Re: csapp, really good?

2019-07-09 Thread Andrew Luke Nesbit

On 09/07/2019 09:09, Dan LaBell wrote:

UNIX for Programmers, and Users, a Complete Guide.
(NOT THE 2nd edition, the 1st ) Graham Glass


Why do you like the 1st edition more than the 2nd?

I understand many reasons why an earlier edition is preferable.  I have 
several examples of titles too, where I prefer an earlier edition.  For 
example, when a beautiful chapter on an obscure, underrated, or 
overlooked topic is removed to make room for information on new 
developments.  (And I appreciate this.  Publishing is full of tradeoffs 
like this.)


Note that there seems to be a 3rd edition of Glass available.


And, Practical C Programming, Steven Oualline
(which I will part with in moment, and never really needed, but I will
still recommend it) because it contains every scold you would know by 
heart,

if you learned programming, in the unix lab.


I see this book often and have skimmed through it once or twice.  I 
never saw anything particularly compelling about it.  I will have a 
closer look next time.


Understanding the dark corners of C is essential to understanding the 
language properly.  More importantly, it's important to know how to 
protect oneself against widely propagated misinfomation.  An example of 
this kind of _misinformation_ is that arrays and pointers are the same.


There aren't many of these dark corners but most of them have a profound 
influence.  I highly recommend getting a copy of "Expert C Programming: 
Deep C Secrets" by Peter van der Linden and reading it with K's 2nd 
edition close by, which it makes meaningful references to.


As for the original book you asked about, CS:APP3e, I think it looks 
fantastic.  It's been on my list of books to read for about a year. 
From what I've seen it's completely different to the Unix book by Bach.


Andrew
--
OpenPGP key: EB28 0338 28B7 19DA DAB0  B193 D21D 996E 883B E5B9


Re: Web + email hosting recommendations

2019-05-09 Thread Andrew Luke Nesbit
On 09/05/2019 08:44, Mayuresh wrote:
> On Tue, May 07, 2019 at 09:20:57PM +, Christopher Pinon wrote:
>> Someone has already mentioned Vultr (who offer KVM VPSes), which is a
>> good choice. I've run a small NetBSD VPS at Vultr in the past.

The idea of cheap VPS'es that offer KVM access is new to me.  This is
welcome news.

> Has anyone tried hetzner[1]?
> 
> It seems KVM, so possibly NetBSD should work.

You need a dedicated server for this.

> The plans look good 2GB/20GB/20TB/1vCPU for Euro 2.49.

I had one of these and inquired about KVM access.  It is not currently
provided.

Andrew
-- 
OpenPGP key: EB28 0338 28B7 19DA DAB0  B193 D21D 996E 883B E5B9


Re: Web + email hosting recommendations

2019-05-08 Thread Andrew Luke Nesbit
On 08/05/2019 18:41, Christopher Pinon wrote:
> Thanks for your extended comment, but I'm not sure that I would conclude
> from it that "Debian is in pretty bad shape these days". (Or, if so,
> then that every distribution is in pretty bad shape these days.)

You've identified a contradiction in my spiel.  I'll think about it and
tighten up my argument.

> Don't take this the wrong way, but: Adélie Linux? Really? Still at beta
> with (I'm guessing) five users?

It's my personal opinion.  I think it's an excellent distribution and I
am seeing it grow in popularity.

> As for Void Linux, it's a rolling release, intended for those who like
> to break and to recompile their system every weekend.

I've never been a fan of rolling releasees and they break for me all the
time too.  But some people seem to know how to make them work reliably.

> Debian stable isn't for everyone, and certainly not for those who want
> to experiment with Btrfs.

In this case, why would they include Btrfs in the default package
repositories?

> Sorry to hear about your networking issue with Debian stable, but you
> didn't say what exactly the problem was.

There's no need to apologise for my networking issue.  I don't even
blame Debian.  Software is very, very difficult to get right.  Sometimes
I'm surprised that computers work at all.

