|
inignot posted:Also, go look at a crossever cable sometime, one end is A, one end is B.
|
# ? Jan 15, 2013 16:11 |
|
|
# ? May 16, 2024 18:09 |
|
Yay passed CompTIA A+ Test was a lot easier than I expected, although the things I didn't know I straight up didn't know.
|
# ? Jan 17, 2013 05:17 |
|
Shoeonhead22 posted:So there's less trivial bullshit on the ROUTE? I won't get asked to do things (as an extreme example) like gently caress with/calculate the reliability K-values? Depends what you consider trivial. Just took mine and I got questions like the metric of bgp when redistributed into ospf, dr/bdr election or neighbor in ospf network types or ipv6 reserved addresses. The exam wasn't that hard but some questions were kind of weird and as for the labs, most of mine didn't work. For example, in one lab I had to type a command that would bring down/up the neighbor relationship and it never did. So, my next course of action was to shut the interface and much to my surprise, the routers were still neighbors. I think I had 5 labs and 4 out of 5 didn't work. I passed with a good grade so I doubt 4/5 of my labs were wrong.
|
# ? Jan 17, 2013 18:13 |
|
Lunaris posted:Yay passed CompTIA A+ Any advice to someone who's planning on taking it? Specific stuff to memorize, out of left field questions, etc?
|
# ? Jan 17, 2013 20:31 |
|
Just passed the CCNP ROUTE this morning. Studied with "Implementing Cisco IP Routing (ROUTE) Foundation Learning Guide: Foundation learning for the ROUTE 642-902 Exam" reading it cover to cover, taking notes, etc. I feel like it did a pretty decent job getting me ready. Also I labbed the poo poo out of BGP and redistribution.
|
# ? Jan 17, 2013 21:45 |
|
^^ Great job, You are a two levels above me. How long did you spend working on getting CCNP ROUTE? I am not a huge note taker when reading books. I'll often go through a book two times to fully grasp the concept, that's just been my learning style. Though I know I will have to reread chapter for example if I get a low score on the end of chapter test. I've noticed a lot of people asking the A+ lately. I wonder it would be ok if I uploaded an exam study guide here from a book I used last year? Edit: Anyway I'm just gonna upload it's not like there's anything wrong with it. DropsySufferer fucked around with this message at 09:31 on Jan 18, 2013 |
# ? Jan 17, 2013 22:35 |
|
I've honestly been preparing for it since early to mid summer. Involving myself in discussions at work and bouncing ideas of colleagues and volunteering to take the lead on a couple of network design projects for mid-sized companies (with some air cover behind me just in case.) Around December or so I started hitting the books hard to fill in the gaps then in the beginning of January started spending time doing a lot of lab stuff. The practical hands-on stuff is tremendously helpful and practically trivializes most of the lab sims on the exam. When I study for something I tend to start way in advance and gradually up the intensity as I get closer to my deadline. Seems to help my retention a boatload. All in all the experience has made me a better network engineer.
|
# ? Jan 17, 2013 23:00 |
|
In the Lammle Network+ chapter on networking devices, he shows network diagrams that have some switches connected to another switch, with the hosts connected to the secondary switches. He doesn't explain this (at least not in this chapter), so what is the reasoning here? Why wouldn't you just connect the four hosts to one switch, instead?
|
# ? Jan 18, 2013 06:16 |
|
Kaethela posted:Any advice to someone who's planning on taking it? Specific stuff to memorize, out of left field questions, etc? Based on my own A+ exam and talking to a few other people that passed it, the questions that caught us all off guard were related to printers. Know the parts, what they do, and the troubleshooting stuff for them. In my exam there were definitely enough printer questions that I wished I had paid more attention to that section.
|
# ? Jan 18, 2013 09:05 |
|
hooah posted:In the Lammle Network+ chapter on networking devices, he shows network diagrams that have some switches connected to another switch, with the hosts connected to the secondary switches. He doesn't explain this (at least not in this chapter), so what is the reasoning here? Why wouldn't you just connect the four hosts to one switch, instead? Assume each of the hosts is simply a model and doesn't truly represent the numbers of hosts you would have on a real network.
