[00:00:02] <kdavy_> Alasdairrr: i didnt know xenserver supported ovf
[00:00:10] <Alasdairrr> kdavy_: http://support.citrix.com/article/CTX124660
[00:00:20] <Alasdairrr> XenConvert is the utility that gave me the ovf to begin with :/
[00:00:25] <kreign> kdavy, I've got a vendor support contract... probably about 20 small bug tickets open with them that they're not fixing.
[00:00:36] <kdavy_> kreign: that's why i placed an intentional bottleneck at the fabric level :) that and i'm cheap
[00:00:39] <kreign> kdavy, you can inport/convert ovf into xs
[00:00:55] <kreign> kdavy, we think alike.
[00:01:06] <kreign> kdavy, I use/have used virtualbox on my workstation to build VMs for XS
[00:01:22] <kreign> a couple more steps but it's quicker for me.
[00:02:17] <Alasdairrr> I'm trying to P2V a Windows Server 2008 box using XenConvert, and it's proving quite difficult
[00:04:13] <kdavy_> Alasdairrr: can't you just do a fresh install? often that's easier than screwing with P2V
[00:05:11] <Alasdairrr> kdavy_: It's a customer box, unfortunately
[00:05:44] <kreign> Alasdairrr, I hate that tool.
[00:06:05] <kdavy_> Alasdairrr: are you at least using XenConvert 2.3?
[00:06:06] <kreign> Alasdairrr, basic trick: uninstall all the drivers on the windows host first.
[00:07:21] <Alasdairrr> kreign: Well right now I'm not even able to get XenCenter or "xe vm-import" to import anything I produce via xenconvert
[00:07:44] <Alasdairrr> And unfortunately I'm not in a position to do anything remotely disruptive to the windows host
[00:08:43] <kreign> fun
[00:08:53] <Alasdairrr> yeah
[00:08:58] <Alasdairrr> lots of fun :)
[00:09:02] <kreign> Alasdairrr, the windows guys get stuck with that job, though i got them to use platespin (which appears the only decent tool that works)
[00:09:09] <kreign> Alasdairrr, i can relate.
[00:10:26] <Alasdairrr> *nods*
[00:10:59] <kdavy_> kreign: lol, while we're on the topic
[00:11:23] <Alasdairrr> If I produce a VHD from XenConvert, will it be bootable?
[00:11:34] <Alasdairrr> Or does it need to be an OVF?
[00:11:47] <kdavy_> how would you clone an FC boot LUN from one SAN to another?
[00:12:04] *** rev78 has quit IRC
[00:12:10] <kdavy_> considering you can't do it at SAN level
[00:14:27] <kreign> kdavy, have two FC cards?
[00:14:44] <kreign> kdavy, I don't know much about FC but I suspect you could dd the damn thing from one host to another
[00:14:49] <kreign> eg. boot it on a machine w/ two FC cards
[00:14:57] <kreign> add the 2nd LUN
[00:15:00] <kreign> dd it over
[00:15:17] <kreign> (sorry, very little FC experience, but that would at least work, if not elegantly)
[00:22:52] <kreign> kdavy, hey, you're an EE by trade, right?
[00:23:02] <kreign> er schooling
[00:24:49] *** archstanton43 has joined #Citrix
[00:33:18] <kdavy_> kreign: yeah, but dd requires a host os that supports FC
[00:33:32] <kdavy_> Physics major
[00:33:35] <kreign> ah
[00:33:39] <kreign> kdavy, how about a linux livecd?
[00:33:50] <kreign> or a liveUSB
[00:33:56] <kreign> easy enough to dish up
[00:34:06] <kreign> you'd need some sort of host, at any rate
[00:34:08] <kdavy_> kreign: i havent found one with native FC support :-/ maybe i just wasnt looking
[00:34:10] <kreign> whether virt or otherwise
[00:34:21] <kreign> kdavy, that may be possible. I'm not familiar w/ FC support, tbh.
[00:34:24] <kreign> kdavy, here's what ya do
[00:34:28] <kreign> kdavy, 8Gb USB stick
[00:34:38] <kreign> install -whatever- on it
[00:34:50] <kreign> debian, cent, fucking ubuntu
[00:34:58] <kreign> then boot it in XS
[00:35:12] <kreign> or just make a new 'utility VM' for FC
[00:36:46] <kdavy_> kreign: not an option - servers are 200 miles away
[00:37:00] <kreign> kdavy, a small VM isn't a possibility?
[00:37:12] <kreign> what do you have connecting them currently?
[00:37:30] <kdavy_> fibre
[00:38:05] <kdavy_> but yeah that was just a theoretical question - i ended up mapping both LUNs to a windows box and used a 15 day trial of Acronis disk clone for home users
[00:38:19] <kdavy_> it's just a pain
[00:39:56] <kreign> kdavy, that'd be the basic gist of it, yeah
[00:40:10] <kreign> kdavy, rawrite for windows might do the trick, too.
[00:43:59] <kdavy_> nope, rawrite couldnt do it and neither could the windows dd clone
[00:44:20] <kdavy_> i needed a full disk sector by sector copy, not just a partition clone
[00:45:07] <kreign> gotcha
[00:45:27] <kreign> guess the 'dd clone' couldn't address the whole LUN?
[00:48:17] <Alasdairrr> anyone here played with blktap?
[00:50:15] <kreign> kdavy, sorry, i don't have an answer for you that doesn't suck, but I'm sure one exists. :)
[00:57:46] <Alasdairrr> great, so the xenconvert process has actually locked up all of the xenserver instances
[00:57:52] <Alasdairrr> thats just great
[00:59:25] <kreign> Alasdairrr, ... all of them? meaning in a pool?
[00:59:43] <Alasdairrr> There are hung blktap2 instances I can't remove
[00:59:51] <Alasdairrr> and the SR can't be destroyed/forgotten
[01:00:03] <Alasdairrr> i'm going to have to reboot all the members of the cluster
[01:00:42] <Alasdairrr> i'm going to have to set up an isolated xenserver instance for doing this kind of stuff
[01:00:56] <kreign> Alasdairrr, that's terrifying
[01:01:09] <Alasdairrr> it is, somehow the import failing left the system in an inconsistent state
[01:01:21] <kreign> urg
[01:01:25] <kreign> see, this is why I dislike XS
[01:01:37] <kreign> little "outside case" bullshit breaking things
[01:01:42] <kreign> not really outside the usecase
[01:01:51] <kreign> just not matching the 1-2 design specifications.
[01:02:15] <Alasdairrr> It's very badly written
[01:02:25] <Alasdairrr> and not terribly well tested
[01:02:34] <Alasdairrr> i'm not surprised they give it away for free :/
[01:02:51] <kreign> Alasdairrr, talk to the people who pay a shitload for it (kdavy)
[01:03:27] <Alasdairrr> kdavy_: Is it less buggy if you pay for it?
[01:03:57] <kdavy_> Alasdairrr: no, it's more buggy
[01:04:06] <kdavy_> paid features introduce more bugs
[01:04:13] <Alasdairrr> Hurrah
[01:04:36] <Alasdairrr> Well I'm glad I'm not the only one having these difficulties
[01:04:47] <kdavy_> :D
[01:05:09] <Alasdairrr> We have two platforms, Citrix XenServer and Solaris Zones
[01:05:28] <Alasdairrr> With Citrix XenServer, our customers are happy because Linux and Windows are "easy to use"
[01:05:33] <Alasdairrr> However Citrix is a nightmare to work with
[01:05:47] <kdavy_> indeed
[01:05:53] <Alasdairrr> Solaris Zones are lovely to work with, but customers whinge because the userland is less friendly
[01:05:55] <kreign> Alasdairrr, so why not use esxi?
[01:06:06] <kdavy_> and it can corrupt itself to the point of being completely unrecoverable too :-P
[01:06:14] <kreign> Alasdairrr, what about brandZ w/ virtualbox?
