On 30/03/24 06:42, Mike Spencer wrote:

> PS: Attempts to connect towww.pmoylan.org time out.

A log file suggests that something went wrong about six hours ago. I've
now rebooted the server.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

Peter Moylan submitted this idea :
> On 30/03/24 06:42, Mike Spencer wrote:
>
>> PS: Attempts to connect towww.pmoylan.org time out.
>
> A log file suggests that something went wrong about six hours ago. I've
> now rebooted the server.

And within the last 5 minutes I accessed it, although response was
slow; response is quicker now.

/dps

--
Who, me? And what lacuna?

On 30/03/24 09:59, Snidely wrote:
> Peter Moylan submitted this idea :
>> On 30/03/24 06:42, Mike Spencer wrote:
>>
>>> PS: Attempts to connect towww.pmoylan.org time out.
>>
>> A log file suggests that something went wrong about six hours ago.
>> I've now rebooted the server.
>
> And within the last 5 minutes I accessed it, although response was
> slow; response is quicker now.

Thanks for the confirmation.

I'm starting to think that I need to replace that machine. It's been
beeping at me now and then, and that can mean a sick hard disk.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

Peter Moylan <peter@pmoylan.org.invalid> wrote:

> On 30/03/24 09:59, Snidely wrote:
> > Peter Moylan submitted this idea :
> >> On 30/03/24 06:42, Mike Spencer wrote:
> >>
> >>> PS: Attempts to connect towww.pmoylan.org time out.
> >>
> >> A log file suggests that something went wrong about six hours ago.
> >> I've now rebooted the server.
> >
> > And within the last 5 minutes I accessed it, although response was
> > slow; response is quicker now.
>
> Thanks for the confirmation.
>
> I'm starting to think that I need to replace that machine. It's been
> beeping at me now and then, and that can mean a sick hard disk.

Server not found error, from here,

Jan

On 30/03/24 20:54, J. J. Lodder wrote:
> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>
>> On 30/03/24 09:59, Snidely wrote:
>>> Peter Moylan submitted this idea :
>>>> On 30/03/24 06:42, Mike Spencer wrote:
>>>>
>>>>> PS: Attempts to connect towww.pmoylan.org time out.
>>>>
>>>> A log file suggests that something went wrong about six hours
>>>> ago. I've now rebooted the server.
>>>
>>> And within the last 5 minutes I accessed it, although response
>>> was slow; response is quicker now.
>>
>> Thanks for the confirmation.
>>
>> I'm starting to think that I need to replace that machine. It's
>> been beeping at me now and then, and that can mean a sick hard
>> disk.
>
> Server not found error, from here,

Some browsers will do that, because my web server is an http:// server,
not https://, and recently several web browsers have a new "feature":
they'll insist on trying https:// even when it's obviously not working.
There are ways to fix this, but it involves changing some options in the
browser, and most people won't even try to find the fix. It looks as if
I'm going to be increasingly isolated until I finally finish my TLS
implementation.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

Peter Moylan <peter@pmoylan.org.invalid> wrote:

> On 30/03/24 20:54, J. J. Lodder wrote:
> > Peter Moylan <peter@pmoylan.org.invalid> wrote:
> >
> >> On 30/03/24 09:59, Snidely wrote:
> >>> Peter Moylan submitted this idea :
> >>>> On 30/03/24 06:42, Mike Spencer wrote:
> >>>>
> >>>>> PS: Attempts to connect towww.pmoylan.org time out.
> >>>>
> >>>> A log file suggests that something went wrong about six hours
> >>>> ago. I've now rebooted the server.
> >>>
> >>> And within the last 5 minutes I accessed it, although response
> >>> was slow; response is quicker now.
> >>
> >> Thanks for the confirmation.
> >>
> >> I'm starting to think that I need to replace that machine. It's
> >> been beeping at me now and then, and that can mean a sick hard
> >> disk.
> >
> > Server not found error, from here,
>
> Some browsers will do that, because my web server is an http:// server,
> not https://, and recently several web browsers have a new "feature":
> they'll insist on trying https:// even when it's obviously not working.
> There are ways to fix this, but it involves changing some options in the
> browser, and most people won't even try to find the fix. It looks as if
> I'm going to be increasingly isolated until I finally finish my TLS
> implementation.

