Need To Understand Booting Process Of CentOS-7

Home » CentOS » Need To Understand Booting Process Of CentOS-7
CentOS 27 Comments


could some help me understand the booting process of CentOS-7

I need to have constructive material where i could able present a presentation on booting process of CentOS-7

With Thanks & Regards,

Keshaba Mahapatra Sr.Technical Consultant

27 thoughts on - Need To Understand Booting Process Of CentOS-7

  • How about I do your presentation for you and you pay me your salary!

    I mean, seriously – according to your signature you are a Senior Technical Consultant. Surely you can read the technical notes / product manual provided by upstream, or if that does not contain enough in depth material then man pages and google should be able to fill in the blanks for you. At the very least I would expect a Senior Technical Consultant to be able to research such things by himself and not post nonsense questions to a public mailing list.

    Jeez, and it’s only Tuesday :-D

  • That was a little rude, don’t you think?

    I agree with your general premises; Keshaba could have started with google and then, perhaps, asked here if s/he got stuck on a particular part. That said though, there was no need for you to come down so hard, either.

  • A little harsh maybe, but not rude. I didn’t swear and I didn’t personally insult. I simply called it as I saw it. But I take your point.

    It was the “Senior” part of the job title that set me off. IMHO the level of questioning didn’t in any way match up with the level of professionalism inferred by the job title. To me it read more like a grad student asking the list to do their latest assignment for them.

  • Especially since 99% of what they need to know is still valid from the TLDP’s Boot to Bash book. The only difference is that instead of SysV init taking over after the initrd, SystemD takes over, which, while sufficiently different from SysV init, is _well_ documented.

  • Ned Slider wrote:

    First mistake: what makes you think we’re your “team”? This is an open mailing list that we’ve all subscribed to, and we’re all over the world, and in various positions and skillsets. We are not your “team”.

    Right – how *did* he get that title (or did he just make it up)?

    Long past history: when I started programming, I was hired as a sr. programmer I. A few months later, I asked my buddy, the systems programmer, why sr. and not jr. He told me that they’d gotten rid of all the jr positions a couple years before, because they were eligible to join the union….

    mark “consultant to his Exalted Highness…”*

    * My cat, of course? I mean, why else are we, and the Internet, here?

  • and hated by some (meaning myself, but I don’t feel I’m alone…


    Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247

  • I agree with Mark here (imaginey that); and to add to that I consider this it poor taste, if not outright insulting, to just assume we are on the same team. We’re not. I for one will not be part of a “team” with people that expect others to do their job for them.

    ^ that.

    I am not trying to come down on the OP here, this is part of a much larger problem. The problem is one of entitlement and laziness. You can couch this in whatever politically correct (ask me about my thoughts on that some time) phrasing you want, but it all boils down to this.

    People have forgotten how to do things on their own and expect everyone else to just drop what they’re doing so they can work on the problems of others. Well, life’s not like that. People damned well need to start doing at least a modicum of research on their own.

    This problem is becoming more and more prevalent as time goes on; and quite frankly the level of competence in IT and surrounding fields has been going steadily down the toilet for quite some time because of it.

    And people that defend, to whatever degree, this type of behavior are part of the overall problem. Three is a huge difference between helping others and coddling them; start helping and stop coddling. If _I_ can manage to do my job without having to depend on others I’m pretty sure it’s a fair bet others can as well.


  • I frequently see “team” as a greeting when, obviously, the poster knows we’re not a team. It always struck me as a cultural quirk rather than some assumption that the questioner expects us to behave like we’re part of the same actual team.

    It sounds to me like this is along the same lines as today’s elders complaining about how kids these days have no respect. In my little tech community, I see a majority of younger people eager to put in the work needed to learn new technologies. They push at least as hard as I did when I was their age, too. Are there some slackers and freeloaders?
    Sure. Are they the majority though? No.

    I think it would be really good if everyone took a step back, took a deep breath, and relaxed. This thread has generated a surprisingly (and to me baffling) amount of negativity.

  • I’m happy to sign under that, only I feel myself plagiarist here… But indeed I hate these days we start solving any problem by creating [google or any other search engine] search line instead of reading documentation first. I hate the GUI interfaces (latest KDE, GNOME,… and lates Windoze as well ;-) switched from logically organized tree of categories, subcategories,… items. … Worst of all I hate myself following it all and behaving as above ;-(


    Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247

  • It’s very simple actually. The first step in the booting process of CentOS-7 is the application of emf to the system, following which it would initiate the cmos process to load the GPT from disk and hand over to LILO which will bootstrap the kernel from the file allocation table into the high memory area before running systemd to complete the booting process.

  • Digimer wrote on Tue, 30 Sep 2014 17:18:17 -0400:

    I don’t know how long you have been on this list, but it’s a really friendly place. Has been, still is. To get a response like the one in this thread you have to “earn” it. In this case it’s obvious that the OP is of that “please do my homework for me” type and didn’t do any research of his own other than finding out there is something different. This is not friendly to the list and it’s not good enough for a “senior” consultant. Ned’s response was appropriate and actually quite friendly.


