It's bad enough already, please no.

I think that's the point: poke more holes in its hull so it finally sinks.

I think OP left off a /s tag; I read their post to be suggesting that they'd love to see Atlassian's products go away.

Spot on! ;-)

Carbon steel is also seemingly indestructible; cleaning if something is stuck means deglazing the hot pan.

I love my carbon steel skillets.

Checking in from the St. Louis, MO area, the outage is in effect here. Two lines on separate accounts are down.

Thankfully, WiFi calling seems to be functioning.

I can't wait to see a postmortem on this outage.

Which is related to the PDX area data center that went hard-down earlier this morning.

This could easily be related as that DC seems to have housed a lot of cloud based service infrastructure.

Which data center is down in PDX?

Flexential PDX02, according to the outages mailing list[0].

[0] https://puck.nether.net/pipermail/outages/2023-November/0149...

Gee, that makes it sound like cloud-hosted products are susceptible to data center outages just like products that rely on colocated and contracted hosting. Impossible!

/s

If half the internet is down, your customers won't complain about you being down as well /s

It looks like some of it is available via RECAP, which can be found here.

https://www.courtlistener.com/docket/66718284/wilson-v-booz-...

Beyond that as another poster noted, you'd have to buy the additional files via PACER.

You can also use BackBlaze B2 to push your own backups with whatever software will support it, similarly to how you'd use S3.

I think it depends on which version you're updating from.

I'm still on 14.5.1 and it shows as a 922.4MB download.

Or for redundancy sake, if you're using any kind of sane setup. (Yes, YMMV bigly with this particular idea.)

That's still nothing even if you do that 4x/day.

Also just because you create a 20GB virtual disk does not necessarily mean you're actually writing out 20GB to the disk.

Many SSDs and NVMEs are designed with total drive writes per day in their specs.

What is the wear method you're measuring by and what's the threshold where you're replacing your drives?

Or probably the filesystem decides to do it? The Refs will just ate adjacent 0 and assuming you want a fallocate here.