| === zmoylan-1i is now known as zmoylan-pi | ||
| === Lcawte is now known as Lcawte|Away | ||
| diplo | Morning all | 07:03 |
|---|---|---|
| MooDoo | morning | 07:25 |
| popey | morning | 08:02 |
| bashrc | g'day | 08:03 |
| TheGeek | mornin | 08:09 |
| popey | yo | 08:09 |
| TheGeek | hoho | 08:10 |
| mjayk | good morning | 08:11 |
| davmor2 | Morning all | 08:11 |
| === Lcawte|Away is now known as Lcawte | ||
| JamesTait | Good morning all; happy Friday, and happy Peanut Butter Cookie Day! 😃 | 08:54 |
| popey | oooh | 08:59 |
| popey | i bought some peanut butter yesterday. I do believe it's peanute butter and banana sandwich day | 08:59 |
| JamesTait | Food of kings. | 09:00 |
| brobostigon | morning boys and girls. | 09:26 |
| bashrc | morning | 09:32 |
| brobostigon | morning bashrc | 09:32 |
| mapps | morninghm | 09:32 |
| brobostigon | morning mapps | 09:33 |
| mapps | was at the casino went to go home | 09:33 |
| mapps | got a pack of cigs out my bag and a drink | 09:33 |
| mapps | and wow | 09:33 |
| mapps | #you have drink he knows the deal' | 09:34 |
| mapps | nope..im just an alcoholic so i ad it with me | 09:34 |
| brobostigon | something is really wrong with bt today, i am getting over 90% packet loss to my vps's, however is working form my phone. | 09:38 |
| mapps | i havw bt and virgin:P | 09:39 |
| brobostigon | ok, | 09:42 |
| popey | level 3 are down apparently | 09:58 |
| brobostigon | level 3? | 09:58 |
| popey | yes, level 3 | 09:59 |
| brobostigon | what does it mean? | 09:59 |
| popey | they provide internet backbone | 10:00 |
| popey | maybe fixed now | 10:00 |
| Laney | I haven't noticed any problems | 10:00 |
| Laney | guess I am level 3 free | 10:00 |
| brobostigon | ah i see. | 10:00 |
| davmor2 | as some of you may have noticed, the Internet at large is seeing some "bad weather" right now. This should resolve shortly as the major network carriers route around the problem | 10:01 |
| Laney | I see what you did there | 10:02 |
| Laney | :) | 10:02 |
| davmor2 | brobostigon: that was Internet Services announcement on the problems :) | 10:03 |
| brobostigon | davmor2: ok, ty. | 10:04 |
| brobostigon | the pebble forums seem to work fine. | 10:07 |
| * popey wonders when his Pebble will arrive | 10:07 | |
| davmor2 | popey: when you going out next? I bet it arrives then | 10:07 |
| popey | not shipped yet | 10:08 |
| popey | so unlikely :) | 10:08 |
| davmor2 | popey: so when you go out next after it ships :) | 10:08 |
| mapps | ;time for russian learning | 10:08 |
| mapps | ;D | 10:08 |
| popey | probably | 10:08 |
| mapps | im spending 10hrs a wek learning laguages | 10:09 |
| mapps | no popey i meant im tryin to learn | 10:09 |
| mapps | trying to lean spanish and russian | 10:09 |
| davmor2 | mapps: you trying to out do Christopher Lee? | 10:09 |
| * brobostigon tests to see what work and what doesnt. | 10:09 | |
| davmor2 | brobostigon: looking at this most things should be up again now ish | 10:10 |
| brobostigon | davmor2: ok, ty. | 10:10 |
| mapps | lol | 10:10 |
| mapps | nah davmor2 | 10:11 |
| mapps | im just a simle southener like popey | 10:11 |
| mapps | i speakk eng/fenchgeman/arbic | 10:11 |
| mapps | so i figure spanish can help | 10:11 |
| davmor2 | mapps: he was born in London so now you have no excuse :P | 10:13 |
| mapps | im gibraltaian mate] | 10:16 |
| mapps | no idea what london is | 10:16 |
| mapps | ;) | 10:16 |
| popey | I'd quite like to learn mandarin | 10:23 |
| bujji | how to find load average ...of a syatem | 10:23 |
| popey | top | 10:23 |
| brobostigon | uptime | 10:23 |
| davmor2 | htop | 10:24 |
| bujji | load average like 0.05 0.12 0.66 what does this mean | 10:24 |
| bujji | tload | 10:24 |
| bujji | w | 10:24 |
| davmor2 | bujji: http://blog.scoutapp.com/articles/2009/07/31/understanding-load-averages | 10:24 |
| popey | lulz https://twitter.com/TMCorp/status/609167065300271104 | 10:35 |
| === Larva is now known as Guest47042 | ||
| bujji | davmor2:how it will calculate for 1 minute 0.05 | 10:39 |