[01:06:26] <Alasdairrr> kreign: Not fast enough
[01:06:34] <kreign> Alasdairrr, due to virtualbox or?
[01:06:40] <kdavy_> Alasdairrr: what about HyperV? :-P
[01:06:50] <kreign> Alasdairrr, have you tested vbox 4 yet?
[01:06:54] <kreign> it's a marked improvement
[01:07:16] <Alasdairrr> kreign: Really? I kind of just view Virtualbox as a desktop product
[01:07:21] <Alasdairrr> kdavy_: Ha :P
[01:07:45] <Alasdairrr> XenServer is free, easy to use, looks okay - I just need to abandon any hope of doing anything remotely complicated with it
[01:07:51] <Alasdairrr> like P2V
[01:07:57] <kdavy_> pretty much
[01:08:57] <Alasdairrr> This latest incident is the worst situation i've gotten myself into having used xenserver for 3 years
[01:10:55] <proprietarysucks> what's your issue
[01:12:02] <Alasdairrr> proprietarysucks: a failed xenconvert import has left a tapdisk2 thingey-me-bob stuck/hung that can't be removed
[01:12:08] <Alasdairrr> which prevents the SR from being unmounted
[01:12:14] <Alasdairrr> so I have an SR sitting there I can't delete/remove
[01:12:39] <archstanton43> anyone here using XenDesktop 5?
[01:12:54] <archstanton43> Ive just started a POC its looking good
[01:12:55] <proprietarysucks> did you unplug the vdi?
[01:13:34] <kdavy_> archstanton43: POC's always look good - Citrix makes sure of that because otherwise they wouldn't sell anything
[01:13:45] <proprietarysucks> vbd*
[01:14:13] <archstanton43> kdavy_ yeah are you using it production or anyone else here?
[01:14:22] <kreign> kdavy, lol
[01:14:25] <kdavy_> archstanton43: nope, i only use XenApp
[01:14:40] <kdavy_> JarianGibson is working with XD5 i think
[01:14:42] <kreign> archstanton43, for a POC/demo or anything like that i insist on sitting at the wheel.
[01:14:44] <kreign> for at least a day.
[01:14:49] * kreign is great at breaking shit
[01:14:59] <kdavy_> maybe also makson - not sure if he's doing 4 or 5
[01:15:18] <kdavy_> kreign: that's another thing we have in common. i break everything
[01:15:59] <archstanton43> id love to work with someone whos going though creating a design or about 400o users
[01:16:04] <archstanton43> 4k users
[01:16:40] <archstanton43> JarianGibson are you around?
[01:18:33] <Alasdairrr> great, this livemigrate has failed too
[01:18:55] <archstanton43> its all fair enough doing a small POC but it doesnt really show you anything as I need to get it working on a netapp with all the flex clones ability and through an access gateway with 2 factor auth
[01:19:54] <archstanton43> kdavy_ do you know if the xendesktop 5 express is only 10 users for 30 days or is it only 10 users after 30 days i couldnt tell
[01:20:20] <archstanton43> kdavy_ it seemed to suggest it was plat for 30 days then allowed 10 after
[01:20:57] <archstanton43> kdavy_ perfect if i can test for longer as i keep having to leave it and come back to it
[01:20:58] <kdavy_> archstanton43: xendesktop express is 10 users free
[01:21:24] <archstanton43> kdavy_ persistent? excellent thanks
[01:21:24] <kdavy_> archstanton43: i'd say start your design with the storage - that's the most important component
[01:21:32] <kdavy_> yep
[01:21:57] <archstanton43> kdavy_ yeah i take it no ones designed a local solution with xd5 with ha
[01:22:11] <kdavy_> nope
[01:22:34] <archstanton43> kdavy_ not a problem was going to use our nas anyhow
[01:22:58] <kdavy_> what nas do you have?
[01:23:20] <archstanton43> kdavy_ IBM N-Series 6040A20 its a netapp rebadged
[01:23:25] <archstanton43> using NFS
[01:23:55] <kdavy_> how many spindles?
[01:24:07] <archstanton43> kdavy_ already using it for large virtual server deployment but we have a spare
[01:24:22] <archstanton43> not enough lol only 48
[01:24:33] <kdavy_> yeah, you'll need a lot more
[01:24:57] <archstanton43> i wish there was an io card which could accelerate read and write on netapp
[01:25:00] <kdavy_> or solid state
[01:25:23] <archstanton43> ive heard o PAMII flex cache but its only read really
[01:25:36] <kdavy_> archstanton43: as a side note take a look at Atlantis Computing - they make I/O acceleration appliances that sit between your NAS and the server farm
[01:25:39] <archstanton43> im basing a shelf on 250 users seem about right?
[01:26:20] <archstanton43> apparently EMC have some solutions also
[01:26:23] <kdavy_> if a shelf is 16 drives @ 150iops, that's roughly 2k iops - you should be fine with 200-250 users
[01:26:50] <kdavy_> i have 8 shelves for 550 users with solid state, but it's way overkill right now
[01:27:26] <archstanton43> kdavy_ that does seem overkill and also how much was that!!!
[01:27:43] <archstanton43> kdavy_ what you using
[01:28:02] <archstanton43> kdavy_ certainly will look at atlantis
[01:30:24] <archstanton43> kdavy_ thing is no-one wants to be waiting longer for their Virtual desktop eh
[01:31:36] <kdavy_> archstanton43: i'm using Nexenta Enterprise - built the SAN myself
[01:31:47] <kdavy_> total about $60-70k software+hardware combined
[01:31:50] <Alasdairrr> kdavy_: What SSDs are you using?
[01:32:09] <kdavy_> a mix of Intel X25M and OWC Mercury Extreme RE
[01:32:36] <Alasdairrr> We're using LSI JBOD arrays, Seagate Constellation 2TB SAS disks plus ZeusRAM for the ZIL and OCZ Vertex2E for L2ARC
[01:33:10] <kdavy_> spindles are all 10k, a mix of Seagate Cheetah 3.5" and Seagate Savvio 10k.4 2.5"
[01:33:27] <kdavy_> Alasdairrr: you're on Nexenta too?
[01:33:29] <Alasdairrr> X25M don't have a supercapacitor though so they lie about fsync()s, in a power loss scenario you might have an issue :/
[01:33:43] <kdavy_> Alasdairrr: i only use them for L2ARC
[01:33:53] <Alasdairrr> ahh okay cool
[01:34:00] <Alasdairrr> kdavy_: I'm project lead for OpenIndiana, so we're actually using a custom homebrew kernel
[01:34:18] <kdavy_> don't have anything for ZIL yet since spindles can handle way more than what's currently on them
[01:34:30] <kdavy_> Alasdairrr: ah, nice, good to know
[01:34:48] <Alasdairrr> But there are people with 300TB arrays running oi_148 without issue
[01:35:00] <Alasdairrr> (Or so RoyK keeps boasting on #openindiana)
[01:35:02] <Alasdairrr> :)
[01:35:03] <archstanton43> kdavy_ great interesting stuff
[01:35:20] <kdavy_> Alasdairrr: i'm sure - nexenta enterprise also has a few multi-PB deployments
[01:35:40] <Alasdairrr> It's quite impressive
[01:35:51] <Alasdairrr> My dayjob (managed hosting) only has 30TB under storage atm
[01:35:58] <kdavy_> i couldn't use vanilla OI/OS since i need the commercial plugins - FC Target and HA Cluster
[01:36:04] <Alasdairrr> *nods*
[01:36:18] <archstanton43> kdavy_ nexenta is that like a better openfiler
[01:36:47] <kdavy_> archstanton43: nexenta is way ahead of openfiler. its beauty is in the ZFS filesystem
[01:37:17] <Alasdairrr> we wrote a homebrew cluster solution for the NFS, took 2 months of developer time - it's not nearly as advanced as you'd get from nexenta but with hosting, nexenta is too expensive
[01:37:57] <kdavy_> Alasdairrr: i dont think its too expensive... at least not compared to an enterprise san
[01:37:58] <archstanton43> kdavy_ yep how easy to use out of box
[01:38:18] <kdavy_> my 16Tb HA Cluster with FC target only cost $15k in licensing
[01:38:20] <Alasdairrr> kdavy_: Oh sure, but we'd never be able to go down the enterprise SAN route, it'd price us out of the market
[01:38:41] <kdavy_> archstanton43: very easy out of the box, but harder if you get into the more advanced stuff
[01:38:57] <archstanton43> kdavy_ impressive how about a snapmirror kind of setup?
