The sad state of sysadmin in the age of containers • Hacker News Board

Re: The sad state of sysadmin in the age of containers

#371

Post by bm1362 » Wed, Apr 22, 2015, 6:48 PM UTC

tedivm wrote: post #311

ksml wrote: post #295

Earlier quoted context omitted.

I think more disconcerting is the rise of "sysadmins" who think they're qualified sysadmins because they know how to bash and docker.

This is hardly a new problem- and in many ways, I'm not sure it's a problem at all compared to the company cultural issues brought up by skywhopper. Whether it's programming or system administration, you're always going to have new people getting excited about the sudden power they've learned. Being able to make computers do things opens up this whole new world, and when people find themselves in that world they may…

I've found that there isn't an easy ramp into system admin from university -- most of the talent comes from dogmatic self learning in computer repair shops or subpar IT shops. All the good guys at $BIG_SOFTWARE_COMPANY seem to be in their 30s after putting in years doing /tedious/, but extremely useful, work for little pay.

Re: The sad state of sysadmin in the age of containers

#372

Post by octref » Wed, Apr 22, 2015, 6:49 PM UTC

I think it's because people are getting worse at explaining things and writing docs.

As a student many times I wanted to learn how things work, but most tutorials/docs just ask you to type in a few magical lines without much explanation. Maybe the authors think their audience won't understand anyway, but I think it's the authors' ineptitude if they can't explain what their programs do in an accessible way.

I really hope there could be more projects like i3[1] and flask[2].

1: http://i3wm.org/docs/userguide.html 2: http://flask.pocoo.org/docs/0.10/

Re: The sad state of sysadmin in the age of containers

#373

Post by yellowapple » Wed, Apr 22, 2015, 6:51 PM UTC

squar1sm wrote: post #314

matt_kantor wrote: post #266

Earlier quoted context omitted.

There are many more depressing examples of this at http://curlpipesh.tumblr.com

Funny tumblr but makes me care-confused. I understand that curl pipe sh could have security problems but I also don't see it as that much different than the "normal" and "ok" way of doing things. I would consider something like the below pretty normal. wget https://whatever.io/latest.tgz tar xzf latest.tgz cd whatever-stable ./configure && make sudo make install Because of familiarity, we aren't going to be too worri…

It's bad because `sh`, `bash`, etc. don't wait for the script that's being piped into it to finish downloading before it starts executing it. So, for example, if you're running a script with something like

    # remove the old version of our data
    sudo rm -rf /usr/local/share/some_data_folder

and the network connection cuts out for whatever reason in the middle of that statement (maybe you're on a bit of a spotty wireless network), the resulting partial command will still be run. If it were to cut off at `sudo rm -rf /usr`, then your system is in all likelihood going to be hosed.

Re: The sad state of sysadmin in the age of containers

#374

Post by gaius » Wed, Apr 22, 2015, 6:51 PM UTC

blfr wrote: post #4

This bothers me as well. Even tasks as simple as adding a repository are now being "improved" with a curl | sudo bash style setup[1]. However, installing from source with make was (and remains) a mess. It may work if you're dedicated to maintaining one application and (part of) its stack. But even then it usually leads to out of date software and tracking versions by hand. Many people have this weird aversion to doin…

Many people have this weird aversion to doing basic sysadmin stuff with Linux

Like developers who won't write SQL and insist on an ORM.

Re: The sad state of sysadmin in the age of containers

#375

Post by patsplat » Wed, Apr 22, 2015, 6:52 PM UTC

Regarding curl PACKAGE | sudo bash...

Just what do you think happens when you run `yum update` or Windows Update?

If you don't trust DNS or the network then you have serious challenges which frankly aren't even solved by air gapping file transfers.

Re: The sad state of sysadmin in the age of containers

#376

Post by swills » Wed, Apr 22, 2015, 6:52 PM UTC

getsat wrote: post #331

Alupis wrote: post #317

Earlier quoted context omitted.

