[00:04:13] *** AlasAway is now known as Alasdairrr [00:25:18] *** cathederal has quit IRC [00:29:38] *** Jenius has quit IRC [00:30:09] *** [M]ax has quit IRC [00:31:04] *** [M]ax has joined #Citrix [00:49:58] <JarianGibson> makson: kdavy mentioned something [00:50:08] <JarianGibson> i am going and done to share a pad [01:00:13] *** [M]ax has quit IRC [01:05:03] *** [M]ax has joined #Citrix [01:14:48] *** Meson has joined #Citrix [01:17:21] *** Alasdairrr is now known as AlasAway [02:56:35] *** Meson has quit IRC [02:57:22] *** Meson has joined #Citrix [03:15:56] <kdavy> JarianGibson: hopefully i can join you. even if i can't come i'd still be willing to help organize a CitrixIRC pad :) [03:21:37] <kdavy> decision will be made before the 22nd, thats for sure [04:17:59] *** Meson has quit IRC [05:14:08] *** KKA has left #Citrix [08:08:49] *** zaf has quit IRC [08:40:05] *** BWMerlin has joined #Citrix [09:59:57] *** brad[] has quit IRC [10:02:39] *** brad[] has joined #Citrix [10:26:45] *** q1x has left #Citrix [11:33:49] *** nikade has quit IRC [11:35:06] *** ackaa has joined #Citrix [12:04:11] *** TheG0blin has joined #Citrix [12:05:27] *** AlasAway is now known as Alasdairrr [12:15:26] *** Trixboxer has joined #Citrix [13:44:22] *** brad[] has quit IRC [13:51:16] <tabularasa> morning peeps [13:58:54] <tabularasa> kdavy[1]: doh, i think you are right! [14:33:20] *** MassiveOverkill has quit IRC [14:33:57] *** MassiveOverkill has joined #Citrix [14:58:41] *** antoon_i has left #Citrix [15:02:52] <JarianGibson> no [15:02:57] <JarianGibson> morning [15:14:47] <tabularasa> yo [15:17:45] <dvsxutact> morning all, what is the easiest way, to determine which vif goes with each vm? [15:20:37] <makson> Morning everyone. [15:21:44] *** zaf has joined #Citrix [15:27:20] *** zaf has quit IRC [15:30:13] *** TuxOtaku has joined #Citrix [15:30:53] *** OmNomDeBonBon has joined #Citrix [15:37:24] <tabularasa> kdavy[1]: you around? [15:37:50] <tabularasa> explain to me if an average SSD gets 5-8k IOPs that you can get 600k out of 24 SSDs [15:39:58] *** zaf has joined #Citrix [16:02:02] <makson> anyone disable outlook cached mode via GPO for 2010? [16:02:23] <jduggan> nope [16:02:25] <tabularasa> nope [16:02:39] <jduggan> is there a reason to gpo cache disable? [16:02:48] <jduggan> i thought not using cache was typically not wise [16:02:56] <tabularasa> i dont't really agree [16:03:13] <makson> :\ microsoft's doc says add OUTLK14.adm and it's under account settings, but that's not in the ADM. [16:03:28] <makson> by default it's disabled outlook cached mode, but i don't sync local settings for outlook [16:04:07] <tabularasa> if everything is local, on gig, and you have a decent server without a zillion users [16:04:24] <tabularasa> thats wierd... i'd imagine its in there [16:05:27] <jduggan> tabularasa: yea it does heavily depend on environment - but it stands to reason its better for exchange server loading if you cache client side [16:05:53] <jduggan> also there are caveats to caching, some delays in gal entries n such [16:05:54] <tabularasa> but then it takes up more local space on the XenApp server [16:06:02] <jduggan> yea thats another caveat [16:06:06] <tabularasa> and with 30 users on each XenApp server, you just transfer that load [16:06:07] <jduggan> effectively duplicating space [16:06:28] <tabularasa> or, storing it on a network drive, for 30 users, you transfer that load again [16:06:31] <tabularasa> yeah, its a balance [16:07:31] <makson> why enable cache mode for citrix servers? it just bloats the profile... [16:08:03] <makson> with UPM they recommend completly excluding appdata\local from the syncing . [16:08:11] <makson> or are you guys talking about in non citrix environments? [16:08:14] <tabularasa> no [16:08:23] <tabularasa> i move the cache to a fileshare [16:08:30] <tabularasa> homedir, actually [16:08:44] <tabularasa> but, we use rackspace for exchange [16:08:50] <tabularasa> so, the exchange is very not local [16:08:59] <makson> tabularasa: ah, ok that makes sense then. [16:09:02] <kdavy[1]> tabularasa, yo [16:09:04] <tabularasa> i need to do more with UPM exclusions [16:09:08] <tabularasa> hey kdavy[1] [16:09:10] <kdavy[1]> had a question? [16:09:14] *** kdavy[1] is now known as kdavy_ [16:09:17] <tabularasa> explain to me if an average SSD gets 5-8k IOPs that you can get 600k out of 24 SSDs [16:09:23] <tabularasa> ala Nexenta [16:09:39] <kdavy_> you mean that thing i wrote about yesterday or the day before? [16:10:21] <tabularasa> yeah [16:10:31] <kdavy_> it was probably metadata-related. reported IOPs include ones that come from ARC - ram cache - which is way faster than an ssd [16:10:50] <tabularasa> gotcha [16:11:21] <tabularasa> i'm comparing like 5 different hardware configurations to calculate cost/vdi [16:11:24] <kdavy_> Out of these 600k, about 5k got past RAM, and only 500 made it to spindles [16:11:30] <tabularasa> blades vs local vs UCS vs etc etc [16:11:47] <tabularasa> so far, believe it or not, its cheapest to go a 2U server with Nexenta [16:11:57] <tabularasa> even at the 5000 vdi level [16:11:58] <kdavy_> :) yeah [16:12:00] <tabularasa> not sure i'd want to manage it [16:12:21] <jduggan> kdavy_: is there some calculation to work out how much ram vs ssd cache vs disks you shoudl use ? [16:12:30] <jduggan> or just as much ram and arc as possible [16:12:39] <tabularasa> would take 12 2U servers with 1 Nexenta... thats 576 cores... at 10 users per core - PVS/DDC etc, its like 5250 VDIs] [16:12:56] <tabularasa> 6144GB of RAM.. with memory sharing i guestimate about the same... 5250 VDIs [16:13:03] <kdavy_> jduggan, there are a couple guidelines. are you gonna be around later? i just got to the office [16:13:23] <tabularasa> its like 400k for all of that, with 10gig switching... thats like $75/VDI [16:13:33] <tabularasa> the UCS comes in next best at $130/VDI [16:13:35] <kdavy_> tabularasa, not bad [16:13:44] <jduggan> kdavy_: i'll leaving the office in 2hours, then i have a 3hour drive home, then i'll be back online [16:13:49] <tabularasa> but i have this feeling that the UCS would perform far better, and i KNOW it would be easier to manage [16:14:07] <kdavy_> have you read about supposed new features for vSphere 5? Local SSD cache that is VDI-aware and dedup-enabled [16:14:25] <tabularasa> dedup? [16:14:47] <kdavy_> yeah, NTFS-aware and pattern-aware dedup for Windows VDI VMs [16:14:49] <makson> kdavy_: wow, where did you see that 5 information? [16:14:56] <tabularasa> holy crap [16:15:01] <makson> kdavy_: did it say for View only :P [16:15:06] <kdavy_> http://virtualization.info/en/news/2011/04/more-details-about-vsphere-5-appear-online.html [16:15:27] <kdavy_> "Swap to SSD" [16:15:35] <kdavy_> "Array Integration: Thin Provisioning enabling reclaiming blocks of a thin provisioned LUN on the array when a virtual disk is deleted " [16:15:52] <tabularasa> as far as cores go... getting up to this level CPU seems a bottle neck.. 12core AMD vs 10core Intel? [16:15:59] <kdavy_> "Accelerator for specific use with View (VDI) workloads, providing a read cache optimized for recognizing, handling and deduplicating VDI client images. " [16:16:07] <tabularasa> or could i assume more VDIs/core with the Intels? [16:16:24] <kdavy_> tabularasa, yes, Intels will give you better density per core because of hyperthreading [16:16:28] <tabularasa> these are the assumptions i'm strugging with [16:16:47] <kdavy_> and because of bigger cores and higher clock speed/calculations per cycle [16:17:49] <tabularasa> Citrix gives you VDIs/core... doesnt help me much without architecture comarisions.. :-. [16:17:52] <tabularasa> :-/ [16:17:57] <kdavy_> but then again, Opteron is wayyyy cheaper [16:18:26] <jduggan> we use 12core opterons in our boxes [16:18:28] <jduggan> because tehyre cheap [16:18:54] <tabularasa> well, at the discount levels that we are talking about here, i can get them pretty close [16:19:22] <tabularasa> plus, if you can get 12 VDIs/core on Intel and 10 VDIs/core on AMD, at scale, is it really a cost savings? [16:19:31] <tabularasa> at 16k desktops, these numbers matter [16:20:11] <jduggan> HP had an offer the last few months of