I didn't go into the details of the problem because they are irrelevant,
boring, and it would take a very long time to accurately describe the issue.

> In any case, we're already OT,
> given that this is a NetBSD list. :-)
Is this whole thread OT because it asks for web hosting recommendations
instead of asking questions specifically about NetBSD?

And I believe that my message was relevant to NetBSD in the sense that
it's about the complexity of software generally.  Some of you may think
I'm drawing a long bow.  I intend this to be my last message on the
topic anyway (although I'm happy to continue to discuss off-list).

Andrew
-- 
OpenPGP key: EB28 0338 28B7 19DA DAB0  B193 D21D 996E 883B E5B9


Re: Web + email hosting recommendations

2019-05-08 Thread Andrew Luke Nesbit
On 08/05/2019 09:09, Mayuresh wrote:
> BTW what is the downside of having IPV6 only? Isn't it the future anyway?

I am trying to go IPv6-native.  It's been very difficult so far.

The worst thing is when an ISP doesn't support IPv6.  How does one work
around this?

Andrew
-- 
OpenPGP key: EB28 0338 28B7 19DA DAB0  B193 D21D 996E 883B E5B9


Re: Web + email hosting recommendations

2019-05-08 Thread Andrew Luke Nesbit
On 08/05/2019 10:19, Christopher Pinon wrote:
> Sijmen J. Mulder  wrote:
> 
>> U'll Be King Of The Stars  wrote:
>>> Debian is in pretty bad shape these days.
>>
>> At risk of getting OT here, but in what way? Not trying to debate, just
>> curiousity.
> 
> I can't say what ullbeking was thinking of, but some were disappointed
> by Debian's decision (for Debian 8) to adopt systemd.

I don't like systemd but I accept it and I embrace it.  And I think it
will get better.  I am trying to learn how to operate it.  But I feel
that a lot of the anti-systemd sentiment comes from people not liking
change.

Systemd is here to stay; the world of computing is one of the
fastest-changing things in the, er, world.  If you want to stay relevant
(if you are a professional) then you have to make at least some effort
to accept changes you don't like or don't agree with.

There are MANY other modern technologies that I dislike more than
systemd.  Systemd is the least of my concerns.

Yes, systemd HAS complicated the operating system and I don't really
understand how it's supposed to be used.  With the benefit of hindsight,
had Debian or Red Hat gone on a campaign to explain what systemd is; how
it is different; where it is better than the old init system; and, most
importantly, _how to use it_; then I think we wouldn't have such
negative sentiment towards to it.

Unfortunately I think many grass roots developers see Red Hat as an
untrustworthy corporate enterprise these days.  I think they are angry
that Red Hat forced systemd onto them, thus signifying the start of an
era where Linux development is no longer a community effort.

So it is true I prefer to use systems that don't use systemd.  But
that's not the core of the issue here.

I boycotted Debian for a few months some time ago out of frustration
because a bug had turned one of my machines into a rogue host and caused
my entire network configuration to collapse.  When I was venting my
frustrations, people were automatically assuming that systemd was to
blame (it wasn't) and I hated systemd.  I didn't and I don't hate
systemd, although I prefer OS'es that don't have it.

I do agree that systemd exacerbated this bug, which was already present.
 And I believe that the complexity and unusability of systemd made it
difficult to understand what the problem was and how to contain it,
before it was too late.

Nevertheless, I think Debian made the right move in adopting systemd.
It's important that we have a distribution with a decent ethical
framework that can stay competetive against Red Hat (in some abstract
way that I am finding it difficult to articulate).

Having said THAT, I actually enjoy using CentOS and RHEL.  I also think
they have decent documentation even though it's not as good as BSD's.

In summary, systemd is correlated with increasing complexity and
increasing numbers of quality issues.  I am not anti-systemd because I
don't think it's the _cause_ of the kinds of issues I have been having
during the past couple of years.

