Hosthatch Sydney: IPv6 "broken" for two weeks and counting

add_iT

Mine is not working yet

Can we exchange to other location with working IPv6 which is closest?
Like Hongkong maybe

jerry_me

root@debian:~# traceroute ipv6.google.com
traceroute to ipv6.google.com (2404:6800:4006:811::200e), 30 hops max, 80 byte packets
1 2406:ef80:100::1 (2406:ef80:100::1) 5.654 ms 6.063 ms 6.032 ms
2 2400:b640:0:3::2 (2400:b640:0:3::2) 0.420 ms 0.392 ms 0.368 ms
3 2406:4f40:10:2::1 (2406:4f40:10:2::1) 0.383 ms 0.377 ms 0.412 ms
4 * * *
5 * * *

RIPE => At 2021-07-27 00:00:00 UTC, 2406:ef80:100::/40 was 63% visible (by 210 of 336 RIS full peers).

what a pity.

bulbasaur

@jerry_me said: RIPE => At 2021-07-27 00:00:00 UTC, 2406:ef80:100::/40 was 63% visible (by 210 of 336 RIS full peers).

How did you obtain this information? I suspect IPv6 is broken in AMS too due to some network issues since I have trouble connecting to certain IPv6 only websites.

I'll ticket them later on in the day, but I don't have high hopes.

jerry_me

@stevewatson301 said:

@jerry_me said: RIPE => At 2021-07-27 00:00:00 UTC, 2406:ef80:100::/40 was 63% visible (by 210 of 336 RIS full peers).

How did you obtain this information? I suspect IPv6 is broken in AMS too due to some network issues since I have trouble connecting to certain IPv6 only websites.

I'll ticket them later on in the day, but I don't have high hopes.

ripe stat

brueggus

@hosthatch said: We will likely stop providing native IPv6 there for now (and in some other problem locations). Most upstreams do not deal with IPv6 issues with the same priority, and the time investment that goes into troubleshooting IPv6-related issues in these locations for a -very- small number of customers who actually use it is huge.

Makes sense from a provider's perspective, I guess. Besides my Hosthatch server in Sydney I currently have IPv6 issues on three other servers (not counting those which lose IPv6 connectivity if I don't ping the gateway every few minutes or so):

• Known Host, Atlanta (AS63410)
• Viettel, Hanoi (AS38731)
• Reliable Hosting (AS29713, could be an upstream issue, really weird...)

All of these issues are ongoing for weeks, Viettel even for months - but there's no affordable alternative in VN sadly. I don't expect any of them to be resolved anytime soon.

My servers on these networks are rented from resellers, paid for a year upfront and are in the Lowend pricing segment, so, as @hosthatch said, it doesn't make a lot of sense for a provider to invest time in pushing the upstreams to fix their stuff. Especially if only a few (or only one?) customers are complaining.

From my perspective it's still frustrating. IPv6 isn't any different to IPv4 routing-wise. Just the subnets are sized differently. I don't get why it takes network providers that long to fix IPv6 issues (if they even care at all)

mcgree

@brueggus said:

@hosthatch said: We will likely stop providing native IPv6 there for now (and in some other problem locations). Most upstreams do not deal with IPv6 issues with the same priority, and the time investment that goes into troubleshooting IPv6-related issues in these locations for a -very- small number of customers who actually use it is huge.

Makes sense from a provider's perspective, I guess. Besides my Hosthatch server in Sydney I currently have IPv6 issues on three other servers (not counting those which lose IPv6 connectivity if I don't ping the gateway every few minutes or so):

• Known Host, Atlanta (AS63410)
• Viettel, Hanoi (AS38731)
• Reliable Hosting (AS29713, could be an upstream issue, really weird...)

All of these issues are ongoing for weeks, Viettel even for months - but there's no affordable alternative in VN sadly. I don't expect any of them to be resolved anytime soon.