| davmor2 | popey: just as well he wasn't on a mountain doing that ;) | 10:41 |
| davmor2 | bujji: read the article if that doesn't tell you have a search on google there is bound to be an article some where that will give you a lot more info. that was just the first I hit on a google search | 10:41 |
| bujji | davmor2:is it based on calculating how many processors running on the system | 10:42 |
| bujji | process* | 10:43 |
| popey | bujji: no, it's how many processes on average are in the queue for a processor. | 10:44 |
| popey | (put simply) | 10:44 |
| popey | if you start 20 long running processes and you have 1 cpu, 1 core, then the number will climb very quickly until those jobs complete | 10:44 |
| popey | if you have 20 cpus (20 cores) then there will be (more or less) one job per core, so the number won't rise. | 10:44 |
| popey | it's a measure of how busy the box is. | 10:45 |
| bujji | how can i count processes running on my sysytem | 10:45 |
| intrbiz | bujji: ps | 10:46 |
| intrbiz | bujji: best results: ps aux | 10:46 |
| popey | top tells you | 10:46 |
| popey | Tasks: 312 total, 2 running, 309 sleeping, 0 stopped, 1 zombie | 10:46 |
| popey | some will be idle | 10:47 |
| davmor2 | popey: kill the zombie | 10:47 |
| popey | or "sleeping" | 10:47 |
| intrbiz | can't kill a zombie | 10:47 |
| bujji | thats good.. | 10:47 |
| davmor2 | popey: it will try to eat your brainz | 10:47 |
| popey | its eating the cpu | 10:47 |
| popey | sd_cicero | 10:47 |
| popey | well, it was | 10:47 |
| bujji | 309 sleeping...how it will calculat for one minute.. | 10:47 |
| popey | calculate what? | 10:48 |
| popey | thats realtime, what's happening right now | 10:48 |
| popey | load average is averged over 1, 5 and 15 minutes | 10:48 |
| bujji | for one minute and 5 minute the load avg will be calculated right.. | 10:49 |
| popey | it's calculated, yes. | 10:49 |
| bujji | how it will calculate i am asking.. | 10:49 |
| popey | dunno, look at the source code | 10:49 |
| intrbiz | load average is exponetntially weighted moving averages for 1 minute, 5 minute and 15 minute | 10:49 |
| intrbiz | as such the 15 and 5 minute values decay slower | 10:50 |
| bujji | that i got it. | 10:50 |
| davmor2 | bujji: http://www.howtogeek.com/194642/understanding-the-load-average-on-linux-and-other-unix-like-systems/ | 10:50 |
| intrbiz | bujji: load average is really not worth worrying about unless la > # cpus | 10:51 |
| intrbiz | ie on a 2 cpu system, a 5 minute load average of 2 mean your maxing out both cpus for 5 minutes | 10:51 |
| davmor2 | bujji: and https://en.wikipedia.org/wiki/Load_(computing) | 10:52 |
| intrbiz | a load average of 10 for 5 minutes effectively means your need 10 cpus to handle the load for that time period | 10:52 |
| popey | also the load average isn't a good measure when you can turn CPUs on and off at will, so the maths gets a little fuzzed | 10:52 |
| bujji | then what is good measure.. | 10:54 |
| intrbiz | lno such thing as one good metric | 10:54 |
| intrbiz | s/lno/no/ | 10:54 |
| bashrc | temperature of the server? | 10:54 |
| intrbiz | a good person considers a range of metrics, as they measure different things | 10:54 |
| bujji | i will come to know this again..)) | 10:55 |
| bujji | intrbiz:i will let you know | 10:55 |
| bujji | bye for now)) | 10:56 |
| === alan_g is now known as alan_g|lunch | ||
| === Lcawte is now known as Lcawte|Away | ||
| === Lcawte|Away is now known as Lcawte | ||
| === alan_g|lunch is now known as alan_g | ||
| popey | https://www.yahoo.com/parenting/the-type-of-parents-most-likely-to-have-a-child-121212598657.html | 13:10 |
| popey | interesting! | 13:10 |
| foobarry | are direct messages between 2 freenoders logged ? or possible to log on the server? | 13:13 |
| awilkins | foobarry, Reading it, you need DCC CHAT for actually private messages | 13:15 |
| awilkins | I think just normal /msg does go through server | 13:15 |