buy one get one free on opteron processors [16:20:20] <jduggan> so we bought 5 new hp boxes single proc and got dual [16:22:47] *** Carpe|Diem has quit IRC [16:31:39] <kdavy_> tabularasa, VDI/Core can be dramatically different in corp. vs. edu [16:32:19] <tabularasa> sure... i need more AMD vs. Intel [16:32:24] <tabularasa> i'm going to check some benchmarks.. [16:32:34] <tabularasa> this is where i lean on you hardware guys. ;) [16:33:22] <kdavy_> tabularasa, do you have the funds for a small POC? Amd vs. Intel for example [16:33:53] <kdavy_> could even get demo units from vendors - at this scale they should be all over this deal. make them do the work :) [16:39:13] <gblfxt> robocopy local profile [16:39:57] <kdavy_> hmmm so just found an interesting bug/feature [16:40:37] <gblfxt> i get the idea local/roaming was not designed w citrix in mind [16:42:32] <kdavy_> under certain conditions (nfc what those are, i'm just observing), a XenApp 5 server with Clamwin installed does not map certain users' local printers [16:42:40] <kdavy_> uninstall clamwin - printers map just fine [16:46:46] <gblfxt> dunno, from what i've seen, different av's seem to pick at random what part of citrix they are declaring a virus for the month [16:48:41] *** OmNomDeBonBon has quit IRC [16:50:54] <tabularasa> kdavy_: just no time [17:03:23] <kdavy_> gblfxt, haha yeah. So far i had zero problems with Kaspersky 8.0 Enterprise though, have it deployed on almost every server in the farm by now [17:03:46] <kdavy_> it's perfect... XenApp optimized, low footprint, etc [17:03:56] <kdavy_> low cost too [17:04:28] <gblfxt> nice, ill look into that, thanks... :) [17:04:50] <tabularasa> kdavy_: cool, thanks for the tip [17:04:58] <gblfxt> you use xenapp 6? [17:05:28] <gblfxt> ah, you mentioned xenapp optimized [17:05:41] <kdavy_> licensing for it is kinda weird. if you get it as standalone licenses, it's something like $200/server, but it is included in Kaspersky Business Space Security, making it something like $40/server w/ 3 year subscription, and no per-user fees [17:06:53] <kdavy_> my Kaspersky rep said this makes no sense and he just recommends Business Space security - then you can use the same license pool for desktops and for servers, and any version of Kaspersky is covered under it except ones for Exchange and ISA [17:07:22] <kdavy_> same licenses cover KAV for File Servers, too [17:07:23] <kdavy_> :) [17:08:03] <gblfxt> ive been using clamav for windows, just because it has a low footprint, but dont know if it works right yet either, lol [17:08:28] <tabularasa> o/names [17:08:48] <tabularasa> oops [17:14:30] <tabularasa> Would love to see a Intel E7-4870 vs AMD 6174 article.... [17:18:50] *** kaffien has joined #Citrix [17:25:47] *** kdavy_ has quit IRC [17:26:59] *** TheG0blin has quit IRC [17:39:16] *** BWMerlin has quit IRC [17:40:41] <tabularasa> Nice, my Citrix SE just looped Daniel Feller into my converstaions about the design... [17:40:44] <tabularasa> squirt..... [17:41:13] <makson> tabularasa: Nice !!! [17:41:18] <tabularasa> big time.... [17:41:39] <makson> tabularasa: wow have him come in the channel and chat about it hehe :) That's awesome he can use u as a case study. [17:44:10] <makson> tabularasa: keep us posted. [18:18:34] *** Alasdairrr is now known as AlasAway [18:33:32] *** eastz0r has quit IRC [18:33:50] *** eastz0r has joined #Citrix [18:59:08] *** eastz0r has quit IRC [18:59:27] *** eastz0r has joined #Citrix [19:25:30] *** OmNomDeBonBon has joined #Citrix [19:29:23] <tabularasa> i'm registered for Synergy... squirt! [19:39:06] <tabularasa> Summit and Synergy [19:45:32] *** eastz0r has quit IRC [19:56:35] *** eastz0r has joined #Citrix [20:00:18] *** KaiForce has joined #Citrix [20:03:28] <KaiForce> I added a 2008 x64 server to my PS 4.5 farm and published mstsc.exe. It launches but just shows up in the taskbar, with the only option being "close window" when I right click. Does this whether I specify an .rdp file or not. What is going on? [20:07:13] <tabularasa> Daniel responded to my email... i'm so excited... lol [20:10:05] <KaiForce> hmmm, notepad, same config, launches but screen is corrupt. I must have some incorrect seamless flags set?? [20:12:55] <KaiForce> Is there an issue with multi monitor and Xenapp 5 on 2008x64? Notepad looks OK on the primary monitor but corrupted when I drag it to secondary. [20:13:09] <tabularasa> add more video memory [20:14:30] <KaiForce> tab: Oh, gotcha. I think we had to make an adjustment to that setting because we were having an RDP session degrade. Thanks for the tip. [20:15:44] <KaiForce> is 8k the limit on PS 4.5? [20:15:47] <tabularasa> yup.. its a citrix policy [20:15:53] <tabularasa> wait.. its a reghack in 4.5 [20:16:02] <tabularasa> you can add like 32k or something in registry [20:16:05] <KaiForce> there is a farm setting, I have it set to 8000 [20:16:13] <tabularasa> yeah, it needs to be bigger [20:16:20] <tabularasa> i rememeber reg hacking mine to 32 [20:16:24] <KaiForce> can I change it there or do I need to hack? [20:16:31] <tabularasa> need to hack [20:16:35] <tabularasa> its a simple reg key [20:16:46] <KaiForce> OK. That actually works better because I can restrict it to this server. [20:18:59] *** frogstarr78 has quit IRC [20:19:20] *** frogstarr78 has joined #Citrix [20:20:06] *** Elias_Rus has joined #Citrix [20:41:58] <KaiForce> tabularasa: that was the issue. thanks, I could have been beating my head against wall on this one. [20:44:52] <KaiForce> did you do this hack on 2003 server? [20:47:07] <tabularasa> yup [20:47:12] <tabularasa> its fixed in XA6 [20:47:19] *** Phibs has joined #Citrix [20:47:20] <tabularasa> i think it is... i forget now.... [20:47:27] <Phibs> is there a specific channel for xenserver? [20:47:41] <tabularasa> you be in it [20:47:47] <tabularasa> :) [20:47:56] <Phibs> hehe [20:48:07] <Phibs> Was wondering what the feature 'memory optimizatino' which is not part of the free version buys me [20:48:18] <tabularasa> we only have a couple XenServer gurus in here, so you may have to be patient [20:48:29] <Phibs> np [20:48:42] <Phibs> i'll search around on the site some more too, although its hard to find info sometimes on it ;0 [20:49:06] <Phibs> nm found it in the pdf ! ;) [20:49:19] <Phibs> XenServer reduces costs and improves application performance and protection by sharing unused server memory between VMs on the host server [20:49:20] <tabularasa> heh, nice [20:49:25] <Phibs> not quite sure what that means though haha [20:49:38] <Phibs> I guess that means if a VM is not using all of it's allocated memory, it can be utilized by another vm [20:49:40] <Phibs> so, oversubbed [20:50:22] <Elias_Rus> hi guys [20:51:26] <dvsxutact> morning all, what is the easiest way, to determine which vif goes with each vm? [21:03:43] <gblfxt> could use robocopy to sync the activex [21:04:15] <gblfxt> but then will prolly need a script to register them all too... :( [21:22:04] *** KaiForce has quit IRC [21:27:07] *** KaiForce has joined #Citrix [21:32:13] *** Trixboxer has quit IRC [21:36:22] *** kdavy[1] has joined #Citrix [21:36:25] *** kdavy[1] is now known as kdavy_ [21:47:28] *** Elias_Rus has quit IRC [21:50:03] *** kdavy_ has quit IRC [21:52:31] <makson> hmm XA6 shutdown issue is back? http://forums.citrix.com/thread.jspa?threadID=263059&start=420&tstart=0 [21:52:49] <makson> i rebuilt lab environment, and it looks like i'm seeing this shutdown issue again. after Patch 018 is installed. [22:18:50] <dvsxutact> when i type ifconfig in the xenserver console, all the vif's have FF:FF:FF etc for mac address, is there anyway to set the vif mac to the mac listed in the vm's network page? [22:36:13] *** OmNomDeBonBon has quit IRC [22:36:23] <kreign> good afternoon people. [22:43:09] *** KaiForce has quit IRC [22:53:55] *** Bl_nK has joined #Citrix [22:55:19] *** Bl_nK has left #Citrix [22:59:07] *** draygo has quit IRC [23:00:49] *** draygo has joined #Citrix [23:11:29] *** draygo has quit IRC [23:13:15] *** draygo has joined #Citrix [23:18:08] *** draygo has quit IRC [23:19:35] *** draygo has joined #Citrix [23:37:44] *** draygo has quit IRC [23:39:06] *** draygo has joined #Citrix