Be more British, it is the world that is isolated,

Jan

Peter Moylan explained :
> On 30/03/24 20:54, J. J. Lodder wrote:
>> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>>
>>> On 30/03/24 09:59, Snidely wrote:
>>>> Peter Moylan submitted this idea :
>>>>> On 30/03/24 06:42, Mike Spencer wrote:
>>>>>
>>>>>> PS: Attempts to connect towww.pmoylan.org time out.
>>>>>
>>>>> A log file suggests that something went wrong about six hours
>>>>> ago. I've now rebooted the server.
>>>>
>>>> And within the last 5 minutes I accessed it, although response
>>>> was slow; response is quicker now.
>>>
>>> Thanks for the confirmation.
>>>
>>> I'm starting to think that I need to replace that machine. It's
>>> been beeping at me now and then, and that can mean a sick hard
>>> disk.
>>
>> Server not found error, from here,
>
> Some browsers will do that, because my web server is an http:// server,
> not https://, and recently several web browsers have a new "feature":
> they'll insist on trying https:// even when it's obviously not working.
> There are ways to fix this, but it involves changing some options in the
> browser, and most people won't even try to find the fix. It looks as if
> I'm going to be increasingly isolated until I finally finish my TLS
> implementation.

You're not writing your own TLS, are you? If you are, that way lies
madness. Consider using one of the sources that the curl utility will
build with.

/dps

--
There's nothing inherently wrong with Big Data. What matters, as it
does for Arnold Lund in California or Richard Rothman in Baltimore, are
the questions -- old and new, good and bad -- this newest tool lets us
ask. (R. Lerhman, CSMonitor.com)

On 31/03/24 06:03, Snidely wrote:
> Peter Moylan explained :

>> Some browsers will do that, because my web server is an http://
>> server, not https://, and recently several web browsers have a new
>> "feature": they'll insist on trying https:// even when it's
>> obviously not working. There are ways to fix this, but it involves
>> changing some options in the browser, and most people won't even
>> try to find the fix. It looks as if I'm going to be increasingly
>> isolated until I finally finish my TLS implementation.
>
> You're not writing your own TLS, are you? If you are, that way lies
> madness. Consider using one of the sources that the curl utility
> will build with.

I am. Call it a personal quirk, but I like to have control of my own
code. Also, the sources I've looked at are too hard to understand. No
documentation, few comments. Some people who work on encryption seem to
believe that their code should be cryptic. Their understanding of
information hiding is different from mine.

I've already written my own SSH, and that works fine. TLS is a bit
harder, because it's not as well designed as SSH, and is heavily
dependent on certificates. I set the TLS work aside when I realised that
I needed to find a source of free certificates, and somehow ended up
detoured onto other projects.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

Saturday, Peter Moylan quipped:
> On 31/03/24 06:03, Snidely wrote:
>> Peter Moylan explained :
>
>>> Some browsers will do that, because my web server is an http://
>>> server, not https://, and recently several web browsers have a new
>>> "feature": they'll insist on trying https:// even when it's
>>> obviously not working. There are ways to fix this, but it involves
>>> changing some options in the browser, and most people won't even
>>> try to find the fix. It looks as if I'm going to be increasingly
>>> isolated until I finally finish my TLS implementation.
>>
>> You're not writing your own TLS, are you? If you are, that way lies
>> madness. Consider using one of the sources that the curl utility
>> will build with.
>
> I am. Call it a personal quirk, but I like to have control of my own
> code. Also, the sources I've looked at are too hard to understand. No
> documentation, few comments. Some people who work on encryption seem to
> believe that their code should be cryptic. Their understanding of
> information hiding is different from mine.

I'm not sure that they believe it should be cryptic as that it is
battle-tested.

> I've already written my own SSH, and that works fine. TLS is a bit
> harder, because it's not as well designed as SSH, and is heavily
> dependent on certificates. I set the TLS work aside when I realised that
> I needed to find a source of free certificates, and somehow ended up
> detoured onto other projects.

Are you familiar with the Feisty Duck Newsletter?

<URL:https:https://www.feistyduck.com/newsletter/>

They've also written a book:
<https://www.feistyduck.com/books/bulletproof-tls-and-pki/>

and a training program --
<https://www.feistyduck.com/training/practical-tls-and-pki>

/dps

--
https://xkcd.com/2704

On 31/03/24 11:57, Snidely wrote:
>
> Are you familiar with the Feisty Duck Newsletter?
>
> <URL:https:https://www.feistyduck.com/newsletter/>
>
> They've also written a book:
> <https://www.feistyduck.com/books/bulletproof-tls-and-pki/>
>
> and a training program --
> <https://www.feistyduck.com/training/practical-tls-and-pki>

Thanks, I've taken a look at these.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

Peter Moylan explained on 3/30/2024 :
> On 31/03/24 11:57, Snidely wrote:
>>
>> Are you familiar with the Feisty Duck Newsletter?
>>
>> <URL:https:https://www.feistyduck.com/newsletter/>
>>
>> They've also written a book:
>> <https://www.feistyduck.com/books/bulletproof-tls-and-pki/>
>>
>> and a training program --
>> <https://www.feistyduck.com/training/practical-tls-and-pki>
>
> Thanks, I've taken a look at these.