| awilkins | As to whether it's logged, I don't know freenode's config / policy on that | 13:15 |
| popey | ask Dave | 13:16 |
| popey | Apparently D ave is in a field somewhere and so cannot respond. | 13:48 |
| popey | Isn't that right Dave ? | 13:49 |
| popey | foobarry: dave says "no" | 13:49 |
| foobarry | really? thnaks | 13:49 |
| foobarry | i had concerns about private chats when both ircers are on the same irc node | 13:50 |
| foobarry | node/server | 13:50 |
| popey | he said they don't log anything like that | 13:50 |
| foobarry | and can't? | 13:50 |
| popey | can't what? | 13:50 |
| foobarry | its technically possible i guess | 13:51 |
| foobarry | if i was on a ircnet server | 13:51 |
| popey | if the server was compromised, and someone ran tcpdump, maybe | 13:51 |
| foobarry | ok thanks | 13:52 |
| popey | np | 13:52 |
| foobarry | thats outside the scope of my concerns | 13:52 |
| popey | Thanks Dave :) | 13:52 |
| foobarry | Dave is awesome | 13:52 |
| shauno | this conversation reads like Dave is popey's imaginary friend o_O | 14:07 |
| davmor2 | I'm here | 14:09 |
| awilkins | Are you doing science, and you're still alive? | 14:14 |
| popey | heh | 14:15 |
| bashrc | science! | 14:19 |
| popey | Telegram! | 14:28 |
| bashrc | I've not used Telegram. I think it's some Russian thing with a centralised server | 14:29 |
| popey | The guy who started it is indeed Russian | 14:31 |
| popey | I don't think that makes it a Russian Thing | 14:31 |
| bashrc | https://telegram.org/faq#q-can-i-run-telegram-using-my-own-server | 14:31 |
| bashrc | over HTTP. That might make it a little more resistant to blocking | 14:34 |
| jpds | bashrc: Based in Berlin. | 14:34 |
| bashrc | with E2EE, apparently | 14:35 |
| davmor2 | bashrc: it can have you can also have messages that self destruct too | 14:38 |
| awilkins | No such thing as a self-destructing message | 14:38 |
| davmor2 | bashrc: that wipes it from yours, the recipient and the server | 14:38 |
| awilkins | There's messages the app will destroy | 14:38 |
| awilkins | But it's an open protocol | 14:38 |
| awilkins | That means you can write a client that doesn't respect the self-destructing | 14:38 |
| davmor2 | telegram for the win | 14:39 |
| bashrc | yes, it would be difficult to self-destruct a message, but it maybe could exist in the system for some amount of time before deletion - similar to Bitmessage | 14:39 |
| awilkins | One way you could do it is post a message encrypted with a key that is only available online that your client obtains live to read it | 14:40 |
| awilkins | But again, nothing stops a modified client retaining the key, or the plaintext | 14:40 |
| bashrc | yes | 14:40 |
| awilkins | It's the DRM problem - once you provide the client with the key and the content, nothing stops them using it however they want | 14:40 |
| awilkins | (eventually) | 14:41 |
| zmoylan-pi | it's the same problem that people had back in the days of actual telegrams where people used personal codes as there was no way to ensure security of the communication links | 14:42 |
| === Lcawte is now known as Lcawte|Away | ||
| === Lcawte|Away is now known as Lcawte | ||
| === alan_g is now known as alan_g|EOW | ||
| awilkins | re: telegram : http://security.stackexchange.com/questions/49782/is-telegram-secure | 17:10 |
| shauno | I thought someone found a few months ago, that it was leaving decrypted convos on disk after they'd been deleted | 17:20 |
| shauno | http://blog.zimperium.com/telegram-hack/ | 17:25 |
| shauno | it has a cache where they're unencrypted at rest, even after they've been "self-destructed". and since the primary threat on a mobile is losing the damned thing, "at rest" is a pretty dodgy hole | 17:27 |
| sebsebsebb | hello | 17:53 |
| bujji | does sym link and hard link takes memory on hard disk? | 18:58 |
| daftykins | mere bytes i should think | 19:00 |
| bujji | ? | 19:04 |
| bujji | hardlink i node number is same right | 19:04 |
| bujji | daftykins:symlink does not taking memory | 19:12 |