|
# ? Jan 18, 2013 09:56 |
|
hooah posted:In the Lammle Network+ chapter on networking devices, he shows network diagrams that have some switches connected to another switch, with the hosts connected to the secondary switches. He doesn't explain this (at least not in this chapter), so what is the reasoning here? Why wouldn't you just connect the four hosts to one switch, instead? He's probably representing distribution and access layer switches.
|
# ? Jan 18, 2013 14:28 |
|
hooah posted:In the Lammle Network+ chapter on networking devices, he shows network diagrams that have some switches connected to another switch, with the hosts connected to the secondary switches. He doesn't explain this (at least not in this chapter), so what is the reasoning here? Why wouldn't you just connect the four hosts to one switch, instead? As others have said, that's a simple diagram to model access and distribution switches. In a reality, there could be a number of reasons for that arrangement. The most common scenario is physical separation of the host machines: different floors or hallways. There's also the issue of the physical limitations of copper. You generally want as short of a distance as possible from your hosts to the access layer switch, because copper can only carry a decent signal around 300 feet (and that's pushing it). From there, you'll have a fiber uplink to your distribution or core switches, which in large buildings and campuses can be several hundred to several thousand feet away.
|
# ? Jan 18, 2013 14:44 |
|
psydude posted:As others have said, that's a simple diagram to model access and distribution switches. In a reality, there could be a number of reasons for that arrangement. The most common scenario is physical separation of the host machines: different floors or hallways. There's also the issue of the physical limitations of copper. You generally want as short of a distance as possible from your hosts to the access layer switch, because copper can only carry a decent signal around 300 feet (and that's pushing it). From there, you'll have a fiber uplink to your distribution or core switches, which in large buildings and campuses can be several hundred to several thousand feet away. Cat5/6 specs says you can run cable to 100m (328ft) for GigE, 10GigE on Cat6 is limited to 55M at guaranteed ratings. After that you run into cross talk and other issues, you can run the cable considerably longer distances if you're not concerned about meeting the performance spec. That said, running copper that distance is never a good idea and fiber will almost always prove to be a better choice
|
# ? Jan 18, 2013 15:34 |
|
1000101 posted:Around December or so I started hitting the books hard to fill in the gaps then in the beginning of January started spending time doing a lot of lab stuff. The practical hands-on stuff is tremendously helpful and practically trivializes most of the lab sims on the exam. I'm working on CCNA Voice at the moment, but hope to start on the CCNP R&S exams right after that. How did you get access to the physical equipment you mention? Was it purely stuff at work, a rack rental, or your own personal lab? Or did you go the simulator/emulator route - GNS3 / Packet Tracer / Boson / some other? I'm wondering how far I'll get without buying gear, and I've heard conflicting reports, so I'm interested to hear what your experience has been.
|
# ? Jan 18, 2013 19:02 |
|
Is doing the upgrade from MCITP:EA (08R2) to the MCSE: Server Engineer (2012)? I just passed the 70-413 exam today and have 70-417 scheduled for 01FEB2013. I don't know why I took it out of order but hey whatever. If anyone has taken the 70-417 upgrade exam any pointers on what to look out for?
|
# ? Jan 19, 2013 04:43 |
|
VR Cowboy posted:Is that an American thing? Where I work we use A exclusively. B is the american standard. The rest of the world uses A. (Including the U.S. military at least overseas)
|
# ? Jan 19, 2013 04:59 |
|
eefster posted:I'm working on CCNA Voice at the moment, but hope to start on the CCNP R&S exams right after that. How did you get access to the physical equipment you mention? Was it purely stuff at work, a rack rental, or your own personal lab? Or did you go the simulator/emulator route - GNS3 / Packet Tracer / Boson / some other? A mix of personal lab and work lab. I think a sim will work mostly for the CCNA but I find that doing break/fix type stuff for CCNP learning is much more effective on physical equipment. I also keep GNS3 around to do some experimenting too.
