FYI - Several Embarcadero services are currently unavailable

[Vincent Parrett 657]

On 1/18/2024 at 9:02 AM, Vincent Parrett said:

The cloud isn't always cheaper (been there, done that, bought new servers). To make the cloud cheaper you really need to leverage the features they provide (not just a bunch of vm's) - and that means re-architecting things (=time + more money) and it also means vendor lock in - once you leverage the cloud features it's not easy to switch providers. 

By pure chance this video came up on youtube this morning which articulates the point I made above ^
 

 

 

My guess is embarcadero are simply spinning up new vm's and not re-architecting their services - they might be in for a cloud bill shock in the future 🤔

[Lars Fosdal 1732]

@Vincent Parrett - I agree.  Simply moving to a cloud hosted VM, running your old services, will be expensive as heck. We have metrics to support that statistic.

 

[David Heffernan 2300]

1 hour ago, Lars Fosdal said:

@Vincent Parrett - I agree.  Simply moving to a cloud hosted VM, running your old services, will be expensive as heck. We have metrics to support that statistic.

 

What's the reputational cost of the current fiasco? 

[Lars Fosdal 1732]

21 minutes ago, David Heffernan said:

What's the reputational cost of the current fiasco? 

Are you thinking about operational cost of new solution vs reputational cost of outage and lack of info?
The suits would probably say those are two different budgets 😛

My personal opinion on the outage and recovery time: priceless, but not in a good way.
As for operational cost: In the context of @Vincent Parrett's observation, if they simply recreated their VMs - that is far more costly than running cloud adapted services, so it will mean a steep increase in operational cost if that solution is maintained for the long run. In the short run, it makes sense in the context of minimizing the reputational cost issue.
 

[Anders Melander 1639]

7 hours ago, Vincent Parrett said:

By pure chance this video came up on youtube this morning which articulates the point I made above ^

I'm sure the scenarios he outlined as a reason for moving to the cloud (basically cost of hardware) and then moving off it again is true for a lot of cases. It doesn't however resonate with any of the scenarios I've been or am involved with; We've moved systems to the cloud not because of hardware cost (which I can't remember having ever been an issue) but because of the cost of managing on-premise servers, providing the required infrastructure (storage, network, recoverability, know-how, etc) and securing global availability.

I think the size of a company is a big factor in whether cloud makes sense or not. If you're already paying for the "overhead" of having dedicated, competent IT staff service your on-premise clients, then it might not make sense to outsource stuff to the cloud. Of course the nature of the services being outsourced also has to be taken into consideration, but I can't see hardware cost being a factor.

[PeterPanettone 154]

Is it worth renewing my Delphi subscription?

[Dalija Prasnikar 1341]

1 hour ago, PeterPanettone said:

Is it worth renewing my Delphi subscription?

Only you know that.

 

But, if you are asking is Delphi dead, then the answer is No, it isn't. This was just a hiccup (even though, a huge one), and the resolution is just around the corner.

[Darian Miller 336]

3 hours ago, Dalija Prasnikar said:

the resolution is just around the corner. 

As far as we know, it could be another 3 weeks.  They could have screwed up the replacement system and it's not ready for the spike in traffic it is about to receive.   It certainly can be difficult to validate new systems.

17,000 views of their blog post and it hasn't been updated since Jan 25th.   That is truly astonishing.  It makes zero sense to me.

[Lars Fosdal 1732]

I have to agree. Their information - or rather, lack thereof - is really bad for their reputation.

 

[Uwe Raabe 1890]

11 minutes ago, Lars Fosdal said:

Their information - or rather, lack thereof - is really bad for their reputation.

Apparently the minds in charge think different.

[Silver Black 23]

4 hours ago, Dalija Prasnikar said:

Only you know that.

 

But, if you are asking is Delphi dead, then the answer is No, it isn't. This was just a hiccup (even though, a huge one), and the resolution is just around the corner.

Nonetheless, even if I love Delphi and I have a very big project with it, I started learning C# and Visual Studio Community...

[Vandrovnik 202]

1 hour ago, Lars Fosdal said:

I have to agree. Their information - or rather, lack thereof - is really bad for their reputation.

 

Lack of information, inability to fix in a reasonable time and inability to fix the "integer division bug" that makes Delphi 12 unusable for Win32 applications 😞

[Dalija Prasnikar 1341]

25 minutes ago, Vandrovnik said:

inability to fix the "integer division bug" that makes Delphi 12 unusable for Win32 applications 😞

You mean the issue with division with $FFFFFFFF constant? That is nowhere near being a showstopper bug. As a workaround it is possible to use variable instead of constant. And this issue is fixed, but at the moment hotfix cannot be delivered due to outage.

