=== skrewler_ is now known as skrewler [02:32] hello [02:32] does anyone tested/using EBS based instance ? [02:33] sure [02:33] i don't use instance-store [02:33] does it work ? [02:33] of course [02:33] otherwise i wouldn't use it [02:34] though this in ec2 not UEC. [02:34] i don't use eucalyptus/uec coz its buggy as [02:35] : so u use ? [02:35] huh [02:36] u dont use eucalyptus/uec then what u using now [02:36] flaccid: though this in ec2 not UEC. [02:37] otherwise openstack or cloudstack for private cloud [02:38] Amazon Elastic Compute Cloud ? [02:38] yes [02:38] need to pay though [02:39] indeed [02:40] there is the aws free tier though tm [02:40] *atm === dendro-afk is now known as dendrobates [02:43] : did u tested EBS based instance exercise ? [02:44] flaccid: sure [02:44] i build images too, that includes ebs AMIs. [02:44] $ euca-run-instances -t m1.large -k [02:44] flaccid: i don't use eucalyptus/uec coz its buggy as [02:44] anything else i should repeat? [02:45] Note: that the EMI used here cannot be the loader EMI provided in step 2 [02:46] it say that EMI not the loader EMI provided in step 2, then what should i put [02:46] if someone can help, they'll respond. please be patient. [02:54] i can give a tip though, read the docs [03:00] i'm confusing thats why i ask for help [03:01] http://www.google.com/search?client=opera&rls=en&q=euca-run-instances+ebs+emi&sourceid=opera&ie=utf-8&oe=utf-8&channel=suggest === dendrobates is now known as dendro-afk [06:37] cant execute this command [$ eval $(blkid lucid-server-uec-amd64.img | cut -f2- -dā€ [06:37] ā€œ) [06:37] ] what wrong ? [06:40] eeek, that looks dodgy [06:40] also most of it came through as incompatible encoding [06:40] and i didn't see an error output or anything. [06:41] then the execise give wrong command ?? [06:59] <_et> ttx: hi. Was there at UCD. Got a clarification to make. good time? [07:03] <_et> can anyone please clarify how control flows when there is an API call to attach a volume to an instance? [07:03] <_et> nova places a request for a volume on the queue and then does the control flow to the scheduler or to the nova-volume? [07:14] eval $(blkid lucid-server-uec-amd64.img | cut -f2- -dā€ [07:14] ā€œ) [07:14] should i replace blkid with value ?? [07:43] davinci11: sounds more like pebkac to me. still i have no idea what you are following. [07:50] another funny thing $ euca-attach-volume $vol -i -d /dev/sdb [07:50] [07:50] if you want help, you are may have to learn to ask better. [07:50] sdb doesn't show up in /dev [07:51] its probably not in your block device mapping for the instance or image [07:51] oh [07:51] this is to attach [07:51] it won't be there until you attach.. [07:51] you are telling it where to attach. [07:51] alr attached [07:52] as shown in hybridfox [07:52] doesn't sound like the attachment was successful. [07:52] if its not in fdisk -l or /dev/sdb it obviously didn't attach [07:53] should i change sdb to other ? [07:53] you could try i guess [07:54] and personally i wouldn't rely on hybridfox either [07:54] though i wouldn't rely on eucalyptus at all :) [07:54] @@ [07:55] i tried /dev/vda [07:55] but.... in /dev [07:55] i found vdb [07:57] davinci11: check the instances disks with $ sudo fdisk -l - the one without partition table should be the volume [08:00] i attach /dev/vda again but this time show me vdc [08:00] when sudo fdisk -l [08:00] yeah it won't be in fdisk -l anyway if not in /dev or /proc/partitions [08:01] Disk /dev/vdc doen't contain a valid partition table [08:01] i'm not familiar with euca, but maybe try a xen virtual device such as /dev/xvdd [08:01] davinci11: sweet it worked [08:01] obviously you need to partition and format it to use it [08:01] think of if like a blank hard disk you just bought from the store [08:02] why i put /dev/sdb(exercise) it never show up ?? [08:02] well xen doesn't use /dev/sd* [08:02] so if you are not using a kernel on the instance that has it patched to support /dev/sd* it won't work [08:02] though i don't know how UEC/eucalyptus works in this respect with ubuntu EMIs etc. [08:08] TeTeT: $ euca-run-instances -t m1.large -k [08:08] Note: that the EMI used here cannot be the loader EMI provided in step 2. [08:11] i still don't know what you are following. [08:11] the loader EMI on step2 is not real EMI right ? i tried to start it and try ssh but it wont work, so now i'm using intance downloaded from store [08:11] smoser: is this on your blog or something? [08:12] wtf is step2 [08:12] davinci11 + flaccid : I think the device id for euca-attach-volume was always only a hint, it could always be a differently named device [08:12] flaccid: it's from an exercise I handed over to him, http://people.canonical.com/~tspindler/UEC/ebs-based-instance.pdf [08:12] oh right i referenced that earlier [08:12] * flaccid goes to read [08:12] flaccid : LOL step2 is one of the exercise [08:13] davinci11: yeah, use one from the store or one from uec-images.ubuntu.com [08:13] davinci11: learn to quote/reference resources so people know wtf you are going on about [08:13] flaccid ur words flaccid [08:13] watch ur words [08:14] * flaccid watches his words [08:14] davinci11: rght so