[01:39:01] <kdavy_> i have yet to encounter something i couldn't easily do from the web gui
[01:39:12] <kdavy_> archstanton43: snapmirror - what's that?
[01:39:22] <archstanton43> kdavy_ can you set up any kind of mirroring or cluster ha?
[01:39:33] <kdavy_> archstanton43: yes to both
[01:39:54] <archstanton43> kdavy_ easily through the web interace because thats where openiler ell flat
[01:40:08] <archstanton43> sorry my f key is a bit bust
[01:40:14] <kdavy_> asynchronous mirroring between units comes with the basic license, synchronous replication is a bit extra, HA cluster is also a commercial plugin
[01:40:44] <Alasdairrr> kdavy_: Any idea how it implements synchronous replication under the hood?
[01:40:57] <kdavy_> i have the HA Cluster setup (although i'm starting to realize i'm probably one of the few people who bought it)
[01:41:01] <archstanton43> kdavy_ i wouldnt use synchronus anyway
[01:41:37] <kdavy_> archstanton43: not sure exactly about synchronous replication, but it's transactional
[01:41:53] <archstanton43> kdavy_ all i care about is a good aync mirror dedupe and some sort of cloning ability
[01:42:11] <kdavy_> archstanton43: check, check and check
[01:42:18] <archstanton43> as in volume cloning im liking it
[01:42:22] <kdavy_> although there are catchas with dedupe
[01:42:34] <archstanton43> kdavy_ always is volume size?
[01:42:38] <Alasdairrr> deduping large datasets is very costly in terms of performance
[01:42:46] <archstanton43> processor intensive yep
[01:42:58] <archstanton43> will be on anything
[01:43:15] <Alasdairrr> but there are issues beyond just that - I've heard reports of deleting deduped datasets taking weeks
[01:43:23] <archstanton43> kdavy_ im limited to 2TB volumes on netapp
[01:43:24] <kdavy_> archstanton43: no, deleting a deduped data set brings the system down to its knees and causes FC/iScsi initiator logouts + port resets
[01:44:06] <archstanton43> kdavy_ ooh nasty
[01:44:13] <kdavy_> Alasdairrr: yep, without L2ARC it might take weeks; with good L2ARC - hours to days depending on size
[01:44:30] <kdavy_> archstanton43: there is no limit on volume size
[01:44:31] <archstanton43> kdavy_ just dont delete make it all the same instead lol
[01:45:13] <kdavy_> archstanton43: yeah, but most people learn this the hard way :-P
[01:45:50] <kdavy_> Nexenta 3.1 is supposed to have that bug fixed, i'm working with them to get my hands on an early alpha
[01:46:06] <archstanton43> kdavy_ it sounds great anyway
[01:46:10] <kdavy_> so i can do my own testing
[01:46:58] <kdavy_> it's great - performance is unbelievable to my old SAN (which cost $250k 3 years ago and would still cost $150k new)
[01:47:23] <kdavy_> *compared to my old san
[01:47:31] <archstanton43> kdavy_ yeah that about the cost of ours
[01:47:34] <Alasdairrr> On a different topic - have any of you used platespin?
[01:47:48] <archstanton43> kdavy_ i keep hearing about it
[01:47:53] <kdavy_> i'm actually thinking about connecting the old SAN to the Nexenta and doing passthrough
[01:48:21] <archstanton43> Alasdairrr sorry i keep hearing platespin but havent started looking at it yet
[01:48:59] <archstanton43> kdavy_ yeah got to put the old san to good use
[01:49:28] <Alasdairrr> I'm going to trial platespin, see if it can solve this P2V issue
[01:49:36] <kdavy_> archstanton43: right now the idea is to put stuff with high disk space requirements on the old san, and all the high-IO stuff on the new one
[01:49:37] <archstanton43> kdavy_ its like leaving an astin martin to rust
[01:50:00] <kdavy_> archstanton43: no, it's more like driving to work on a semi every day
[01:50:31] <archstanton43> kdavy_ that sounds ideal and exactly what i will probably end up doing
[01:51:04] <kdavy_> mm, well i'm usually around both here and in #nexenta, #nexentastor and #opensolaris - ping me if you have questions
[01:51:04] <archstanton43> kdavy_ ive got 2 projects which are going to need massive io but i have lost of space but not much io at present
[01:52:14] <archstanton43> kdavy_ all it shops in same boat really just hope these SSD's fall faster
[01:52:20] <archstanton43> in $
[01:52:36] <archstanton43> kdavy_ thanks i might do that
[01:52:41] <kdavy_> Sandforce 2400 series ssds should be on the shelves soon, as well as the Intel 510. then i can toss the X25M's
[01:53:04] <Alasdairrr> OCZ looked like they were going to ship SAS based OCX Vertex2 Pro drives
[01:53:10] <Alasdairrr> but then they disappeared from their website
[01:53:23] <Alasdairrr> and now the only SAS products OCZ list are on their OEM website
[01:53:28] <kdavy_> and by "toss" i mean do a 12-drive RAID10 in my work computer :)
[01:54:16] <kdavy_> yeah, i wish there were reasonably priced SAS ssd's... i have to use the stupid interposers now
[01:54:41] <Alasdairrr> What disk shelves are you using and which interposers?
[01:55:02] <kdavy_> Alasdair, LSI 620J and LSI 9252 interposers
[01:55:23] <Alasdairrr> ah yes, we're using their 3.5" shelves
[01:55:36] <Alasdairrr> Are the LSI 9252 interposers any good?
[01:55:38] <kdavy_> shelves are pricey but no headaches whatsoever
[01:55:55] <kdavy_> Alasdairrr: they work, never had any issues with them. that's all i can say
[01:55:56] <Alasdairrr> I have no idea where to get the interposers from, my supplier failed to find them
[01:56:06] <Alasdairrr> my usual supplier, I should say
[01:56:13] <kdavy_> i got mine from either serversupply or provantage... forgot
[01:56:53] <kdavy_> actually no, i got them from wiredzone.com - that's my usual supplier for the supermicro gear as well
[01:57:06] <Alasdairrr> perhaps i'll get them from there too
[01:58:55] <Alasdairrr> anyway, 1am
[01:58:57] <Alasdairrr> bed :)
[01:59:05] <archstanton43> Alasdairrr yeah same
[01:59:05] <Alasdairrr> thanks all for the help earlier, night!