> Is it really the best choice, or has someone done something cleaner in golang or c++11? What does the language have to do with the program? Hadoop is what it is because it's a complex problem with a fittingly complex solution. Simply re-writing it in your pet language won't somehow make it "better".

Go and modern C++ are both quite a bit more terse than Java. They also produce binaries which don't necessarily require a runtime to be available on every server (just ABI compatibility). (I have no horse in this race, I am just writing what I think the grandparent comment was referring to)

C++ does usually require a runtime.

Re: The sad state of sysadmin in the age of containers

#377

Post by yoklov » Wed, Apr 22, 2015, 7:08 PM UTC

swills wrote: post #376

getsat wrote: post #331

Earlier quoted context omitted.

Go and modern C++ are both quite a bit more terse than Java. They also produce binaries which don't necessarily require a runtime to be available on every server (just ABI compatibility). (I have no horse in this race, I am just writing what I think the grandparent comment was referring to)

C++ does usually require a runtime.

C++'s runtime is small and ubiquitous. Depending on how the software is written (if it allows disabling exceptions and rtti), it might be the same size as C's runtime, which is practically (but not totally) nonexistant.

I'm not an expert on Java, but my experience with it is that it's runtime is fairly huge and requires custom installation.

Re: The sad state of sysadmin in the age of containers

#378

Post by pjmlp » Wed, Apr 22, 2015, 7:11 PM UTC

getsat wrote: post #368

pjmlp wrote: post #347

Earlier quoted context omitted.

> They also produce binaries which don't necessarily require a runtime to be available on every server Just like Java[0]. It is just a matter of choosing the right compiler for the use case at hand. [0] - http://www.excelsiorjet.com/ (one from many vendors)

Cool concept, I didn't realise this existed. Can you run Hadoop and friends under this? I've worked at companies with over 500 servers in a Hadoop cluster and literally never once heard about anything other than using Oracle's JRE aside from one proposal to use OpenJDK which was shot down pretty quickly.

I don't have experience with Hadoop.

Almost all commercial JVMs have some form of AOT or JIT caching, specially those that target embedded systems.

Sun never added support to the reference JVM for political reasons, as they would rather push for plain JIT.

Oracle is now finally thinking about adding support for it, with no official statement if it will make it into 9 or later.

JEP 197 is the start of those changes, http://openjdk.java.net/jeps/197

Oracle Labs also has SubstrateVM, which is an AOT compiler built with Graal and Truffle.

Re: The sad state of sysadmin in the age of containers

#379

Post by mercurial » Wed, Apr 22, 2015, 7:14 PM UTC

lmm wrote: post #263

mercurial wrote: post #165

Earlier quoted context omitted.

I don't know any tooling that turns (recursively) a Maven pom file into a Debian repository of Debian policy-abiding .debs, which are magically updated when the pom file changes.

If you mean Debian policy-abiding in the sense of "signed by a Debian developer in the Debian WoT" then no, but that's not something you could ever do automatically. But for the rest, I've done all the individual pieces before: it is trivial to generate a .deb from a maven pom and put it in a debian repository, it's trivial to do some operation on all the dependencies of a maven project, and it's trivial to hook some…

If they are not public, that's obviously not the same thing, though you're still going through a lot more complexity than "here is this .war, put it on the server and reload the webapp".

Re: The sad state of sysadmin in the age of containers

#380

Post by Lord_Zero » Wed, Apr 22, 2015, 7:19 PM UTC

rudolf0 wrote: post #367

dikaiosune wrote: post #342

Earlier quoted context omitted.

The alternative is that many of the startups don't learn this in their own time, and they go on to become bigger, more successful companies who can set the tone and shift the market. Of course, if they're actually able to succeed by doing so, then that says something too. Although the trend of many data breaches certainly wouldn't decline in that case.

>Although the trend of many data breaches certainly wouldn't decline in that case. Exactly. Successful and profitable are not mutually exclusive with "secure" or "well-architected". At least until those last two come to bite you later and start eating into your profits.

Sony is a great example of this.