|
# ? Jan 20, 2013 00:29 |
|
For MS certs, is it recommended to start off with MTA? Or can I dive straight in to MSCA? Looking at the paths for MSCA it looks like to get certified its 3 exams? The website doesn't specifically stat the requirement. It just lays out three steps.
|
# ? Jan 20, 2013 21:41 |
|
Kaethela posted:Any advice to someone who's planning on taking it? Specific stuff to memorize, out of left field questions, etc? I'd say pay special attention to the group policy/NTFS permission type questions. Basically the way they word some of the effective permission questions can be tricky. Remember IP address class ranges. Know the 802.11 versions and their broadcast ranges and speeds. Know the basic ports. As far as left field questions, I don't recall anything too crazy. Taking the practice test on the CompTIA website is similar to what you can expect when taking the real thing. If you're passing those with a high percentage, you'll be fine. It really isn't a hard test, its just that some of the questions asked require you to remember minor things, and multiple instances at that. I used the Mike Meyers all in one books for A+ and Network+. Good luck!
|
# ? Jan 20, 2013 21:56 |
|
Frag Viper posted:For MS certs, is it recommended to start off with MTA? Or can I dive straight in to MSCA? MTAs are quicker and cheaper, if you're not sure where you're headed or don't have the resources. Think: Associate's Degree vs Bachelor's. Plenty of people take the starter, and plenty of people skip it. Yes, for the Server and DB MCSAs, you take 3 specific exams. There's a pretty large knowledge base necessary so you wouldn't want to sit through a single 6-hour test (and then have to pay $500 to re-take it if you only bombed one section). I can't speak to any others (though, the desktop one appears to be only 2).
|
# ? Jan 21, 2013 01:42 |
|
Frag Viper posted:For MS certs, is it recommended to start off with MTA? Or can I dive straight in to MSCA? You're welcome to drive into the MCSA/MSCE track the MCSA for 2012 should be 3 tests 70-410, 70-411, and 70-412 It doesn't matter at all what order you take them in.
|
# ? Jan 21, 2013 02:05 |
|
How much of a rhyme or reason is there to the "X" part of the NBaseX naming scheme? I've noticed T for twisted pair, F for fiber, and S/L/E for shortwave/longwave/extended wave lasers, but haven't caught on to any patterns besides those.
|
# ? Jan 21, 2013 03:12 |
|
Alright, I'm switching to a new IT job that will not reimburse me for certs like my current one will. Trade-off is I should be getting hands-on experience with some Cisco equipment and Windows servers on-the-job. So I'll hold off on my plan to get an Enterprise Administrator cert at least until I get used to this new position.
|
# ? Jan 21, 2013 17:01 |
|
I kind of feel like an idiot because I failed the 70-648 (Upgrading MCSA to Server 2008) because I didn't realize there were two separate sections and you're required to pass both (I didn't pass the 70-640 section). I'd just like to ask, one more time, if anyone who has passed the 70-648 in the past has a good recommendation for a non-MS Press book? I've always gotten by just reading the MS books, studying the things I don't use through MS links, doing the labs, and from my own admin experience. It's also just kind of embarrassing that I'm just now getting around to trying to finish the 70-648. Also, gently caress Microsoft for not offering second chance vouchers for upgrade exams. tadashi fucked around with this message at 17:48 on Jan 21, 2013 |
# ? Jan 21, 2013 17:44 |
|
Sylink posted:Also, can anyone recommend me books to learn Apache/MySQL optimizing? I'm working with a magento webhost now and while I have setup LAMP before I would like to know more tips and tricks if you will. Weeks late on this, but I didn't see anyone reply to you. High Performance MySQL is really really good. It describes a lot of poorly understood and underappreciated features like partitioning and explains confusing configuration options in great detail. But more importantly, it drives home the idea that 99% of performance problems in DB's are related to terrible query design and/or terrible schema design That 100x performance gain you need is much more likely to come from rewriting your query to use an index than by finding the perfect magic value of some obscure setting in the config file.