[01:59:25] <archstanton43> kdavy_ yeah its been really usefull
[01:59:31] *** Alasdairrr is now known as AlasAway
[01:59:52] <archstanton43> AlasAway see ya
[02:00:19] <archstanton43> he must be in uk
[02:01:23] <archstanton43> i need to find a wiredzone.com but in the uk
[02:05:01] <archstanton43> anyone tried one of these http://www.wiredzone.com/itemdesc.asp/ic/32024190/HP/600281-B21
[02:07:00] <archstanton43> i know capacity is crap but if you could dedupe and thin prov on it and only keep non presistent desktops it apprently gives you 100,000 iops
[02:07:59] <archstanton43> thats about 5k desktops it cant be
[02:12:14] <kdavy_> archstanton43: drivers will be your main issue
[02:12:47] <kdavy_> and it's still way overpriced for what it is
[02:13:18] <kdavy_> and there are reliability concerns even beyond raid0
[02:15:09] <kdavy_> and in VDI, writes are your main concern, not reads, and you can't really dedup shit like pagefiles and all the other random write workloads
[02:24:57] <kdavy_> mmm reading about the new OCZ Vertex 3 Pro - supercap, 70,000 random 4k write IOPs, under $4/GB, SATA 6Gbps
[02:26:56] <kdavy_> and native compression in the Sandforce controller. cool
[02:45:07] <archstanton43> kdavy_ that sounds ace
[03:11:05] *** archstanton43 has quit IRC
[03:20:01] <Rienzilla> is this a 320G storage for 18k$?
[03:26:02] *** MSilva01 has quit IRC
[03:28:54] *** ScottCochran has joined #Citrix
[03:29:18] <ScottCochran> hello al
[03:29:19] <ScottCochran> all
[03:30:12] *** MSilva01 has joined #Citrix
[03:46:13] *** lesrar has joined #Citrix
[03:49:15] *** waynerr__ has quit IRC
[04:41:48] *** Zed` has quit IRC
[04:47:26] *** The_Butcher has joined #Citrix
[04:52:31] *** Zed` has joined #Citrix
[05:00:37] *** The_Butcher has quit IRC
[05:18:45] <kdavy> ScottCochran: what up?
[05:19:00] <kdavy> Rienzilla: yeah, 320G for 18k$, crazy
[05:19:41] <ScottCochran> kdavy not much, you?
[05:19:49] <kdavy> chillin'
[05:20:01] <kdavy> recovering VM templates from a failed SR
[05:20:33] <ScottCochran> outch... what kinda SR
[05:20:50] <kdavy> xenserver, what else is there?
[05:21:00] <kdavy> i.e. what else is there that is doomed to fail
[05:21:34] <ScottCochran> lol I mean what kinda SR?  iSCSI, NFS, SLG...
[05:21:46] <kdavy> Fibre Channel
[05:22:13] <ScottCochran> ahh, old school.  I have never used Fiber with Xen
[05:22:23] <kdavy> i don't even want to touch SLG - too unreliable from what i've heard
[05:22:52] <kdavy> fiber actually works really well with xen, much less of a pain to deal with than iScsi in my opinion
[05:23:12] <ScottCochran> I am using it in my current project with XenServer 5.6 FP1... Working good for me
[05:23:33] <kdavy> you don't have to worry about any NIC business, separate vlans for storage and network, management interfaces etc
[05:24:45] <kdavy> and i have 8G FC adapters in every blade, so i can easily quadruple my throughput by upgrading the fabric (2G FC core at the moment)
[05:24:55] <ScottCochran> yea but you need separate fiber switches, HBAs in servers, HBA's in SAN, and fiber expertise
[05:25:10] <ScottCochran> a lot more money to scale with fiber
[05:25:18] <kdavy> fiber is easier than network - less layers in the OSI mode
[05:25:19] <kdavy> model
[05:25:27] <kdavy> and 8G fiver is cheaper than 10GbE
[05:25:37] <kdavy> *fiber. what's up with those typos?
[05:26:51] <kdavy> fiber switches are cheap too if you go for previous generation models. 20 port 2Gb FC switches are $600-$800, 8-port 4Gb switches can be found for about $800-1000
[05:27:22] <kdavy> no need to have the latest and the greatest
[05:27:43] <ScottCochran> yea but you NEED ethernet, you do not need fiber.  Why run two networks?  IMO fiber will be dead in the next 5 years, maybe less once 10Gb, 20Gb, and even 100Gbs ethernet drop in price
[05:28:02] <kdavy> even 10Gb ethernet essentially runs over fiber
[05:28:13] <kdavy> and there is IP over FC, if you insist :)
[05:28:19] <kdavy> i just don't use it
[05:28:42] <ScottCochran> Yea but only to be used for people to slowly migrate off fiber
[05:29:48] <kdavy> anyway, what's the point of arguing? both points of view make sense
[05:29:59] <ScottCochran> I guess I see it different as a consultant and going in to large companie and feds.  They always buy new and fiber just cant cut it
[05:30:15] <kdavy> i use blade servers so fiber is A LOT cheaper than upgrading to 10GbE
[05:30:29] <kdavy> IBM blade servers, even
[05:31:11] <ScottCochran> Yea we see a lot more HP and UCS, which are using 10Gb with ther network virtualization technologies
[05:33:32] <kdavy> CNA's, right?
[05:34:50] <kdavy> i keep track of the latest trends, i just see no point in using them yet. if i needed more throughput/better latency, I'd go with DDR Infiniband to begin with :)
[05:43:23] <gladier> CNAs are cool
[05:44:01] <gladier> was talking with one of the networking guys ... he had a few cisco nexus 7000's he was playing with
[05:45:02] <kdavy> gladier: how much are those again?
[05:46:03] <kdavy> probably $100k for a decently populated chassis
[05:46:28] <gladier> something nuts like that
[05:47:31] <kdavy> in my opinion a 6500 is still fully capable of datacenter-level tasks
[05:47:45] <kdavy> that's why i have a 6506 and a 6509
[05:47:56] <kdavy> and FC for storage :)
[05:49:27] <kdavy> hahahahaha look at this... found by accident - http://www.neobits.com/cisco_mem_xcef720_1gb_cisco_1gb_ddr_sdram_memory_module_1gb_1_x_1gb_ddr_sdram_p218972.html?atc=gbs
[05:49:36] <kdavy> AWESOME DEAL 30% OFF!!!
[05:51:07] <kdavy> and http://www.neobits.com/cisco_mem_prp2_4g_cisco_4gb_sdram_memory_module_p256658.html
[05:51:23] <kdavy> deal so good it's already out of stock
[06:47:52] <ScottCochran> kdavy You use WLB in XenServer?
[06:48:22] <kdavy> ScottCochran: yes
[06:49:09] <kdavy> i don't use the automated portion though - just look through and manually apply recommendations from time to time, my workload is fairly static
[06:49:57] <ScottCochran> I have had it set to optimize performance at the highest level automation and has not moved one VM or made a recommendation
[06:49:59] <kdavy> my favorite part about it is the "Optimal Server" and the star ratings of optimal servers when you live-migrate - perfect for maintenance
[06:50:09] <ScottCochran> been almost 3 weeeks
[06:50:17] <kdavy> ScottCochran: what is your highest scoring metric - CPU?
[06:50:19] <ScottCochran> yea it does the optimal server
[06:51:17] <ScottCochran> I guess, the metric is like 90 percent.  Does it only use the highest metric?
[06:51:36] <kdavy> no, it uses all of them, but the algorithm is strange
[06:52:03] <kdavy> set CPU metric to like 50% for it to start making recommendations - it is very conservative
[06:54:01] <kdavy> i have my thresholds set to 50% CPU, 500MB of free memory, 52 MB/s of network read/write and 100MB/s of disk read/write
[06:54:22] <kdavy> but of course it depends on your environment
[06:54:49] <kdavy> i usually do half of maximum
[06:54:58] <kdavy> that will balance everything evenly
[06:55:19] <ScottCochran> cool... Yea VMware DRS is more intuitive
[06:55:24] <kdavy> i have yet to see it make a recommendation based on anything but CPU/RAM
[06:55:57] <kdavy> havent messed with the paid vmware editions - i only use ESXi in 1+1 groups with shared storage
[06:56:28] <kdavy> ghetto-HA :)
[06:57:01] <ScottCochran> 1+1? You cant even VMotion with the free can you?