| bujji | hardlink does | 19:12 |
| daftykins | makes sense | 19:13 |
| daftykins | hard link means 'copy the file here too but link it to the original for changes' right? | 19:13 |
| bujji | here why we need of creating hard link files.....in real way ???? | 19:16 |
| bujji | daftykins:rc3.d directory contains sym link files...thats ok | 19:18 |
| daftykins | sorry can't understand you at all. | 19:19 |
| bujji | runlevel 3 directory conatains sym link files... | 19:19 |
| daftykins | that's a statement not a question | 19:20 |
| daftykins | and i'm the wrong person to ask :) | 19:20 |
| bujji | usecase ok for sym links over there.. | 19:22 |
| bujji | why we need creating hard link here..? | 19:23 |
| daftykins | too busy to look things up for you i'm afraid | 19:24 |
| daftykins | and the language barrier is still an issue | 19:24 |
| bujji | are you understanding my question.. | 19:25 |
| daftykins | no, but as i say i'm too busy anyway :) | 19:27 |
| bujji | take your time)) | 19:28 |
| daftykins | no i can't help you :) | 19:29 |
| bujji | let expect from others. | 19:30 |
| shauno | hardlinks are a whole different kettle of interesting. symlinks are like shortcuts. hardlinks are .. weird | 19:30 |
| daftykins | we would probably need to know the task at hand to know which makes more sense | 19:31 |
| daftykins | http://bit.ly/1L3ntX4 | 19:31 |
| daftykins | bah, £75 difference between two Lenovo X1 Carbons | 19:31 |
| bujji | shauno:use case of hardlink)) | 19:31 |
| daftykins | cheaper one has a 1080p screen but it's TN... dearer one has a WQHD (2560x1440) but drops to an i5 | 19:31 |
| shauno | hardlinks are interesting because they're all equal. if you create a symlink to a file, and then delete the file, the symlink is now useless | 19:32 |
| shauno | with hardlinks, if you link to a file, and then delete the original, the hardlink you created still points to the file | 19:32 |
| intrbiz | hardlinks are multiple names to the inode, so would likely take space in the dentry | 19:33 |
| bujji | intrbiz:inode number for this same for hard link but taking memory i am not understanding here use case | 19:34 |
| bujji | intrbiz:hardlink both acts like a original files | 19:35 |
| intrbiz | bujji: what do you mean 'taking memory' ? | 19:35 |
| bujji | du | 19:35 |
| intrbiz | bujji: inode tracks the extends of a file, a hard link, maps multiple names to an inode, it does not duplicate the inode nor the extents | 19:36 |
| intrbiz | s/extends/extents/ | 19:36 |
| bujji | intrbiz:inode is same right | 19:37 |
| intrbiz | bujji: ? | 19:38 |
| bujji | intrbiz:inode is same for hd link | 19:39 |
| intrbiz | bujji: yes, as I said, a 'hard link' is just mutliple names pointing to a file | 19:39 |
| intrbiz | bujji: the name of a file is not part of the inode | 19:39 |
| bujji | but if we delete original it still there right. | 19:41 |
| intrbiz | bujji: when you delete a hard link, the link count of the inode is decremented, when it reaches zero (IE: nothing points to the inode) it will be removed from disk | 19:41 |
| intrbiz | bujji: so if you have a hard link to a file, and you remove either the hard link or the file, the data will still remain | 19:42 |
| intrbiz | bujji: why all the questions on symbolic / hard links? | 19:43 |
| bujji | how to delete hard link..? | 19:43 |
| shauno | the same way as any other file. it's literally the same thing. this is why the OS actually calls delete 'unlink' | 19:43 |
| bujji | intrbiz:run level directories contains sym link files.. | 19:44 |
| intrbiz | bujji: yes, sym links make more sense for that use case | 19:45 |
| bujji | intrbiz:these are pointing to the /etc/init.d services | 19:45 |
| intrbiz | bujji: on old distros yes | 19:45 |
| bujji | intrbiz:yes,right | 19:45 |
| bujji | intrbiz:in the same way hard link use case why wwe need that one | 19:46 |
| intrbiz | bujji: hard links are useful in certain situations, for example not wanting the traversal overhead of a symlink and to avoid issues where with a symlink the target can be deleted leaving a dead link | 19:47 |