|
# ? Jan 21, 2013 23:54 |
|
Ah, ok on the DBs. Maybe I don't give a poo poo then since I am supporting a known web framework/app and its not like I can change customer's website for them (we don't develop the sites, we only support the hardware as a webhost). Still probably a good book to check out. Sylink fucked around with this message at 15:06 on Jan 22, 2013 |
# ? Jan 22, 2013 15:03 |
|
Oh I definitely still recommend getting it. It discusses a lot of other important topics like backups, high availability and how to keep your DB's from growing forever, turning into a disk space black hole. It's a great book.
|
# ? Jan 22, 2013 16:16 |
|
Right now I'm learning how to subnet. And I have a question because its driving me crazy right now. So basically I'm having trouble understanding how to find the network id's after finding the subnet mask. Right now its asking me to break 192.168.4.0 /26 in to 3 (4) subnets. I know that since its /26 its going to be 255.255.255.192 It then says to see each of these network id's, first convert the original network id into binary. Which is: 192.168.4.0 11000000.10101000.00000100.00000000 Then it states to add the 4 different id extensions to the end. It then broke the binary down to: 11000000.10101000.00000100.00000000 11000000.10101000.00000100.01000000 11000000.10101000.00000100.10000000 11000000.10101000.00000100.11000000 This is where it confuses me because I'm not sure why or how it got arranged like that. How am I supposed to know that its supposed to be the following? 00000000 01000000 10000000 11000000 I'm completely fine if I use the "Happy Chart" to figure this out, but not in the sense this book is trying to explain it. Id like to have a firm understanding how to do it without using cheater charts and stuff.
|
# ? Jan 23, 2013 04:37 |
|
Oh ok, I can explain that now. It's because in a /26 you have 2 bits that get "borrowed" from the host bits to be used as network bits. the 00, 01, 10 and 11 combos are the four possible combinations for those two network bits.
|
# ? Jan 23, 2013 04:44 |
|
Inspector_71 posted:Oh ok, I can explain that now. It's because in a /26 you have 2 bits that get "borrowed" from the host bits to be used as network bits. the 00, 01, 10 and 11 combos are the four possible combinations for those two network bits. Ok, but how do I figure out the possible combos? Thats the ONE thing driving me crazy right now and I want to learn. This book is treating subnetting like Rosetta stone, it gives you enough information to have an idea, but it doesn't give you the vital information. Even that how to subnet a network .pdf doesn't fully explain it. It just gives you the jist and tells you to use the "happy chart"
|
# ? Jan 23, 2013 05:09 |
|
Frag Viper posted:Ok, but how do I figure out the possible combos? Thats the ONE thing driving me crazy right now and I want to learn. It's been a while since I've had to explain this so someone correct me if im wrong If you take the subnet mask and turned it into binary it would be 11111111.11111111.11111111.11000000 Take the network and just lay it on top like the slut it is 11000000.10101000.00000100.00000000 11111111.11111111.11111111.11000000 The last dotted quad has 2 bits as a part of the network mask so 2^2 is 4 which are 00, 01, 10, and 11. So you have four networks. You get their boundaries by just doing the math: 00 = .0 01 = .64 10 = .128 11 = .192 Have you touched v6 yet?
|
# ? Jan 23, 2013 05:24 |
|
Frag Viper posted:Ok, but how do I figure out the possible combos? Thats the ONE thing driving me crazy right now and I want to learn. Well, with the 2 bits it's simple since there are only 4 possible combos. I would probably stick with the 256-Mask=Block size way (that I posted about in the other thread) to figure out the networks, since if you really need the binary you can work backwards from there rather than trying to figure out all possible combos for 6+ bits.