[06:57:47] *** ScottCochran_ has joined #Citrix
[06:59:06] <kdavy> ScottCochran: nope, i use MS failover clustering or NLB (depending on use case) between the guest VMs
[06:59:11] *** ScottCochran_ has quit IRC
[07:00:41] <ScottCochran> lol, nice.  Does that require Server Enterprise licensing, in which case 2 ESX licenses would be the same $ or cheaper
[07:00:46] <kdavy> i really only need nines of uptime during business hours - after 9pm almost no one cares if something is down for a few minutes
[07:01:17] <kdavy> ScottCochran: yes it requires server enterprise licensing, but SPLA makes it easy since you can license on a per-seat basis instead of per-socket
[07:01:31] <kdavy> or per user, whatever
[07:02:16] <ScottCochran> ahhh SPLA... how doe use license MS Clustering per user?
[07:02:43] <kdavy> MS clustering is not licensed per user. you license applications that run on those servers
[07:03:12] <kdavy> for example SQL Standard (doesnt support failover clustering, cheaper per-seat) vs. SQL Enterprise (supports clustering, more expensive)
[07:03:24] <kdavy> same with Exchange Std. vs Enterprise, and so on
[07:04:15] <ScottCochran> oh so your saying you need application level fault tolerance for some, not just server HA like in VMware or Xen
[07:04:19] <kdavy> the beauty of it is, MS doesn't care how many actual instances of Windows Server you use to accomplish the task
[07:04:31] <kdavy> yep, i do
[07:05:12] <ScottCochran> they dont care about the # of Windows Servers used?  Hows that?
[07:05:15] *** ScottCochran has left #Citrix
[07:05:27] <kdavy> server HA i really only need for things like XenApp/XenDesktop farm, and the odd server-side crap that is legacy or doesnt support any kind of clustering
[07:05:37] *** ScottCochran_ has joined #Citrix
[07:06:06] <kdavy> ScottCochran_: in my case the actual Windows Server licensing for back-end apps is included in the SPLA user CALs
[07:06:59] <kdavy> so for a full hosted XenApp environment you'll need a Windows Server CAL + MS RDS CAL (+ Exchange CAL + SQL CAL and so on depending on the mix of users)
[07:07:54] <ScottCochran_> Oh really?  Nice!  So you work for a hosting company?
[07:08:15] <kdavy> ScottCochran_: yep, full-service hosting and IT outsourcing
[07:09:02] <kdavy> we put in-house IT departments on the street, essentially :)
[07:09:08] <ScottCochran_> Sweet, where are you based
[07:09:14] <kdavy> Houston, TX
[07:09:48] <ScottCochran_> Cool
[07:10:05] <kdavy> yep, no indian accents
[07:10:30] <ScottCochran_> You working for a big co. Like IBM ?
[07:11:01] <kdavy> nah, small company - we only support 600-ish seats at the moment. but we have ambitions :-P
[07:12:09] <ScottCochran_> Good, IBM sucks at hosting.  One of our current consultants worked for them and said how bad it was...
[07:12:32] <kdavy> the only thing i respect IBM for is the hardware, i didnt know they even did hosting
[07:13:03] <kdavy> regardless, getting hosting from a fortune 500 corporation is like selling your kidney to a stranger
[07:13:50] <kdavy> they lure you in but you immediately regret it after you've signed the contract
[07:14:06] <ScottCochran_> Yea they actually won the contract to host the entire state of TX data centers.  Screwed it up really bad, made news... Google IBM Hosting State of Texas
[07:14:44] <kdavy> i know, and you know what sucks most... they fucking did it thanks to my company
[07:15:47] <ScottCochran_> Did what?
[07:15:53] <ScottCochran_> Hosting?
[07:16:18] <kdavy> our president originally started talking to the state of TX about datacenter-hosted computing - it was years and years ago, a very new concept at the time. we almost had a deal where we would migrate over one small TX department at a time - no more than 200 seats at a time, since we didn't have enough people for such rapid expansion
[07:16:54] <kdavy> but then some idiot came and decided it's a good idea to move EVERYTHING at once, and we were too small to compete with IBM
[07:17:07] <kdavy> now IBM fucked up and we're talking to the government again
[07:17:41] <ScottCochran_> Yea I just read the article last week about how bad IBM did
[07:19:30] <kdavy> according to some people i spoke to, that article doesn't even come close to describing the full extent of disaster
[07:19:31] <kdavy> lol
[07:19:44] <ScottCochran_> Ouch
[07:21:02] <kdavy> all IBM was interested in was selling the hardware+software, the hosting portion just came as a side-deal since they couldn't find anyone else to do the project
[07:21:32] <kdavy> and, of course, lots and lots of consulting
[07:21:53] <ScottCochran_> Supposedly they are big into hosting IT now, same with EMC
[07:22:50] <kdavy> eh, they can't compete with smaller companies i think
[07:23:55] <kdavy> or at least with competent smaller companies, where qualified decisions are made on the spot instead of going through a chain of management and constant fingerpointing
[07:25:37] <ScottCochran_> I agree but it also dpe
[07:27:17] <ScottCochran_> Depends on the infrastructure required by a client.  Could you provide a 5k VM environment with geographically distributed DR sites? In a reasonable amount of time
[07:27:20] <kdavy> maybe we're talking about different target clientele though - our sweet spot is SMB with 20-200 users, too small for IBM to give a shit about
[07:28:31] <ScottCochran_> Yea
[07:29:35] <kdavy> 5k VM - no, we wouldn't be able to take such a large project at this point in time. but we do have an infrastructure capable of scaling to that range - already have two geographically distributed colos, and the design itself can scale easily if you add more capacity
[07:31:05] <ScottCochran_> Gotcha...
[07:31:50] <kdavy> i'd say for a 5k VDI project, we'd need 3-6 months of implementation and $300k up-front investment in hardware, in addition to hiring like crazy for the support department. that's a rough estimate but the scale should be adequate
[07:33:11] <kdavy> but more importantly, we wouldn't get a project like this because we lack references of similar scale. we'd need to scale out first before getting to this level
[07:35:05] <kdavy> wait, no, the $300k figure isn't right. way way way too optimistic
[07:35:50] <kdavy> $1M is definitely doable though
[07:43:25] *** ScottCochran_ has quit IRC
[07:43:44] <kdavy> regardless, scaling from 50 to 500 is much harder than from 500 to 5000 in my opinion, and who cares about cost - if it makes sense it makes sense
[07:45:46] *** gblfxt has quit IRC
[07:46:13] *** gblfxt has joined #Citrix
[08:37:34] *** MSilva01 has quit IRC
[08:53:43] <censor> ThinkTank: slept over my problem for a night, got into the office this morning, and had a VM running within 20 minutes ... yay! =)
[08:55:04] <censor> my problem was that i a) had to set the Default SR in the other_config and b) i didn't see that there was other XML in that field, so i overwrote the whole 'disks' field with just 'sr=...', including the provisioning XML ... doh!
[08:55:30] <censor> i'll put my code in my wiki, may other people learn from my mistakes ;)
[11:03:44] *** cmarreel has joined #Citrix
[11:42:16] *** AlasAway is now known as Alasdairrr
[12:37:40] *** theobalu has joined #Citrix
[12:40:09] <theobalu> Hi! Does anybody knows how i can set up XenDesktop 5 with host type vmware esx?
[12:45:56] <pak21> http://support.citrix.com/proddocs/index.jsp?topic=/xendesktop-rho/cds-vmware-rho.html
[12:51:42] <theobalu> Thanks pak21! This should solve my problem.
[12:53:12] *** theobalu has quit IRC
[12:59:08] *** sanket__ has joined #Citrix
[12:59:14] <sanket__> hello
[13:00:22] <sanket__> how to install citrix client for ARM9? can anyone help me....