| bujji | intrbiz:not getting | 19:48 |
| shauno | time machine is actually an interesting use of hardlinks | 19:50 |
| intrbiz | pg_upgrade uses hardlinks too | 19:50 |
| shauno | I know y'all meant to hate macs and stuff, but this is all pretty generic to just unix | 19:50 |
| shauno | each time it makes a backup, it puts the backup in a timestamp'd folder. (think, /backups/2015-06-12-2051) | 19:51 |
| intrbiz | rsnapshot also makes use of hardlinks | 19:51 |
| shauno | then for each file that's unchanged, it hardlinks it to the previous version | 19:51 |
| shauno | this way each backup is 'complete', but you don't actually have 10 copies of the same file on disk | 19:52 |
| shauno | you have 10 copies of the filename, but they're all using the same actual inodes on disk, so they don't consume 10x the space | 19:53 |
| shauno | yeah, I think it's quite similar to rsnapshot. time machine as a backup mechanism isn't very fancy at all. the actual 'features' are how well it's integrated into the rescue process; and that they've made it so simple even my mother has no excuses | 19:54 |
| intrbiz | opensuse integrates btrfs based snapshot really well into the boot process now | 19:55 |
| intrbiz | can just select an older copy of the system to run from grub | 19:56 |
| bujji | hardlinks are not applicable for directories right.. | 19:56 |
| shauno | right, a directory doesn't actually 'exist' | 19:56 |
| intrbiz | bujji: nope | 19:56 |
| intrbiz | bujji: a directory and a file, are merely names | 19:57 |
| bujji | when i execute it is saying that | 19:57 |
| shauno | the filesystem is made up of two almost unconnected systems | 19:57 |
| intrbiz | directories and files are dentrys, the contents of a file are inodes | 19:57 |
| shauno | one is actually storing data; this is just inodes tracking where the contents of a file actually live on disk | 19:58 |
| shauno | and then you have the actual organization of sticking names, permissions, directories, namespaces, etc so you can find these inodes | 19:58 |
| bujji | ls -id . and cd <dir name> ls -id .. is having same inode number..? | 20:02 |
| bujji | here the hard link applicable for directories ...? | 20:03 |
| intrbiz | bujji: hard links are not applicable to directories | 20:03 |
| bujji | intrbiz:. and .. are directories right | 20:04 |
| intrbiz | bujji: they aren't real directories, '.' just means the current directory and '..' parent directory | 20:05 |
| bujji | those are existing right.. | 20:06 |
| bujji | intrbiz:cd .. in the sense going to parent directory right. | 20:06 |
| intrbiz | bujji: cd = change directory , .. = parent directory | 20:07 |
| bujji | cd / nad type; ls -id . and then cd opt:and type; ls -id .. both are having same inode number.. | 20:09 |
| intrbiz | bujji: of course they will | 20:11 |
| intrbiz | bujji: /. = / and /opt/.. = / | 20:11 |
| bujji | intrbiz:that doesnt mean hard applied for that | 20:12 |
| intrbiz | bujji: no | 20:12 |
| bujji | hard link | 20:12 |
| intrbiz | bujji: . and .. are not real, they do not physically exist, they are a presentation for navigation | 20:13 |
| bujji | intrbiz:if i try to delete that one saying can not remove directory. | 20:15 |
| bujji | intrbiz:that doesnt mean hard coded.. | 20:16 |
| shauno | which one? | 20:16 |
| bujji | . nad .. | 20:16 |
| bujji | and* | 20:16 |
| intrbiz | bujji: what are you actually asking? | 20:19 |
| bujji | . and .. is having same inode number that means hard applied over there. | 20:21 |
| intrbiz | bujji: no | 20:24 |
| shauno | hm, directories do appear to have inode numbers. that's completely not how I understood it lol | 20:26 |
| shauno | bujji: I'm not sure this will help, but you might want to look at 'stat' instead of trying to remember all ls's flags :) | 20:26 |
| bujji | intrbiz:then how it can be linked...is that soft link(no) because inode is same | 20:26 |
