http://mirror.0x.sg/

[andr3wyong]

Is the mirror down?


edit
Its back up

I moved house, just moved my SI 10G line over to my new place today. Didn’t expect such a long downtime, it was due to high loss from NLT.

 

[dryteletubby]

Down again?

 

[andr3wyong]

Down again?

SI WAN down not sure if my switch died or if it’s due to high loss again.

 

[dryteletubby]

SI WAN down not sure if my switch died or if it’s due to high loss again.

It's back up

Maybe next time you should have redundancy

 

[andr3wyong]

It's back up

Maybe next time you should have redundancy

Because paying $150/month isn't enough?

Edit: It was due to NLT making a mistake and changing patching without informing ahead.

 

[andr3wyong]

It's back up

Maybe next time you should have redundancy

I’ve actually self-funded the UPS, ATS and other relevant infra to keep the power stable. However WAN redundancy isn't so straightforward since any other uplink would have a different IP.

 

[CoolRock]

It's back up

Maybe next time you should have redundancy

Maybe next time you should help fund the redundancy line

 

[dryteletubby]

Because paying $150/month isn't enough?

Edit: It was due to NLT making a mistake and changing patching without informing ahead.

I’ve actually self-funded the UPS, ATS and other relevant infra to keep the power stable. However WAN redundancy isn't so straightforward since any other uplink would have a different IP.

I assumed it was your switch. Apparently NLT was being unreliable.

 

[davidktw]

I’ve actually self-funded the UPS, ATS and other relevant infra to keep the power stable. However WAN redundancy isn't so straightforward since any other uplink would have a different IP.

Appreciate your contribution. Just a possible suggest if you do really want to make ur uplink redundant. Anycast would be very feasible redundancy but that would be a different layer of routing. Alternative is have ur DNS setup with a low TTL like 5mins?

So if any of your primary uplink is down by pinging against an external probe, your internal probe can update the DNS record via ur another uplink. This can be achieve is ur DNS name servers service provider offer some kind of API for such updates.

I guess ur secondary uplink doesn’t need to be 10Gbps

Another way is using of Amazon Route 53 which offers health check for failover. Works with external public IPs. Last way is simply use a dynamic IP service

 

[andr3wyong]

thanks everyone for your suggestions. I've considered it but I'm not willing to sink the recurring costs in. My SI line had actually been consistently stable for over a year.

The recent spate of downtime was because I moved house and physically relocated the server. Subsequently we (SI and myself) found excessive bit error rate on my line. Due to investigation by NLT there were occasions (like today) where NLT made changes to the fiber splicing without adhering to the appointed time with SuperInternet.

In any case, other than rectification works on the fiber I'm not expecting any prolonged reliability issues.

In fact I've taken my moving house to also replace the core switch with a much newer Juniper.

 

[dryteletubby]

Mirror down?

 

[andr3wyong]

Migrating it to new hardware, give it another 2 hours

 

[chankl78]

Good Day

Migrating it to new hardware, give it another 2 hours

Good Day sir.

Will you be updating Kali Linux to 2019?

Thanks.

Smile
Chankl78

 

[rtwair]

I have been a user of your arch linux mirror for some time and really appreciate the reliability & performance. After trying hard to find a reliable freebsd pkg mirror (currently switching back and fourth between several in Asia) I am wondering whether you still have any plans to host a freebsd mirror?

Thanks!