[13:59:52] *** sanket__ has quit IRC
[14:04:16] *** tom_wurm has quit IRC
[15:06:10] *** Zed` has quit IRC
[15:13:06] *** Zed` has joined #Citrix
[15:45:36] *** rev78 has joined #Citrix
[15:51:57] *** deshantm_xenorg has quit IRC
[15:52:41] *** zaf has quit IRC
[15:53:09] *** zaf has joined #Citrix
[15:53:37] *** deshantm_xenorg has joined #Citrix
[15:53:56] *** Zed` has quit IRC
[15:54:18] *** Zed` has joined #Citrix
[16:39:22] <kreign> omg march is not supposed to be this cold
[17:26:07] *** censor has quit IRC
[17:30:16] <kdavy_> kreign: it's warm here
[17:35:12] <kreign> kdavy, yeah, well, it's always warm there. You're in Texas. :)
[17:35:35] <kreign> it's obscenely cold here for march
[17:35:42] <kreign> colder than last month, mostly
[17:35:50] <kreign> 10-day forecast does not look good, either.
[17:39:15] <kdavy_> sucks
[17:39:22] <kdavy_> how cold?
[17:47:29] *** Trixboxer has joined #Citrix
[18:09:00] <kreign> kdavy, -10F I Think?
[18:09:07] <kreign> kdavy, cold enough a diesel barely starts with a block heater.
[18:09:19] <kreign> and the batteries were not happy
[18:09:42] <kreign> and there's wind
[18:09:48] <kreign> like I said, so much for march
[18:14:28] *** cmarreel has quit IRC
[18:18:06] <kreign> kdavy, you wouldn't happen to be willing to rewrite this horrible PHP tool/interface for me, would you?
[18:20:23] <kreign> fucking page takes 30+s to load due to massive db queries on an un-normalized db
[18:20:28] <kreign> select * from foo
[18:21:03] <kreign> and it's done really 'cleverly' too.
[18:26:44] <kdavy_> heh, no thanks
[18:26:59] <kdavy_> is the db backend mysql?
[18:30:34] *** Freeborn has joined #Citrix
[18:36:39] *** Freeborn has quit IRC
[19:04:28] *** OmNomDePlume has joined #Citrix
[19:04:40] *** OmNomDePlume has quit IRC
[19:05:06] *** OmNomDePlume has joined #Citrix
[19:25:26] *** nikade has quit IRC
[19:54:13] *** kreign has quit IRC
[19:56:17] *** kreign has joined #Citrix
[19:57:20] *** MSilva01 has joined #Citrix
[19:57:54] <MSilva01> hi all ... I need some help with PVS ...
[19:58:22] <MSilva01> I dont know why but in the middle of day some vms are losting this machine account
[19:59:01] <MSilva01> when this happens i need to shutdown the vm and reset the computer account
[19:59:16] <makson> MSilva01: i actually saw smae thing the other day
[19:59:24] <makson> do you have machine account password disabled?
[19:59:31] <makson> machine account password disabled?
[19:59:53] <MSilva01> yes ... i do
[20:00:07] <makson> MSilva01: 5.6 SP1 ?
[20:00:19] <MSilva01> 5.6 only
[20:00:20] <makson> i saw the same thing reset computer account and waiting to see if it happens again.
[20:00:27] <makson> MSilva01: i think there is a hot fix for that pre SP1
[20:01:09] <makson> MSilva01: http://support.citrix.com/article/CTX127313
[20:01:12] <makson> check that out.
[20:01:24] <MSilva01> I dont know, I'll look for right now
[20:03:20] <MSilva01> good to hear ... i cant believe that in the middle of day the machine lost the computer account ...
[20:04:02] <MSilva01> makson ... thanks,  i'll let you know if worked
[20:08:09] <MSilva01> makson ... i remembered that PVS Server is 5.6, but target device is 5.6 SP1
[20:11:19] <MSilva01> i saw this issue few times in the past, now a lot of times after brazil daylight changes ...
[20:17:07] <Splatone_> Pfsense released 2.0RC1...!
[20:28:06] <Splatone_> anyone know how to create a single icon on customers desktop that will point back CSG app without installing full online plugin.  IE just a ica file?
[20:37:44] <MSilva01> splatone ... if you create um ica file the ticket will expire and ask for credentials
[20:38:20] <Splatone_> Splatone_: There has to be a way to do this..
[20:38:41] <Splatone_> I know you could do it before with just program neaborhood but I want to hit the CSG..
[20:38:45] *** RidaGee has joined #Citrix
[20:38:59] <Splatone_> I think there is a way to hit the app directly on the wi server.
[20:39:43] <MSilva01> did you try to logon on csg and save the ica file ?
[20:39:49] <Splatone_> I expect the users to put credentials in somewhere.
[20:44:54] <MSilva01> splatone ... i did a test saving the ica file and worked for me ...
[20:45:41] <Splatone_> hmm..
[20:46:01] <Splatone_> and your going through a CSG?
[20:46:08] <Splatone_> so hitting external IP?
[20:46:29] *** blood_ has joined #Citrix
[20:46:49] <blood_> how do I boot an iso with a linux vm in xenserver? Can't seem to get it to work
[20:46:55] <MSilva01> the only thing that changed was the application asking for credentials
[20:50:54] <kreign> blood_, not sure what you mean
[20:51:17] <blood_> we have a CentOS VM in XenServer and trying to boot the CentOS Live iso but are unable to
[20:52:15] <kreign> blood_, 'unable to'
[20:52:19] <kreign> not sure what that means.
[20:52:39] <kreign> these are questions I expect from level 1 tech support, at worst, or users.
[20:52:44] <blood_> We set startup options to "Boot from DVD" in XenServer but then the CentOS VM fails to start
[20:52:48] <kreign> (questions I expect to have to ask)
[20:52:57] <kreign> blood_, getting warmer.
[20:53:07] <kreign> "there are angry lemmings flowing from the monitor"
[20:53:20] <kreign> "I get an error code xyz and a picture of pam anderson, circa 2011"
[20:54:02] <Splatone_> kreign: I think that was a dos app I put on the computers in highschool.
[20:54:03] <blood_> kreign: ok one second, let me get the exact error message when attempting to startup from DVD
[20:55:40] <kreign> blood_, we'll get you to a Tier 1 level of competence yet, I promise.
[20:56:34] *** blood_ has left #Citrix
[20:56:37] <kreign> Splatone_, CGA?
[20:56:42] *** blood_ has joined #Citrix
[20:56:44] <blood_> oops
[20:56:51] <kreign> Splatone_, I had an awesome strip poker game that did full CGA
[20:57:05] <kreign> when that floppy died, I was upset.
[20:58:24] <blood_> kreign: error message is "Warning /dev/xvda is not a disk image. Traceback most recent call last : file /usr/bin/pygrub line 746 in question mark raise runtime error unable to find partition containing kernel"
[20:58:37] <blood_> kreign: That shows up in XenCenter alert
[20:58:45] <kreign> :|
[20:58:55] <blood_> kreign: the VM fails to start when trying to boot from DVD with that error message
[20:59:04] <kreign> DVD or DVD iso?
[20:59:07] <blood_> iso
[20:59:15] <kreign> selected it from the drop-down?
[20:59:19] <blood_> yes
[20:59:32] <kreign> run 'file imagename.iso' from console on it
[20:59:51] <kreign> if that looks correct try: mount -o loop isoimage /mnt
[21:00:03] <kreign> see if that succeeds
[21:00:24] <blood_> kreign: the console doesn't even appear when startup options are set to DVD. Should I set it back to HD then try in console?
[21:00:42] <blood_> kreign: currently, the console is not showing since the VM fails to start
[21:01:15] <kreign> blood_, the XS console not the guest console.
[21:01:21] <blood_> ah ok
[21:01:22] <kreign> the scary looking #
[21:01:26] <blood_> one second
[21:03:49] <blood_> kreign: asked my coworker, he said the iso's are located on network and not local to the XS server so XS is unable to see it from the console.
[21:04:02] <kreign> blood_, that is incorrect.