Edited February 5 by Dalija Prasnikar

[Anders Melander 1639]

1 hour ago, Dalija Prasnikar said:

And this issue is fixed, but at the moment hotfix cannot be delivered due to outage.

"the check is in the mail" 🙂

 

1 hour ago, Dalija Prasnikar said:

You mean the issue with division with $FFFFFFFF constant? That is nowhere near being a showstopper bug.

It is for us. Some of our projects are huge, with lots of integer math optimizations involving bit shifts and funky multiplications and divisions and we simply cannot risk it when the compiler has known bugs in that area. So basically Delphi 12 has been DOA for us.

[Darian Miller 336]

Delphi 12 seems to have more than a normal amount of regressions.  I've listed some of them on its wiki page on GitHub:

https://github.com/ideasawakened/DelphiKB/wiki/D29.ATHENS.12.0.0.0

 

GetIt being down stops fixes for these regressions being released.  There aren't workarounds for all of these issues.

 

 

[Vincent Parrett 657]

14 hours ago, Lars Fosdal said:

@Vincent Parrett - I agree.  Simply moving to a cloud hosted VM, running your old services, will be expensive as heck. We have metrics to support that statistic.

I went through the same exercise last year with our internal systems (CI etc) and it absolutely made sense to purchase new servers and put them in a data center (we closed our office when covid hit and went fully remote). Server tech has evolved a lot since I last purchased some (7yrs ago) - you can get some incredible bang for buck - all nvme lots of cores & lots of ram for AU$20K each.   

[Vandrovnik 202]

2 hours ago, Dalija Prasnikar said:

You mean the issue with division with $FFFFFFFF constant? That is nowhere near being a showstopper bug. As a workaround it is possible to use variable instead of constant. And this issue is fixed, but at the moment hotfix cannot be delivered due to outage.

I mean this one:

https://quality.embarcadero.com/browse/RSP-43274 (Arithmetic operations on record fields return incorrect results in certain cases if the "Optimization" compiler option is enabled.)

This looks like a showstopper bug for me, because it can generate wrong result anywhere in the application. Turning off optimizations will "solve" the problem, but the result will be that all my customers will write to me about how the new version of the app is slow.

 

And as for distribution of the hotfix - is it really impossible to distribute the fix in year 2024 other than through a non-functioning Getit? Downloading files from my.embarcadero.com works normally, so why not put the patch there?

 

[Dalija Prasnikar 1341]

9 hours ago, Vandrovnik said:

I mean this one:

https://quality.embarcadero.com/browse/RSP-43274 (Arithmetic operations on record fields return incorrect results in certain cases if the "Optimization" compiler option is enabled.)

This looks like a showstopper bug for me, because it can generate wrong result anywhere in the application. Turning off optimizations will "solve" the problem, but the result will be that all my customers will write to me about how the new version of the app is slow.

I somehow missed that one. Yes, this one looks serious. I don't know what is the status of this issue.

 

9 hours ago, Vandrovnik said:

And as for distribution of the hotfix - is it really impossible to distribute the fix in year 2024 other than through a non-functioning Getit? Downloading files from my.embarcadero.com works normally, so why not put the patch there?

I thought that my.embarcadero.com was also impacted by the outage. Anyway, I phrased my sentence a bit awkwardly, like the outage could be the only reason why fixes are not released yet as patch also needs to go through testing phase before it is finally released.

[Sherlock 632]

16 hours ago, Lars Fosdal said:

I have to agree. Their information - or rather, lack thereof - is really bad for their reputation.

 

There is fitting german proverb for exactly this

"Ist der Ruf erst runiniert, lebt es sich gänzlich ungeniert"

Which roughly translates to

"Once the reputation is shot you can live quite unabashedly"

 

Soooo... I'm guessing they are headed in that direction. 😄

[Lars Fosdal 1732]

I hope not. I hope they get their house in order ASAP.

[Sherlock 632]

10 minutes ago, Lars Fosdal said:

I hope not. I hope they get their house in order ASAP.

Yeah, I really hope so too. But having read the word reputation at least twice in this read just triggered the association.

[Skrim 11]

I think I know what is going on.

 

Embarcadero have abanonned GetIt?

[Uwe Raabe 1890]

1 minute ago, Skrim said:

Embarcadero have abanonned GetIt?

Ehm, no!

[Skrim 11]

11 minutes ago, Uwe Raabe said:

Ehm, no!

Why then isn't GetIt up and running?
For a company within the IT industry it should be simple enough, right?
In my industry (finance) we are required to have plans for downtime situations to minimize the damage. Doesn't Embarcadero have plans for such cases? Or is the philosophy, let's just wait and see what happens should something arise 

[PeterPanettone 154]

12 minutes ago, Skrim said:

Why then isn't GetIt up and running?

Perhaps the damage was greater than reported - and maybe the damage also occurred at a different level.