| intrbiz | bujji: its not a link, '.' and '..' do not exist | 20:27 |
| bujji | intrbiz:we are changing from directory to directory right | 20:28 |
| intrbiz | bujji: ? | 20:28 |
| shauno | bujji: what he means is that . and .. don't actually exist | 20:29 |
| shauno | like, / is a directory, /tmp is a directory. /tmp/. isn't a directory, it just *means* /tmp | 20:29 |
| shauno | and /tmp/.. isn't a directory, it just means / | 20:29 |
| bujji | intrbiz:like...cd /etc/init.d here etc and init.d have some link? | 20:29 |
| shauno | it's like, today and yesterday aren't on the calendar. the 11th and 12th are. 'today' and 'yesterday' are just convenient ways we can reference them | 20:30 |
| shauno | the same way, . and .. aren't actually on the filesystem | 20:31 |
| bujji | shauno:past u have some actions and present you have some actions you had some link. | 20:32 |
| shauno | I mean they're just concepts. they're not actually things | 20:35 |
| bujji | shauno:here present directory you have done some actions and parent directory have done some actions the time stamp will be change if you observe. | 20:36 |
| shauno | right, say I'm in /tmp | 20:37 |
| shauno | and I touch a file named 'something' | 20:37 |
| bujji | okey | 20:37 |
| shauno | the timestamp on /tmp has changed. you can see this with '.' because '.' just means 'here' | 20:37 |
| shauno | . hasn't actually changed. . doesn't actually exist. but if I look at . I see /tmp | 20:38 |
| bujji | cd .. and see current directory. | 20:38 |
| bujji | shauno:got it? | 20:40 |
| shauno | heh, I've had it for almost 20 years. I apparently just have no idea how to clearly explain it :) | 20:41 |
| bujji | can you recollect)) | 20:43 |
| intrbiz | bujji: the path '/home/test/..' actually means '/home/' | 20:44 |
| intrbiz | bujji: the path '/home/test/.' actually means '/home/test/' | 20:44 |
| intrbiz | bujji: the . and .. are used to represent here and parent respectively, the are dealt with when canonicallising the path | 20:46 |
| bujji | intrbiz:absolute path | 20:47 |
| intrbiz | bujji: statements != questions | 20:48 |
| bujji | intrbiz:for example chain link...that does have a link between one after the other right..like /home/test/ | 20:49 |
| bujji | intrbiz:if we delete "test" there is no link with that. | 20:50 |
| intrbiz | bujji: . and .. are not links | 20:50 |
| intrbiz | if I remove June 12th from the calendar, the concepts today and tomorrow still exist | 20:51 |
| bujji | intrbiz:can you tell me why hard links are not applicable for directories. | 20:52 |
| intrbiz | bujji: . and .. are just concepts | 20:52 |
| bujji | intrbiz:you are removing there 12th actions | 20:53 |
| intrbiz | bujji: directories and files are both merely names. they exist in a heirarchial tree structure. internally these names are stored as a dentry structure | 20:53 |
| intrbiz | bujji: an inode structure is used to track disk blocks used to store stuff | 20:54 |
| intrbiz | bujji: a link maps a dentry to an inode | 20:54 |
| bujji | intrbiz:yes | 20:54 |
| bujji | can you tell me why hard links are not applicable for directories. | 20:55 |
| intrbiz | bujji: because directories are merely an entry in the tree, if you allowed directory hard links, then all sorts of chaos could ensue, such as circular references, etc | 20:57 |
| intrbiz | note the inode of a directory merely tracks the disk blocks used to store the dentry structures | 20:57 |
| intrbiz | bujji: another issue with hard links for directories would be having multiple parents | 20:58 |
| bujji | intrbiz:yes you come to the point now..got it | 20:59 |
| bujji | intrbiz:here in linux file system is having single parent right. | 21:00 |
| bujji | intrbiz:/ | 21:01 |
| intrbiz | bujji: / is the Virtual File System (VFS) root yes | 21:02 |
| bujji | intrbiz:here you have diff sub directories | 21:02 |
| bujji | intrbiz:each and every directory reffered to the parent directory right. | 21:03 |
| intrbiz | bujji: a child knows its parent | 21:04 |