[21:04:07] <kreign> your coworker is an idiot
[21:04:15] <blood_> kreign: ok so what do I tell him=)
[21:04:22] <kreign> a rhetorical question
[21:04:40] <kreign> "how do you presume the XenServer can boot the iso if it's not 'able to see' it from the console?
[21:04:57] <kreign> IIRC an ISO directory displays itself as an xapi mount when you do a df
[21:05:01] <kreign> I could be mistaken on that
[21:05:37] <blood_> kreign: ok I'll ask him =)
[21:05:45] <kreign> or you could look yourself
[21:05:49] <kreign> he's likely just as dangerous as you are
[21:06:20] <blood_> kreign: :)
[21:06:36] <kreign> ssh in
[21:06:38] <kreign> df
[21:06:40] <kreign> done
[21:06:43] <blood_> k sec
[21:08:35] <blood_> ok you were right=)
[21:08:50] <blood_> kreign: the iso shows up as bootable when we execute the "file name.iso"
[21:10:23] <kreign> the 'iso store' is just an nfs or cifs mounted path.
[21:10:28] <kreign> added to their xapi
[21:10:29] <kreign> fwiw
[21:10:32] <kreign> I hate xapi
[21:10:35] <kreign> :P
[21:11:03] <blood_> kreign: ok what do you want me to try next?
[21:11:39] <kreign> blood_, ok, so it's bootable. I'm not sure what's causing your error, exactly. is it a cifs or nfs mount/share?
[21:11:49] <blood_> cifs
[21:11:52] <kreign> blood_, I might guess it's permission related.
[21:12:02] <kreign> blood_, see if you can mount the ISO on the (XS) console
[21:12:13] <kreign> mount -o loop filenamepath /mnt
[21:12:21] <blood_> kreign: well if I log into the VM in question, I am able to mount the ISO just fine
[21:12:30] <kreign> ehhhh?
[21:13:01] <blood_> kreign: I just can't get it to boot as it starts up
[21:13:56] <blood_> kreign: basically, by the time the guest console refreshes, it has already passed the initial startup and has begun starting services
[21:14:34] <blood_> kreign: I know in VMware, they allow you to set a startup delay so you can see the initial startup
[21:15:33] <blood_> kreign: let me clarify something.. I can mount the iso just fine only if I set startup options to "Boot from HD". If i "Boot from DVD" it gets the above error I mentioned
[21:17:15] <blood_> so basically I am never able to see the screen which will allow me to hit a key to boot from the DVD
[21:17:36] <kreign> ah
[21:17:43] <kreign> how long is this delay?
[21:17:49] <tabularasa> freaking AGEE isn't authenticating to my AD... wtf
[21:17:54] <kreign> blood_, it sounds to me that XS isn't even seeing the ISO/mounting it, tbh
[21:18:03] <kreign> tabularasa, AGEE?
[21:18:05] <blood_> kreign: well the console refreshes right before I am able to get the login prompt in my VM
[21:18:11] <JarianGibson> tabularasa: service account?
[21:18:38] <blood_> kreign: Is there a way I can have XS delay the startup and allow me to see the console right when it restarts?
[21:19:33] <tabularasa> yeah
[21:19:42] <tabularasa> DN is identical to a working one
[21:19:52] <tabularasa> i'm adding a second AG Virtual Server to my netscalers
[21:19:57] <tabularasa> kreign: access gateway
[21:21:14] <tabularasa> JarianGibson: is there a good authentication troubleshooting tool?
[21:21:24] <tabularasa> i can ping it.. i can telnet to the server on 389 and it opens the socket
[21:21:26] <JarianGibson> ssh to appoliance
[21:21:28] <JarianGibson> shell
[21:21:31] <JarianGibson> cd /tmp
[21:21:32] <tabularasa> i verified the username/password
[21:21:34] <tabularasa> kk
[21:21:40] <JarianGibson> cat aaad.debug
[21:21:47] <tabularasa> ah
[21:21:49] <tabularasa> awesome
[21:21:53] <tabularasa> thats what i needed
[21:22:10] <tabularasa> Hey... hows workspace control work on AGEE in ICA proxy mode?
[21:22:39] <JarianGibson> should work the same as ag or wi
[21:23:48] <tabularasa> http://i51.tinypic.com/2dhzqbn.png
[21:24:10] <kreign> blood_, i'm having a difficult time following you here and working on a couple things right now in addition
[21:24:25] <kreign> blood_, is this a linux VM?
[21:24:34] <blood_> kreign: yes running CentOS 5.5
[21:24:36] <kreign> ok
[21:24:39] <kreign> following a bit better now
[21:24:50] <kreign> blood_, vm boots, just not from the ISO, correct?
[21:24:57] <kreign> and you can mount it fine
[21:25:03] <kreign> from within the guest
[21:25:04] <blood_> kreign: right because I am unable to see the startup screen on console
[21:25:21] <blood_> kreign: not sure if the CentOS livecd requires me to press a key
[21:25:29] <blood_> kreign: correct I can mount it just fine in guest
[21:27:00] <kreign> blood_, how long is the delay before the screen flashes?
[21:27:03] <tabularasa> send_reject sending reject to kernel
[21:27:04] <kreign> ie how long are visuals impeded
[21:27:23] <tabularasa> ok, it think binddn bindpw failed with invalid credentials
[21:27:28] <tabularasa> and i know its right.. wtf
[21:29:30] <tabularasa> JarianGibson: you never seen that?
[21:29:34] <tabularasa> i'm googling now
[21:30:18] <blood_> kreign: around 15 seconds. We noticed the Boot Startup Options for linux VMs are different then Windows VMs. We also tried using a few other iso's that we know work with our windows VMs and the same error occurs as I stated above.
[21:31:03] *** Kroms has joined #Citrix
[21:31:09] <kreign> 'boot startup options'? explain
[21:31:20] <Kroms> Hi.
[21:31:53] <blood_> kreign: XenCenter->Linux VM->General Tab->Startup Options
[21:31:55] <Kroms> I'm trying to get XenServer to mount my USB Storage device to a Trixbox/CentOS 5.5 Virtual Machine and am having issues.  Can anyone lend some assistance?
[21:32:33] <blood_> kreign: We select "Boot from DVD" in that tab and the error occurs where the VM does not start at all. It will only start when startup options are set to "Boot from HD"
[21:33:40] <kreign> blood_, do you have boot order on the VM set to preference the DVD drive?
[21:33:54] <kreign> rather you've tried that, i presume
[21:34:10] <blood_> kreign: yes if we select that then the VM fails to start and XenCenter spits out the error above
[21:34:19] <kreign> ok, on the same page
[21:34:31] <kreign> just getting restatement so I was sure we're talking about the same thing
[21:36:13] <blood_> kreign: what if we created a new VM from a Windows template then attached the CentOS VM virtual disk to it and try to boot from iso ?
[21:37:00] <blood_> not a pretty solution but if it's the only thing that will work I'm all for it
[21:37:33] <kreign> blood_, there should be no difference, but i can't say for sure.
[21:37:41] <kreign> blood_, i'm not that familiar with the ugly side of XS
[21:38:00] <kreign> mostly, that it's there, sorta like a mother-in-law (tryt not to provoke it because it can make life miserable)
[21:38:12] *** uncon has quit IRC
[21:42:19] <gblfxt> i tried booting from centos before as well, apparently the boot options made available in the gui arent enough to tweak it to boot properly
[21:43:37] <gblfxt> ended up installing ubuntu, but i was fairly new to citrix back then, now i know there are more options on the cmd line level
[21:44:52] <blood_> kreign: We just realized it may in fact be our CentOS template which is causing the Boot DVD not to work properly. We used the CentOS 5 (64 bit) template. I also remember someone here mentioning that they only see CentOS 5 (32bit)/(64bit) and not 5.x templates. We actually see CentOS 5, 5.0, 5.1, 5.2, 5.3
[21:45:12] <blood_> We are running XenServer 5.6 FP1
[21:45:36] <kreign> ah
[21:45:42] <kreign> blood_, well there ya go, maybe
[21:46:00] <blood_> kreign: are you able to check if you see CentOS 5.x templates?
