket.

I sent you a private message. Check it.

zed · April 8

well i've requested a refund, i guess we'll see what happens.

this entire process has been just weird.
the explanation for why we weren't notified makes no sense at all.
i only bought the service because it was in south carolina.

Quazi · April 9

I've initiated a cancellation myself. Whether my service is back up or not. I think in my experience, there was a clear omission of truth especially with the responses I got from my support tickets to report my service being down. The email and what RAVNIX posted here was just a poor attempt to save face. Their website was never down throughout this and they had control over ravnix.gg and could have sent their customers emails prior and during this ordeal that they had. I just hope they honor my cancellation/refund request. I'm not losing much and probably can just eat those costs but I've currently have services with other hosts that I've continued to be with for years w/o issues like this. I'm starting another service with them in light of this b.s. fiasco.

Quazi · April 9

I'm pasting this here for LET community posterity.

Internal Management
Staff
Apr 9, 2026, 03:33 AM
You were sent an email regarding the outage, there was no point of pasting the same thing into the ticket so instead you were just told to check (after it was online again).

Unfortunately due to how our ticket system works we're unable to have more than 50 tickets per category so we are trying to close tickets as fast as possible to avoid tickets being dropped and not processed on our end.

Either way I have sent your request to billing and it'll be processed within 7-10 business days.

Client
Apr 9, 2026, 03:46 AM
Look, I don't want to get into a pissing match with whomever replied.

1) I generated the tickets BEFORE I received any official email from RAVNIX alerting their customers about the outage. In fact, I saw RAVNIX post on LET before the email as well.

2) It doesn't matter on your "limitation" on tickets per category, it means nothing due to the circumstances there. You guys failed to notify your customers, period. Do you see the responses I got when my VPS first went down? "We've encountered a catastrophic failure on the hypervisor your server is residing on. We've now fixed the issue and booted your server for you."

A catasrophic failure on hypervisor. You expect me to believe this after what was stated in the email and on LET? Give me a f'ing break. I don't buy this b.s.

I have a screenshot of the time/day stamp of the email. This ticket response is just emotion on your part. Yeah, either way, I look forward to getting a response from your billing department. Sorry you all had to go through this but RAVNIX had an opportunity to set things and the record straight and failed.

RAVNIX LLC no-reply@ravnix.gg
Tue, Apr 7, 10:16 PM (2 days ago)

TEC-46767325-949
Created: 3/29/2026

GEN-29622880-033
Created: 4/4/2026

KeyGenMe · April 9

@zed, any updates on your side? Did you get the refund or at least some kind of response?

@Ravnix, I reached out to support right after your message, but it’s been two days and still nothing. Is that normal? How long does it usually take — like a week, a month? And do they actually reply at all?

aphex · April 9

@Quazi said: Unfortunately due to how our ticket system works we're unable to have more than 50 tickets per category so we are trying to close tickets as fast as possible to avoid tickets being dropped and not processed on our end.

ai slop ticket system?

zed · April 9

@KeyGenMe said: @zed, any updates on your side? Did you get the refund or at least some kind of response?

Yes I've received my refund.

LowEndSupport · April 13

@KeyGenMe said:
@zed, any updates on your side? Did you get the refund or at least some kind of response?

@Ravnix, I reached out to support right after your message, but it’s been two days and still nothing. Is that normal? How long does it usually take — like a week, a month? And do they actually reply at all?

I reached out as well since zed's post for a refund and it's been 5 days now with no reply. Clientarea.gg is also down so can't even check.

KeyGenMe · April 13

Just wanted to add an update to my previous post — I’ve received compensation from @Ravnix, which I appreciate.

I’m hopeful that after the move to the new data center, things will finally settle down and we’ll see a period of stability and smooth operation. Fingers crossed for a calmer and more reliable experience going forward!

Quazi · April 14

I have not heard back from them since April 9th and getting pretty annoyed that you all have gotten refunds/compensations and I'm still in limbo.

https://supportarea.gg/ seems to be up but getting an "Unable to connect to authentication service" when I try to log in.

https://clientarea.gg/ appears to be down.

@RAVNIX What's going on this time?

Saragoldfarb · April 14

@Quazi said:
I have not heard back from them since April 9th and getting pretty annoyed that you all have gotten refunds/compensations and I'm still in limbo.

https://supportarea.gg/ seems to be up but getting an "Unable to connect to authentication service" when I try to log in.

https://clientarea.gg/ appears to be down.

@RAVNIX What's going on this time?

Down again? Thats a annoying. My server seems to be up fortunately.

Quazi · April 16

Can anyone who still has service with Ravnix please check to see if they can log into the client portal? Because I cannot. I haven't gotten any word on my ticket or refund.

JohnnySac · April 16

@Quazi said:
Can anyone who still has service with Ravnix please check to see if they can log into the client portal? Because I cannot. I haven't gotten any word on my ticket or refund.

I get 502 Bad Gateway for clientarea.gg.

tenpera · April 17

everything is fine.

Howdy, Stranger!

Categories

In this Discussion

RAVNIX offline?

Comments

Howdy, Stranger!

Quick Links

Categories

In this Discussion

RAVNIX offline?

Comments