My servers on these networks are rented from resellers, paid for a year upfront and are in the Lowend pricing segment, so, as @hosthatch said, it doesn't make a lot of sense for a provider to invest time in pushing the upstreams to fix their stuff. Especially if only a few (or only one?) customers are complaining.

From my perspective it's still frustrating. IPv6 isn't any different to IPv4 routing-wise. Just the subnets are sized differently. I don't get why it takes network providers that long to fix IPv6 issues (if they even care at all)

6to4 or 6in4 tunnels are also good if they must be required.

brueggus

@mcgree said:

@brueggus said:

@hosthatch said: We will likely stop providing native IPv6 there for now (and in some other problem locations). Most upstreams do not deal with IPv6 issues with the same priority, and the time investment that goes into troubleshooting IPv6-related issues in these locations for a -very- small number of customers who actually use it is huge.

Makes sense from a provider's perspective, I guess. Besides my Hosthatch server in Sydney I currently have IPv6 issues on three other servers (not counting those which lose IPv6 connectivity if I don't ping the gateway every few minutes or so):

• Known Host, Atlanta (AS63410)
• Viettel, Hanoi (AS38731)
• Reliable Hosting (AS29713, could be an upstream issue, really weird...)

All of these issues are ongoing for weeks, Viettel even for months - but there's no affordable alternative in VN sadly. I don't expect any of them to be resolved anytime soon.

My servers on these networks are rented from resellers, paid for a year upfront and are in the Lowend pricing segment, so, as @hosthatch said, it doesn't make a lot of sense for a provider to invest time in pushing the upstreams to fix their stuff. Especially if only a few (or only one?) customers are complaining.

From my perspective it's still frustrating. IPv6 isn't any different to IPv4 routing-wise. Just the subnets are sized differently. I don't get why it takes network providers that long to fix IPv6 issues (if they even care at all)

6to4 or 6in4 tunnels are also good if they must be required.

No. Just no. I'm aware of a single tunnelbroker that's still actively maintained, and that's HE. HE, however, has the huge disadvantage that you're single-homed HE and therefore I won't be able to reach my backup server which is single-homed Cogent as long as Servarica doesn't get their second upstream back up and running.

bdl

Can confirm the increase in latency between Sydney and LA Hosthatch services since 20th July, looks like packets are going to the Olympics via Tokyo, also noticed route to my Chicago Hosthatch services from Sydney doing the same thing.

brueggus

Device <redacted> recovered from Devices up/down
Severity: critical
  Time elapsed: 1w 3d 19h 15m 49s
  Timestamp: 2021-07-31 05:31:35

Hallelujah! Seems like Psychz have fixed their network. Hope, they'll touch it never again.

digitalwicked

@brueggus said:

Device <redacted> recovered from Devices up/down
Severity: critical
  Time elapsed: 1w 3d 19h 15m 49s
  Timestamp: 2021-07-31 05:31:35

Hallelujah! Seems like Psychz have fixed their network. Hope, they'll touch it never again.

Wow! wasnt expecting after their email.

add_iT

this weird

My product name in panel before is Promotional Package NVMe 8G now it change to Package NVMe 4G, anyone have same issue? , i know it just the name but i afraid there is some impact on my vm

on sydney btw

DP

@add_iT said:
this weird

My product name in panel before is Promotional Package NVMe 8G now it change to Package NVMe 4G, anyone have same issue? , i know it just the name but i afraid there is some impact on my vm

on sydney btw

I don't think there'll be an impact to your VM or invoice, but you can always check to confirm.

digitalwicked

@add_iT said:
this weird

My product name in panel before is Promotional Package NVMe 8G now it change to Package NVMe 4G, anyone have same issue? , i know it just the name but i afraid there is some impact on my vm

on sydney btw

Good pickup, same issue on both VPS in Sydney - resources halved?

digitalwicked

Interestingly under bandwidth, this has remained unchanged however the promo name has 1/4th bandwidth shown.