  • You don’t write like a Komputer Xpert despite your impressive “senior”
    title :-)


    Paul. England, EU.

  • I have been following this – and other discussions here and elsewhere – and I
    have some mixed feelings about what is happening.

    I started with Linux at about Red Hat 5.2 in 1997 or so. I was looking to move away from OS/2 as I could see it was dying at that point but I had no desire to learn anything about (my opinion) that piece of crap many call WinBloze. There were no classes on Linux and almost no books. So I taught myself using what was available, usually with the aid of the few books that were extant then and a couple local people who were more advanced than I. This was before I realized that, in Gnu/Linux and Unix, ANYTHING can be done and all I have to do is find the right command(s) to make it happen. Not that I don’t need a bit of a nudge now and then.

    But that assistance was almost always in the form of me asking a question about how to do something and a response that I should check out a particular resource or simply the name of a command to look up. No one ever EVER gave me a complete command or script that would solve my problem. Nor did I want them to.

    I always learn best by doing things, exploring and experimenting. Most people I
    have taught in my own classes as well as Red Hat training and IBM classes I have taught over the years are the same way. At least those who students who actually
    “get it.”

    I believe that it is important to help those who are just beginning the long road towards being a “real” sysadmin, however you might define that. But I also believe that giving little nudges in the right direction and letting them figure out the rest is ALWAYS the best way to help them learn.

    I also think that too many people just starting with Linux are still thinking along the lines of WinBloze and don’t even know about the Unix/Linux Philosophy. Learning and keeping that philosophy in mind can go a long way towards helping solve any problem and understanding any part of Gnu/Linux.

    The attitude of many new to Linux and unfamiliar with the philosophy seems to be that they just want a specific map to follow, a set of steps or a formula with no variables to follow to solve every problem. Well that is NOT the Linux way. Nor can it ever be. This attitude is typical of WinBloze admins, IT and project managers as well as MBA’s who have no real understanding of the underlying technologies they “manage.” Admins coming out of other environments need to realize that computers and operating systems – especially “real” ones like Linux
    – are complex and there is no one way to do anything nor is there one single answer to most questions. Learning how things work is important.

    There is a great deal of really good information on the Internet that can help almost anyone looking to learn or to resolve a particular problem. There is also an unbelievable amount of garbage out there that is incomplete or just plain wrong and that can be very frustrating to anyone wishing to learn. Sorting through it and deciding which is which can be nearly impossible, especially for a newcomer.

    I have also been very frustrated by the poorly written and allegedly “free” crap some people peddle on places like LinkedIn and hope that very few actually use them. These documents can make learning the reality of Linux daunting and probably cause many to give up and stay with what they know.

    Perhaps some questions get asked here because people are so frustrated with the huge amounts of bad data that they want to come to the source first and not waste time experimenting their way through the garbage. I always keep a set of VMs available for experimentation, but not everyone has a VM or non-critical hardware available for experimentation.

    It can also be important to understand why a question is being asked. Is this person trying to write a book? Fix a problem? Just understand how it works?

    As for job titles, most are ridiculous anyway. As owner of my own business I
    have given myself a couple titles including “President,” “Lead Trainer” and
    “Senior Consultant.” The adjectives “senior” and “lead” are meaningless in reality because I am a one person shop. But they imply things to the people to whom I try to sell my services.

    I once worked for a large bank that had literally hundreds of “vice presidents.”
    The only qualification was to contribute to the bank’s political PAC.

    So, though I mostly lurk and learn, I think it important to point people in the right direction without doing their jobs for them.

    And I also enjoy a good discussion, however heated. ;-)

  • I have used Linux (RedHat mostly but others too) for at least a decade and a half too. I agree, there often is really little documentation. But I
    would say reading man pages (and re-reading them several times) was not only last resort but very often close to comprehensive information you need on a command. There were often nice “HWOTO”s, the one about xfs comes to mind, “linux on steroids” was its title.

    As far as *nix (|based) system documentation is concerned I would like to mention here _my_ comparison. On the best side is FreeBSD Handbook (found on their website). You can follow it to do almost anything. On the worst side (of what I’ve tried to use) is MacOS Server “documentation”. It is bloated in size to about FreeBSD Handbook if not larger. Tons of pictures. Consists almost to 100% of the following constructs: “Go there, click this, check that box, click next… You are done”. And if before “you are done” you get any error, you are dead and there is no way to troubleshoot what is wrong, or get clues from this “documentation”. I would put the documentation RedHat has for RHEL 7 right in the middle between FreeBSD
    Handbook and MacOS Server “documentation”.

    Just my little rant ;-(


    Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247

  • Sorry about that “Sr” thing. I hope no one is offended. I’m sysadmin working for 2 Departments of the university. At some point there was a
    “group” of (two) sysadmins that was doing this consisting of myself and
    “junior” sysadmin. During that period my position title sounded something like “IT manager” or some other manager. Which I never considered myself
    (a manager). I guess at that point I went a minimal resistance path and just added Sr in front of sysadmin (though that Sr was in position title long before).