|
# ? Jan 23, 2013 05:24 |
|
Inspector_71 posted:Well, with the 2 bits it's simple since there are only 4 possible combos. Is it really as simple as how doomisland just said? So in this case there's only 4. Is it simply asking me to count out the number of created id's in binary and lay them out in order and add the remaining 6 zeros to complete the octet? As in 00 =0 01 =1 10 =2 11 =3 and then .00000000 =0 .01000000 =64 .10000000 =128 .11000000 =192
|
# ? Jan 23, 2013 05:58 |
|
Frag Viper posted:Is it really as simple as how doomisland just said? So in this case there's only 4. Is it simply asking me to count out the number of created id's in binary and lay them out in order and add the remaining 6 zeros to complete the octet? Yeah. I still think it's a rather backwards way to figure out the networks, especially if you're dealing with a /23 and get to work out all possible combos for 7 bits rather than going from a known decimal to binary if you really need it. Inspector_666 fucked around with this message at 06:13 on Jan 23, 2013 |
# ? Jan 23, 2013 06:08 |
|
Jesus tap dancing christ... It was that all along. I can't wait until V6! Thanks for taking the time to answer my questions, both of you. This was driving me crazy. I'm going to go pour some scotch now and take the night off.
|
# ? Jan 23, 2013 06:28 |
|
Frag Viper posted:Ok, but how do I figure out the possible combos? Thats the ONE thing driving me crazy right now and I want to learn. There's an easier way to get the network addresses in decimal. 192. is a /24, your subnet is a /26. That means the octet in question has 2 bits of subnet and 6 bits of host. Each subnet in that octet therefore has 2^6 = 64 combinations of host bits, or in other words each one is sliced at a multiple of 64, starting at 0 and ending before 256. So 0, +64=64, +64=128, +64=196.
|
# ? Jan 23, 2013 06:46 |
|
Go even simpler. /24 is what we deal with every day and you know there are 256 IPs available. So every number higher than that halves it. Every number lower doubles the number of hosts. Just work it from there. So if /24 is 1 group of 256, /26 is 4 groups of 64, take out the network ID and broadcast, and you've got 1-62, 65-126, 129-190, 193-254. Those are the available IPs. Swear to god, I don't do it any more cleverly than that. If someone came to me and said, hey stripe, I need you to carve out a /1 subnet on 1.0.0.0, how many hosts would we have available, I'd just be like, oh hey calc.exe. 256. Times two. Times two. Times two.
|
# ? Jan 23, 2013 06:55 |
|
MC Fruit Stripe posted:Go even simpler. 2^(Host bits) = Available hosts, which is probably easier if you're doing a /1 (2^(Subnet bits) = Available subnets, also!) I think the easiest way to figure out block size is the Lammle method of 256 - Interesting octet = Block size.
|
# ? Jan 23, 2013 07:11 |
|
Networking is fun! Or at least it will be eventually.
|
# ? Jan 23, 2013 07:32 |
|
|
# ? May 16, 2024 18:09 |
|
psydude posted:Seriously, CompTIA? Who the gently caress still uses T568A. That's like asking for a network diagram on a token ring network. http://en.wikipedia.org/wiki/T568A#T568A_and_T568B_termination posted:The United States National Communication Systems Federal Telecommunications Recommendations explicitly forbid T568B, as does all US Federal guidance on new construction for the US Government. T568B is deprecated (removed) in the current TIA/EIA-568-C standard[citation needed] When I was taught (here in Australia), A was the standard one and B is the one you use for crossovers. Either way, T568A is not exactly obsolete like Token Ring or coax physical bus networks! Edit: Passed the CCNA exam last week. Had some questions that were definitely not CCNA material (checked through cert guides, looked in CCNP books). I'm guessing they were the beta unmarked questions the intro mentioned. Or maybe they are mentioned in the official 3rd edition cert guides. I studied from the 2nd edition which is from 2008. PancakeTransmission fucked around with this message at 12:00 on Jan 23, 2013 |
# ? Jan 23, 2013 11:52 |