| bujji | intrbiz:yes,there is a link? | 21:04 |
| intrbiz | bujji: the dentry structure has a pointer to the parent dentry | 21:07 |
| bujji | intrbiz:how it can be able point? | 21:08 |
| intrbiz | bujji: by pointer I mean C pointer | 21:08 |
| bujji | which points to the address of another.. | 21:10 |
| bujji | intrbiz:if we allow h link... issue with hard links for directories would be having multiple parents here i got the point thanks | 21:12 |
| bujji | intrbiz:unclear for . and ..(because of having same i node number)i thought here hard link applied | 21:14 |
| bujji | intrbiz:its not allowing me to delete . and .. why so? | 21:15 |
| intrbiz | bujji: what directory are you in? | 21:15 |
| shauno | if I start crying, please promise not to tell anyone ;) | 21:15 |
| intrbiz | ok | 21:16 |
| bujji | intrbiz:/home/bujji | 21:17 |
| bujji | shauno:? | 21:17 |
| intrbiz | bujji: so . would mean remove '/home/bujji' and .. would mean '/home' | 21:18 |
| intrbiz | bujji: you'd only be able to remove them if they are empty | 21:19 |
| bujji | intrbiz:let me try if they empty | 21:19 |
| intrbiz | bujji: but as we've said, '.' and '..' mean here and parent. if you do rmdir on . or rmdir on .. then you remove the directories that represent that | 21:20 |
| bujji | intrbiz:not working if i used if they empty "rm -rf ." | 21:23 |
| intrbiz | bujji: note no same person would do 'rmdir .' or 'rmdir ..' | 21:23 |
| intrbiz | s/same/sane/ | 21:24 |
| bujji | intrbiz:it saying that "cant remove directory" | 21:25 |
| bujji | intrbiz:that does mean | 21:27 |
| intrbiz | bujji: if you look at the manpage for rmdir (2), it states that the path cannot end in '.' and will produce invalid error if it does | 21:28 |
| bujji | intrbiz:yes,if i use "rmdir ." but i used "rm -rf ." it is saying "rm:cant remove directory" | 21:31 |
| intrbiz | bujji: it looks like rm will refuse to remove . and .. | 21:32 |
| bujji | intrbiz:yes? | 21:33 |
| intrbiz | bujji: yes | 21:33 |
| bujji | intrbiz:why? | 21:34 |
| intrbiz | bujji: because . and .. and real, so it makes no sense to ask to remove them, so the command ignores the request | 21:35 |
| bujji | intrbiz:rm by default doesnot allow to remove directories. | 21:37 |
| intrbiz | bujji: if you want to remove directories with 'rm' you need to use recursive '-r' | 21:38 |
| bujji | intrbiz:same thing | 21:39 |
| bujji | intrbiz:can you give me use case in applying hard link on files. | 21:41 |
| intrbiz | bujji: we listed a few earlier, i'm not listing them again | 21:41 |
| === ahayzen_ is now known as ahayzen | ||
| bujji | intrbiz:can you give me example hard link files in system uses. | 21:47 |
| bujji | intrbiz:dfault hard link files in linux machine. | 21:50 |
| intrbiz | bujji: not sure of any default uses, hard links are usually sparingly used | 21:50 |
| bujji | intrbiz:like sym links there in /etc/rc3.d/? | 21:52 |
| intrbiz | ? | 21:52 |
| intrbiz | bujji: sym links are alot more common than hard links | 21:52 |
| bujji | intrbiz:use case only i can able to get the point. | 21:54 |
| bujji | intrbiz:thanks man you make me clear some points. | 21:56 |
| bujji | intrbiz:i got understand that hard link not applicable for directories because ( issue with hard links for directories would be having multiple parents) right | 21:59 |
| intrbiz | one of the reasons yes | 22:00 |
| bujji | intrbiz:and circular references. | 22:01 |
| bujji | intrbiz:thanks for your time man bye:))) | 22:03 |
| ball | Are "Unity Web apps" intended for Ubuntu on a phone? | 22:41 |
| ging | i have an ubuntu server which at the end of booting tty1 just hangs never gets to the login prompt, i can't find any errors and eveyrthing else works, except tty1, anyone have an ideas on what might cause this or where to look? | 22:46 |
| ball | Is tty1 an actual serial port or a virtual console on the monitor and keyboard? | 22:46 |
| ging | well it's a vm | 22:47 |
| ball | What hypervisor? | 22:47 |
| ging | kvm | 22:47 |