    My apologies for being too lazy and not deleting my standard signature when posting to the list. Whoever read my comments probably noticed I’m often rather sceptical about myself, and often that indeed is justified
    ;-) As far as Experts are concerned, I usually consider the list as Experts when I ask something (and usually I address the list “Dear Experts” – which proved quite justified, BTW)


    Valeri Galtsev Sr System Administrator Department of Astronomy and Astrophysics Kavli Institute for Cosmological Physics University of Chicago Phone: 773-702-4247

  • The difficulty being that for much of the Linux infrastructure the official documentation is excessively exalted by being called ‘documentation’ in the first place. The reason I go to StartPage or DuckDuckGo first is that I have read a lot of the ‘documentation’ and whatever semantic content much of it contains is too frequently buried under jargon and dependent upon presumed foreknowledge. It does not help that many documentation writers seem to consider the the Laconians’ reply to Phillip as being too wordy. The next step no doubt is to quote: “Use the source, Luke. Use the source.” And let us not even consider the useful semantic content of 99.99% of all error messages generated by defective software; since there is none.

    I can hardly justify criticising someone who has limited time and infinite demands thereon turning to the experiences of others as documented on the web and helpfully indexed by Google et al first; before attempting to find the answer in one of the obscure and obfuscated texts that pass for technical documentation in most fields. Computers simply being the one I encounter most frequently and which happens to concern this issue.

    I use man pages and find most of them helpful. But, frankly, they are of more use as an aide memoire than as a source of instruction. I defy anyone who has never built an RPM package to build one from scratch using nothing more than their source code files and the relevant man pages. If they can get so far as to identify those to begin with.

    The existence of technical mailing lists, such as this one, is a mute testament to the deficiency of official documentation. No matter how good it may be a technical document will always be incomplete and subject to misunderstanding. One needs context to grasp what another is trying to convey. The experience of others provides that.

    As to the OP. Well, you have to start some place. Even if the request was awkwardly put and apparently presumptive one must consider that English may not be their native tongue; and that social mores and customs respecting forms of address vary considerably by country. I find it best to give people the benefit of the doubt. If I reply at all I try to provide a pointer to some resource relating to the query that either I have found helpful or has a good reputation for same.

  • You have nothing for which to apologize; anyone genuinely offended is overreacting. (I suspect that the responding poster was joking that a
    “senior” sysadmin wouldn’t be so knowledgeable about our field.)


  • Yup :D.

    See simple.

    Or at least, simple to you and me, and most sys admins who know Linux and computers well (To be fair, if you’re a decent Linux admin, you probably do).

    But to a newcomer, the above would just be a random blob of text which makes them give you a blank stare.

    Also, not necessarily over to Lilo, could go to grub or grub2 or any other bootloader… depends on the system configuration ;-)

    Kind Regards, Jake Shipton (JakeMS)
    GPG Key: 0xE3C31D8F
    GPG Fingerprint: 7515 CC63 19BD 06F9 400A DE8A 1D0B A5CF E3C3 1D8F

  • Timothy Murphy wrote:
    Geez! I bet he didn’t think that he’d have to add
    around that…. I’ve lost track of who posted that, but *I* thought it was funny….


  • Never ever heard anyone in the computer world refer either DC or AC
    (single or three phase) as EMF regardless of the voltage, amperage, resistance or impedance.

    When applying EMF does one spread it on like peanut butter, water it on or just press a EMF switch ?

    You make an invaluable point. Newcomers to any subject know little, if anything, about the topic. It is encouraging they are curious and wish to know more. If one doesn’t frighten them away who knows what significant contributions to our world they may make in years to come.

    I’m off to swap-over my EMF conversion unit which arrived by special delivery this evening. Some might call it a 100-240v AC power adaptor with 19v DC output.

  • I’d admit to a certain level of personal bias and snappish mood when I
    wrote that reply, due in part to the fact the OP’s name and style comes across as similar to somebody I had to deal with earlier, although I still lean, perhaps unfairly, that he is that same sort. Guys with very impressive (and frequently to the point of being absurd) resume/credentials who don’t actually know shit about the work they need to do, then try to leech off others (esp colleagues) and pass off as their own work for credit.

    Usually, I’m more than happy to share what I know, especially since much of what I learnt are from the efforts of others, but when it comes to this sort who don’t even bother to show the slightest effort, I like tripping them up :)

  • By your own admissions, you exemplify everything wrong with anonymous postings to public lists.

    I’d recommend you head over to Stack Exchange where safeguards have been designed to help you stay on topic or just not respond at all.

    Or just step away from the keyboard for the weekend, go outside and take a walk around. Stimulate the parts of your brain not associated with operating systems and on-line interactions. Perhaps even smile at someone you don’t know. Try to remember a time when you were younger… a time when you were hopefully less of and ass and more useful. Then come back and be that person. Take as long as you need. The Internet will still here when you get back.