> At the same time, the truth is that for any Debian release in its
> history, you can find people saying that Debian is in pretty bad shape,
> so unless more is said, it's hard to know what is meant.

I have a lot of respect for the Debian project -- its ethics and its
technical quality.  All Linux distributions have major flaws.  This is
software we're discussing, after all.  I have been using Debian since
the late 1990's, and I know that if I were to trade it in for something
else I'll just get a whole new set of issues to cope with.  (Disclaimer:
I am currently spending more and more time with Adélie Linux and intend
to move more of my end user Linux machines to Adélie.)

The main problem with Debian is one that I sympathise with greatly.  For
many years I have felt that the size and complexity of the project is
out of control.  It's an under-resourced project, despite the best
efforts of a lot of talented developers who often have to make difficult
(and sometimes unpopular) decision.

THIS is causing the kinds of defects that hit me the other day.  It
caused problems on my network that took weeks to fix.  Problems like
this are happening more and more frequently as time goes by.  There's
just far too much software to nurse along, but the Debian developers
have been doing a great job given the difficult circumstances.

Another issue, which I think is quite serious, is the outdated Stable
kernel that is continually recieving back-patches.  I know there are
real reasons for this, and I don't have a better solution.  Regardless,
this causes extremely subtle and mysterious problems that are almost
impossible to predict or detect, excpect for the fact that "my computer
is acting weirdly when I use this program that has very strict
requirements of correctness" (such as flashrom).

These old kernels also make it impossible to run things like Btrfs
reliably.  Btrfs generally requires as recent a kernel as one is able to
deploy, at least 4.14.  

Re: amd64 SBCs on which NetBSD would run ?

2019-05-07 Thread Andrew Luke Nesbit
On 07/05/2019 13:23, David Brownlee wrote:
> On Sat, 4 May 2019 at 18:16, Mike Pumford  wrote:
>>
>> On 04/05/2019 15:30, Mayuresh wrote:
>>> On Sat, May 04, 2019 at 05:49:58PM +0800, Travis Paul wrote:
 You mentioned that you were looking for an amd64 board.  Have you looked
 at the PCEngines APU2 boards[1]? I have not personally tried them but
 perhaps they fit your needs.

I hope you don't mind me jumping in on the conversation at this late
stage.  I've always had a bit of an allergic reaction to "high end"
SBC's like the APU2.  I feel they are overpriced, sometimes very much so.

For the same money as the APU2 you can get a real mainboard, one with a
much more solid construction and better performance.  Similarly with
many other SBC's in that price level.  If you look hard enough you can
get an entry-level serverboard with IPMI for not much more money.

I do like SBC's very much.  My favorite SBC is the OPi+2E, and I hope to
get NetBSD running on it in the near future.  But the advantage of SBC's
(for me) is when they are very cheap and/or promote creative things.

Andrew
-- 
OpenPGP key: EB28 0338 28B7 19DA DAB0  B193 D21D 996E 883B E5B9


Re: Web + email hosting recommendations

2019-05-07 Thread Andrew Luke Nesbit
On 07/05/2019 13:39, Mayuresh wrote:
> On Tue, May 07, 2019 at 01:37:51PM +0100, Sad Clouds wrote:
>> Hello, do you know what hypervisor they use for your NetBSD VPS?
> 
> Sorry, no idea.

Also, don't forget SDF.  They offer NetBSD VPS's.

Andrew
-- 
OpenPGP key: EB28 0338 28B7 19DA DAB0  B193 D21D 996E 883B E5B9


Re: Web + email hosting recommendations

2019-05-07 Thread Andrew Luke Nesbit
On 07/05/2019 12:46, Sad Clouds wrote:
> Hello, could anyone recommend web hosting providers
Perhaps you will find something suitable on this list?

https://trac.torproject.org/projects/tor/wiki/doc/GoodBadISPs

Andrew
-- 
OpenPGP key: EB28 0338 28B7 19DA DAB0  B193 D21D 996E 883B E5B9