| ball | Ah. I've never used that. No idea how it works. | 22:47 |
| ging | i think it emulates a vga monitor, or atleast the way i have it setup | 22:48 |
| ging | i can switch to any other tty and they are fine | 22:48 |
| ball | That's a curious choice. | 22:49 |
| ging | it is so you can connect to it via vnc | 22:49 |
| ball | VNC doesn't require a monitor. | 22:49 |
| ball | (at least at the server end) | 22:50 |
| ball | ...are you saying that it uses VNC to simulate a VGA monitor? | 22:51 |
| ball | (and adaptor) | 22:51 |
| ging | yeah pretty much | 22:53 |
| ball | Wierd. | 22:53 |
| ging | it has a vga adaptor a a pci device | 22:53 |
| ball | I suppose it made sense to someone. | 22:54 |
| ging | seems to be the default way or atleast using virt manager | 22:54 |
| ball | Is that really a KVM thing or just something that you've chosen to do in the vm? | 22:54 |
| ball | That's pretty "out there". | 22:54 |
| ging | i think vmware and virutal box are usally setup similarly | 22:56 |
| daftykins | does dmesg or the boot log shed any light? | 22:56 |
| ball | VMware isn't. | 22:56 |
| ball | I haven't seen Virtual Box in years. | 22:57 |
| daftykins | never found it performs the same as vmware personally | 22:57 |
| ging | daftykins: not really, i have an identical server and the logs look the same until the point that this one just stops | 22:57 |
| ging | the stuff before it stops gives no reall clue | 22:58 |
| ball | ging: Is this after you've installed the guest OS? | 22:58 |
| ball | ging: Does the OS you're installing have a text install option? | 22:58 |
| daftykins | version and kernel version? | 22:59 |
| ging | ball: yes it's ubuntu server 14.04 | 22:59 |
| ball | ging: Did you try the text install? | 22:59 |
| ging | Ubuntu 14.04.2 LTS (GNU/Linux 3.13.0-53-generic x86_64) | 23:00 |
| ging | ball: i am not having a problem installing it, my problem is with how it boots | 23:01 |
| daftykins | is the one that works identical? | 23:01 |
| daftykins | ging: there's a -54 kernel now so #1 i'd update at least | 23:01 |
| ball | ging: fwiw the first time I tried Ubuntu Server I was disgusted that it used a graphical splash screen. I've mostly got over that but I wouldn't be surprised if it caused issues in some environments. | 23:03 |
| ball | ging: I just put it down to me being an old fart. | 23:03 |
| daftykins | plymouth on a server is indeed pants-on-head retarded | 23:04 |
| ball | What is "plymouth"? | 23:04 |
| ging | yeah what is plymouth that may be the issue | 23:04 |
| ging | it repeatedly spawns on the one with the issue | 23:05 |
| daftykins | is your VM on solid state storage? | 23:05 |
| ging | init: plymouth-upstart-bridge main process (200) terminated with status 1 | 23:05 |
| daftykins | no that's not related | 23:05 |
| ging | no it's on a ceph cluster | 23:05 |
| daftykins | i don't have a clue what that is :) | 23:06 |
| ging | it's an rbd storage cluster | 23:06 |
| ball | What is "ceph"? | 23:06 |
| ball | Linux is strange. | 23:07 |
| daftykins | ok well on fast storage you tend to see the above error due to a timing issue, there are plenty of blog posts around online about tweaking a config to stop it thinking something is broken | 23:07 |
| ball | Still, it works well enough on my daughter's PC. | 23:07 |
| daftykins | if one doesn't ask too much of things they can often seem great :) | 23:08 |
| ging | daftykins: they are almost exactly the same, they are a pair, one is redundant the one with the issue, they have a keepalived script which mounts another drive and starts some services on the primary | 23:12 |
| ging | it is like the other one thinks it's not finnished booting because this other stuff hasn't happened, but i don't see why | 23:13 |
| daftykins | anywho as i say there's a newer kernel so a dist-upgrade is in order before pursuing that one further | 23:13 |
| ging | i think it is my config | 23:25 |
| ging | well not mine i didn't build it | 23:26 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!