i see no problem with that pdf. [08:14] we can help if you cite the errors from the commands you are using [08:14] didnt u see i point to TeTeT: ? [08:14] i.e. paste all teh commands and their response output on a pastebin otherwise we don't know where you failed [08:15] no [08:15] no if just follow the pdf it wont work [08:15] also note that a $() is a subshell so you won't see the output from whats in it, so its better to run it manually [08:16] i will see if i can find a euca cloud [08:17] TeTeT: can you help him out to try to get the commands run individually so the output is not in a sub shell, hidden. its bad scripting to assume a command in a sub shell will always work [08:17] see the step2 i mentioned above , pdf didnt tell beginner [08:18] and the how to ssh too [08:18] yeah, i didn't make the pdf. i would of done some better error handling here [08:20] hmm i've only got cloudstacks available atm sorry [08:20] so every command that uses $(), run manually [08:20] then if successful, assign it to the variable [08:20] running in sh -x will also help [08:21] cloudstacks free ? may be i should try it later , after done with this UEC :P [08:21] cloudstack and openstack both have free/oss versions [08:22] RightScale also recently released myCloud though we are still in the transition to GA [08:22] great [08:23] if you already have UEC setup, you only need to see why the commands are failing [08:23] for example [08:23] do this [08:23] foo=$(echo bar) [08:23] as you can see you won't see that echo because its in a sub shell. [08:24] its also a good idea to learn sh/bash basics before using it. [08:24] so what should i type [08:24] echo bar ? [08:26] arggh [08:26] must be a language barrier here, not sure if i can help you sorry [08:26] TeTeT: you wanna give it a try? this guy is always hard to help [08:29] flaccid + davinci11 : was afk. where are we at now? [08:33] see scrollback [08:40] test [08:45] TeTeT: pls look pdf step 11 $ snapshot=$( euca-create-snapshot $vol | cut -f2 ) [08:45] davinci11: whats the output of euca-describe-snapshots? [08:45] TeTeT this step should create snapshot right ? [08:46] davinci11: yes [08:46] davinci11: and the snap id stored in snapshot variable [08:46] but in step 10 alr terminate instance so ..... [08:46] good luck TeTeT [08:46] TeTeT : can still create snapshot ? [08:47] davinci11: did you create the volume the way it was meant to be? E.g. copied everything over? [08:47] yes [08:48] davinci11: so, is there a snapshot? [08:48] but in step10 it ask to detach volume [08:49] no snapshot [08:50] : isnt it should snapshot while volume attached ? [08:50] davinci11: nope, you detach the volume and then create a snapshot [08:50] oh [08:51] instance need to terminate too ? [08:51] davinci11: doesn't really matter, but we don't need the utility instance anymore once the volume has been populated [08:53] euca-describe-snapshots [08:53] doesnt show anything [08:54] : cant see in hybridfox too [08:54] davinci11: what's the content of $vol? [08:55] how to check ? [08:56] echo $vol [08:56] nothing [08:57] davinci11: so what's the output of euca-describe-volumes? [08:59] Volume vol-5A4C0637 1 cluster1 [09:02] well echo "$vol" good idea to quote :) [09:03] davinci11: vol="vol-5A4C0637" [09:03] davinci11: and then try the snapshot creation again [09:04] flaccid: yeah, I usually don't specify the quotes in the exercises as any additional character leads to a chance a student mistypes it [09:06] yeah though if the returns chars that need escaping, you need it e.g. [ or ] [09:07] greybot: "USE MORE QUOTES!" They are vital. Also, learn the difference between ' and " and `. See and . [09:11] TeTeT: how long need to take to create snapshot, because my PC's led keep blinking i'm not sure snapshot still in progress [09:12] davinci11: check with euca-describe-snapshots [09:12] nothing [09:13] davinci11: weird, the command should return soon, though the snapshot creation can take some time [09:37] TeTeT: i know what is going wrong [09:38] TeTeT: $ snapshot=$( euca-create-snapshot $vol | cut -f2 ) [09:38] $vol should replace my own volid [09:39] now i can how much % [09:39] see [09:44] davinci11: ok, great to see progress === daker_ is now known as daker === dendro-afk is now known as dendrobates === RoAkSoAx_ is now known as RoAkSoAx === daker is now known as daker_ === dendrobates is now known as dendro-afk [17:40] kim0: === dendro-afk is now known as dendrobates [17:50] SpamapS, we should have redirected stderr to /dev/null on 'mkdir /var/run/network/static-network-up-emitted' [17:51] smoser: DOH I thought we did [17:52] smoser: I have another small change I need to push up. I assume it can wait until after A3 [17:52] you had it in yoru hunk but when we joined to the if we used mine which did not have it. [17:52] i think so [17:53] actually, yeah, it will never get seen [17:53] Probably only annoying messages on ifup at the cmdline [17:53] right? [17:53] as upstart output does not go anywhere [17:53] it goes where good log messages go to die [17:53] oh yeah, on ifup on the command line it would i htin === dendrobates is now known as dendro-afk === robbiew is now known as robbiew-afk