I also wonder if you're ready to deal with side channel and timing
attack issues? It isn't just buffer management you worry about.

The rule of thumb for crypto is "Never Roll Your Own". Even cribbing
from the Demming book isn't enough; I'm pretty sure that even the open
source stuff like OpenSSL rely heavily on paid professionals with years
of experience in crypto and related issues, either from companies or
universities.

I believe Daniel Sternberg, papa of curl, works for WolfSSL these days,
but doesn't roll his own.

/dps

--
Trust, but verify.

On 30/03/2024 20:03, Snidely wrote:
> You're not writing your own TLS, are you?  If you are, that way lies
> madness.

It is also the definition of masochism, as defined by sadists.

On Fri, 29 Mar 2024 22:50:12 -0000, Peter Moylan <peter@pmoylan.org.invalid> wrote:

> Peter Moylan http://www.pmoylan.org
> Newcastle, NSW

That's a big light fixture, looks like you could bump your head on it.

On Sat, 30 Mar 2024 10:22:08 -0000, Peter Moylan <peter@pmoylan.org.invalid> wrote:

> On 30/03/24 20:54, J. J. Lodder wrote:
>> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>>
>>> On 30/03/24 09:59, Snidely wrote:
>>>> Peter Moylan submitted this idea :
>>>>> On 30/03/24 06:42, Mike Spencer wrote:
>>>>>
>>>>>> PS: Attempts to connect towww.pmoylan.org time out.
>>>>>
>>>>> A log file suggests that something went wrong about six hours
>>>>> ago. I've now rebooted the server.
>>>>
>>>> And within the last 5 minutes I accessed it, although response
>>>> was slow; response is quicker now.
>>>
>>> Thanks for the confirmation.
>>>
>>> I'm starting to think that I need to replace that machine. It's
>>> been beeping at me now and then, and that can mean a sick hard
>>> disk.
>>
>> Server not found error, from here,
>
> Some browsers will do that, because my web server is an http:// server,
> not https://, and recently several web browsers have a new "feature":
> they'll insist on trying https:// even when it's obviously not working.
> There are ways to fix this, but it involves changing some options in the
> browser, and most people won't even try to find the fix. It looks as if
> I'm going to be increasingly isolated until I finally finish my TLS
> implementation.

https is an unnecessary annoyance. Welcome to the world of expiring certificates. I run science projects and there's always one with an expired certificate, nobody remembers to renew them.

On Sat, 30 Mar 2024 13:36:14 -0000, J. J. Lodder <nospam@de-ster.demon.nl> wrote:

> Peter Moylan <peter@pmoylan.org.invalid> wrote:
>
>> On 30/03/24 20:54, J. J. Lodder wrote:
>> > Peter Moylan <peter@pmoylan.org.invalid> wrote:
>> >
>> >> On 30/03/24 09:59, Snidely wrote:
>> >>> Peter Moylan submitted this idea :
>> >>>> On 30/03/24 06:42, Mike Spencer wrote:
>> >>>>
>> >>>>> PS: Attempts to connect towww.pmoylan.org time out.
>> >>>>
>> >>>> A log file suggests that something went wrong about six hours
>> >>>> ago. I've now rebooted the server.
>> >>>
>> >>> And within the last 5 minutes I accessed it, although response
>> >>> was slow; response is quicker now.
>> >>
>> >> Thanks for the confirmation.
>> >>
>> >> I'm starting to think that I need to replace that machine. It's
>> >> been beeping at me now and then, and that can mean a sick hard
>> >> disk.
>> >
>> > Server not found error, from here,
>>
>> Some browsers will do that, because my web server is an http:// server,
>> not https://, and recently several web browsers have a new "feature":
>> they'll insist on trying https:// even when it's obviously not working.
>> There are ways to fix this, but it involves changing some options in the
>> browser, and most people won't even try to find the fix. It looks as if
>> I'm going to be increasingly isolated until I finally finish my TLS
>> implementation.
>
> Be more British, it is the world that is isolated,

But he's an Aussie.

On Sun, 31 Mar 2024 08:43:35 +0100, occam <occam@nowhere.nix> wrote:

> On 30/03/2024 20:03, Snidely wrote:
>> You're not writing your own TLS, are you? If you are, that way lies
>> madness.
>
> It is also the definition of masochism, as defined by sadists.

Wouldn't masochists and sadists make the same definitions?

On Fri, 29 Mar 2024 22:50:12 -0000, Peter Moylan <peter@pmoylan.org.invalid> wrote:

> Peter Moylan http://www.pmoylan.org
> Newcastle, NSW

That's a big light fixture, looks like you could bump your head on it.