[21:46:10] *** GrimdinWork has quit IRC
[21:46:10] <kreign> on 5.6fp1?
[21:46:14] <blood_> kreign: yes
[21:46:23] <blood_> we are wondering if our current install has some corruption
[21:46:33] <kreign> I actually don't have an XS system quick at hand.
[21:46:47] <kreign> blood_, was it an upgrade?
[21:46:50] <blood_> yes
[21:46:51] <kreign> I don't believe in XS upgrades
[21:47:00] <kreign> IMO: wipe and reinstall (it's quick enough)
[21:47:13] <kreign> i've done half a dozen upgrades, all of them ugly
[21:47:27] <kreign> lots of little natty shit
[21:47:35] <kreign> and IMO the templates suck anyway.
[21:48:08] *** chimera has joined #Citrix
[21:48:26] <chimera> hi folks
[21:48:28] *** chimera is now known as Guest51860
[21:48:43] *** Guest51860 is now known as _chimera
[21:49:21] <blood_> kreign: so it's possible to create a custom template to use instead?
[21:49:58] <kreign> blood_, hell if I know. :P
[21:50:05] <kreign> going to say 'yes'
[21:50:07] <kreign> but i've never done it
[21:50:12] <kreign> (or it's been a while, i can't remember which)
[21:51:36] <blood_> anyone have access to XS 5.6 FP1? Wondering if you have CentOS 5.x templates
[21:52:12] *** Ninthek has joined #Citrix
[21:56:15] <blood_> kreign: http://hardforum.com/showthread.php?t=1582205  same issue I'm having
[21:56:16] <blood_> funny
[21:56:56] <kreign> blood_, give me a bit, i'll take a look
[21:56:57] *** MSilva01 has quit IRC
[21:57:04] <blood_> k
[22:01:15] *** OmNomDePlume has quit IRC
[22:01:44] *** uncon has joined #Citrix
[22:02:27] *** Ninthek has quit IRC
[22:02:36] <blood_> i'll be back later
[22:03:27] *** blood_ has quit IRC
[22:13:28] <_chimera> here comes a tricky one
[22:15:29] <_chimera> is there a way to read the bios setting of a HP dl380 gX installed with vsphere 4.1U1 without going down in maintenance
[22:15:57] <_chimera> means does anyone know a tool for reading the bios setting live?
[22:23:05] *** MSilva01 has joined #Citrix
[22:37:30] <gladier> _chimera: not possible afaik
[22:42:52] <Kroms> Can anyone assist me?  I'm trying to get my USB storage device to be recognized by Linux that is in a Virtual Machine with XenServer.  I'm having a bit of trouble with this.
[22:57:46] <tabularasa> JarianGibson: i figured out the workspace control
[22:58:00] <tabularasa> so, the following things need to be configured for workspace control to work
[22:58:34] <tabularasa> workspace control itself, client detection, AGEE in trusted sites, initiated NOT from an existing ICA session (inluding XD)....
[22:58:47] *** _chimera has quit IRC
[23:00:20] <tabularasa> The planets are aligned!
[23:00:49] *** bark0de has joined #Citrix
[23:01:05] <bark0de> hey all
[23:01:11] <tabularasa> howdy
[23:03:07] <bark0de> i am having a problem with a XenDesktop implementation at one of our clients. They had misunderstood how thin provisioning works in XenServer, and so they only gave their virtual servers small disk allocations thinking they could grow. They ran into issues when the roaming profiles directory grew too large and the system volume became full. We fixed this by making more local storage available
[23:03:07] <bark0de> in XenCenter and creating a new partition, and moving over the userprofiles directory to the new partition so they can continue to grow without consuming the system volume
[23:03:36] <bark0de> but now, when the XenDesktop users login, they seem to be pulling the same profile from the day that we made the change
[23:04:06] <bark0de> so as an example, their outlook mail is getting purged everyday as it is pulling the outlook.pst from their AppData folder which is now 2 weeks old
[23:04:20] <bark0de> as if it is reading from the userprofile correctly, but not writing back to it?
[23:04:48] <tabularasa> are you using PVS ?
[23:05:58] <tabularasa> are you using Citrix Profile Manager?
[23:07:30] <bark0de> yes
[23:07:38] <bark0de> Citrix Profile Manager
[23:07:46] *** Trixboxer has quit IRC
[23:09:18] <tabularasa> do you delete cached copys of profiles upon logoff ?
[23:09:37] <bark0de> i had not noticed that option
[23:09:51] <tabularasa> thats a wierd issue... its like the sync back isn't working...  are you using active write-back on the profile?
[23:09:52] <bark0de> but i am assuming that i would want that
[23:09:58] <tabularasa> yes, i agree
[23:10:12] <tabularasa> something is amuck as its not syncing upon logoff
[23:10:19] <tabularasa> i'd go through your GPO settings and make some tweaks
[23:10:23] <bark0de> perhaps its not reading from the new profile location at all, and it is simply the last working cache
[23:10:48] <bark0de> yeah i knew it had to be more of an Active Directory issue than a XenDekstop issue
[23:11:03] <bark0de> :/ their whole AD setup is a mess
[23:11:06] <tabularasa> yes, it seems so
[23:11:55] <bark0de> on the bright side, I found this citrix channel :)
[23:12:02] <tabularasa> nice.. how did you find it?
[23:12:21] <bark0de> i googled citrix irc and found this wiki http://citrixwiki.us/index.php/Citrix_IRC_Channel
[23:12:30] <tabularasa> nice
[23:13:14] <tabularasa> welcome, i hope you stick around
[23:13:34] <bark0de> thanks! now that i know one exists, i'd like to become a regular
[23:13:48] <tabularasa> we are always in need of regulars
[23:13:54] <tabularasa> what are your skillsets?
[23:13:55] <bark0de> we do a lot of virtualization so hopefully i can provide some knowledge
[23:14:02] <tabularasa> you obviously know mine, if you read the wiki.. heh
[23:14:04] <tabularasa> great
[23:14:15] <tabularasa> we have a lot of XenServer traffic in here
[23:14:49] <bark0de> i am real strong when it comes to XS, just getting deeper into XenApp and XenDesktop recently
[23:15:29] <tabularasa> Awesome.. Its great to have another XS guy in here
[23:15:31] <tabularasa> where you from?
[23:15:41] <bark0de> :) glad to be here
[23:15:51] <bark0de> i'm from beautiful and sunny Miami, FL
[23:15:57] <bark0de> near citrix HQ lol
[23:16:03] <tabularasa> nice
[23:16:08] <tabularasa> SC here
[23:16:40] *** Grimdin has joined #Citrix
[23:18:10] <bark0de> oh cool
[23:21:33] <bark0de> so this channel has no Ops or voice?
[23:22:38] <tabularasa> there are ops
[23:22:41] <tabularasa> we are just hidden
[23:22:45] <tabularasa> :)
[23:23:52] <bark0de> thats cool :D
[23:26:54] <bark0de> well Ryan it was nice to meet you, i am done for the day but i'll come back tomorrow when i get back to the office. have a good night bud
[23:27:14] <bark0de> i'm Michael by the way
[23:27:17] <bark0de> c'ya
[23:27:29] <tabularasa> sounds good... have a good one
[23:27:33] *** bark0de has left #Citrix
[23:38:45] *** jamesd2 has quit IRC
[23:39:59] *** jamesd2 has joined #Citrix
[23:44:35] *** gblfxt has quit IRC
[23:45:03] *** gblfxt has joined #Citrix





top