On Fri, 29 Mar 2024 22:50:12 -0000, Peter Moylan <peter@pmoylan.org.invalid> wrote:

> Peter Moylan http://www.pmoylan.org
> Newcastle, NSW

That's a big light fixture, looks like you could bump your head on it.

On 31/03/2024 11:56, Vladimir Putin wrote:
> On Fri, 29 Mar 2024 22:50:12 -0000, Peter Moylan
> <peter@pmoylan.org.invalid> wrote:
>
>> Peter Moylan                         http://www.pmoylan.org

>
> That's a big light fixture, looks like you could bump your head on it.

You know Vlad, if you don't have anything useful to say, you are not
obliged to post anything. Especially the same inane message - three
times over.

I'm guessing you are technologically less than savvy. Anyone who does
not know the security differences between https and http should not be
allowed to use a browser. 'Opera Mail/1.0 (Win32)", hmm. I pity the
people you "run science projects for".

Sun, 31 Mar 2024 10:54:28 +0100: "Vladimir Putin" <russia@will.rule>
scribeva:
>https is an unnecessary annoyance.

Not if logging in and entering passwords is required.

>Welcome to the world of expiring certificates. I run science projects and there's always one with an expired certificate, nobody remembers to renew them.

With Let's Encrypt, if set up right (which can be daunting), that is
automatic.

--
Ruud Harmsen, https://rudhar.com

On Sun, 31 Mar 2024 16:36:20 +0200
Ruud Harmsen <rh@rudhar.com> wrote:

> Sun, 31 Mar 2024 10:54:28 +0100: "Vladimir Putin" <russia@will.rule>
> scribeva:
>

I see no reason to encourage this recividist.

--
Bah, and indeed Humbug.

On 31/03/24 20:53, Vladimir Putin wrote:
> On Fri, 29 Mar 2024 22:50:12 -0000, Peter Moylan
> <peter@pmoylan.org.invalid> wrote:
>
>> Peter Moylan http://www.pmoylan.org
>> Newcastle, NSW
>
> That's a big light fixture, looks like you could bump your head on
> it.

Now I'm trying to remember where that photo is from. The door looks a
bit like a hotel door. We don't have anything like that in our present
home, although I do have to be careful about raising my arms when a
ceiling fan is running.

The other photo was taken on a North Coast vacation, and it's old enough
to lie about the colour of my hair and beard.

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW

On Sun, 31 Mar 2024 13:28:50 +0100, occam <occam@nowhere.nix> wrote:

> On 31/03/2024 11:56, Vladimir Putin wrote:
>> On Fri, 29 Mar 2024 22:50:12 -0000, Peter Moylan
>> <peter@pmoylan.org.invalid> wrote:
>>
>>> Peter Moylan http://www.pmoylan.org
>
>>
>> That's a big light fixture, looks like you could bump your head on it.
>
> You know Vlad, if you don't have anything useful to say, you are not
> obliged to post anything.

What's wrong with commenting on an unusual light fixture?

> Especially the same inane message - three times over.

Blame my news provider.

> I'm guessing you are technologically less than savvy. Anyone who does
> not know the security differences between https and http should not be
> allowed to use a browser.

It worked before https. Now we have red tape everywhere.

> 'Opera Mail/1.0 (Win32)", hmm. I pity the
> people you "run science projects for".

The science projects are mostly 64 bit, that's up to them. I use an old email client because I can't find a newer one with all the features.

On Sun, 31 Mar 2024 15:36:20 +0100, Ruud Harmsen <rh@rudhar.com> wrote:

> Sun, 31 Mar 2024 10:54:28 +0100: "Vladimir Putin" <russia@will.rule>
> scribeva:
>> https is an unnecessary annoyance.
>
> Not if logging in and entering passwords is required.

But we always managed it before.

>> Welcome to the world of expiring certificates. I run science projects and there's always one with an expired certificate, nobody remembers to renew them.
>
> With Let's Encrypt, if set up right (which can be daunting), that is automatic.

Daunting. Not something someone more concerned with the science wants to deal with. And no they can't afford to hire techys.

On Sun, 31 Mar 2024 16:06:56 +0100, Kerr-Mudd, John <admin@127.0.0.1> wrote:

> On Sun, 31 Mar 2024 16:36:20 +0200
> Ruud Harmsen <rh@rudhar.com> wrote:
>
>> Sun, 31 Mar 2024 10:54:28 +0100: "Vladimir Putin" <russia@will.rule>
>> scribeva:
>
> I see no reason to encourage this recividist.

Then why haven't you killfiled me?

That's one of the things Opera Mail (which someone else said was no good) does - if I killfile you, I see no replies to you, why would I? Apparently some lesser newsreaders let you see replies, which quote the original you didn't want to see! Doh!