March 30, 2011  
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31

[00:03:01] <gblfxt> tabularasa, ah, i meant which version of citrix does UPM come with, or is it its own product?
[00:08:17] *** rev78 has quit IRC
[00:13:30] *** skister has joined #Citrix
[00:21:03] *** skister has quit IRC
[00:40:24] *** Meson has joined #Citrix
[02:04:54] <rocky439> does anyone have a good backup script they like to run
[02:27:47] <tabularasa> gblfxt: it is its own product
[02:27:50] <tabularasa> anyone around ?
[02:28:19] <tabularasa> Looking for some scripting logic... I'm setting up my automatic reboot schedules for my xenapp servers
[02:28:29] <tabularasa> the new policies for that are freaking sweet...
[02:28:34] <makson> tabularasa: it's all built into GPO's !!
[02:28:35] <makson> oh haha
[02:28:48] <tabularasa> I added a logoff.bat file that will do a query session and then do a "logoff" for anyone thats already logged in
[02:29:05] <tabularasa> which, i totally want clean logoffs before I bounce the servers, so the roaming profiles will resync and delete upon logoff
[02:29:19] <tabularasa> but i'm looking for some logic that i can delete any of the profiles that don't get cleaned when the server comes back up
[02:29:28] <tabularasa> withoug deleteing adminsitrator/ctxuser/blahblah
[02:29:37] <Meson> Anyone going to TechEd this year?
[02:29:45] <tabularasa> where is it?
[02:29:50] <Meson> Atlanta
[02:30:05] <tabularasa> query session >session.txt
[02:30:05] <tabularasa> for /f "skip=1 tokens=3," %%i in (session.txt) DO logoff %%i
[02:30:05] <tabularasa> del c:\session.txt
[02:30:08] <makson> tabularasa: i believe there is a policy that loggs off those user sessions. When you reboot the server PVS will clear out the disk...
[02:30:12] <makson> ah wait
[02:30:12] <tabularasa> found this off Microsoft KB... hot
[02:30:13] <makson> u don't use VPS right ?
[02:30:14] <makson> PVS
[02:30:15] <tabularasa> no PVS
[02:30:21] <tabularasa> Meson: when is it?
[02:31:18] <makson> tabularasa: no UPM?
[02:31:19] <kdavy_> heh. i just accidentally the whole thing with my Splunk install
[02:31:27] <tabularasa> UPM, yeah.. why?
[02:31:39] <tabularasa> am i missing some functionality?
[02:31:49] <Meson> tabularasa May 16-19
[02:31:52] <kdavy_> the whole thing = the entire network
[02:31:54] <makson> tabularasa: just confused when you said roaming profile, there is option to delete profiles on logoff
[02:32:06] <tabularasa> sure... unless shit happens
[02:32:10] <tabularasa> and it does
[02:32:14] <kdavy_> note to self: do not use TCP for syslog on Cisco PIX
[02:32:19] <tabularasa> sometimes a logoff goes bad and ntuser.dat doesn't clear
[02:32:24] <tabularasa> kdavy_: heh
[02:32:32] <tabularasa> i don't see why you'd ever want to use TCP for syslog
[02:32:45] * tabularasa failed the CCIE written v4 today... :-/
[02:33:04] <makson> tabularasa: darn sorry to hear that... it's a killer test though.
[02:33:10] <tabularasa> i was totally unprepared
[02:33:23] <tabularasa> i killed the v3 written... thought it "wont be that different"
[02:33:29] <tabularasa> i was wrong....
[02:33:29] <makson> tabularasa: so you wan't a script that will delete all profiles except the ones above.. seems like that would be a shutdown script u would write.
[02:33:49] <kdavy_> tabularasa, well, i don't see why either. our network admin is on vacation and i decided to play with Splunk for managing PIX access lists because looking through a catch-all text log file is a real pain
[02:33:54] <tabularasa> it would need to run after reboot, as the ntuser.dat's get locked
[02:34:12] <kdavy_> * managing access lists = managing deny logs.
[02:34:13] <makson> tabularasa: http://www.theshonkproject.com/index.php?option=com_content&task=view&id=25&Itemid=30
[02:34:20] <tabularasa> i personally like syslog-ng + tail + grep
[02:34:22] <makson> tabularasa: saw that a while ago, no need to reinvent the wheel
[02:34:46] <tabularasa> i'll check that out makson
[02:35:02] <tabularasa> makson: holy crap
[02:35:28] <tabularasa> not bad
[02:35:52] <tabularasa> actually, freaking awesome
[02:36:03] <tabularasa> Const ProfileExclusionList = "|administrator|all users|default user|localservice|networkservice|ctx_smauser|ctx_cpuuser|ctx_cpsvcuser|ctx_streamingsvc|ctx_configmgr|"
[02:36:08] <kdavy_> tabularasa, well to be honest, even though i should've used UDP for syslog... is it predictable behavior for PIX to disallow all new connections if a TCP syslog server is down?
[02:36:09] <tabularasa> perfect
[02:36:21] <tabularasa> no
[02:36:32] <tabularasa> not at all
[02:37:19] <kdavy_> tabularasa, apparently that's how it works though. http://kb.prismmicrosys.com/evtpass/evtPages/MessageCode_PIX-3-201008_53050.asp
[02:37:28] <kdavy_> that was the exact issue and exact message code
[02:38:26] <kdavy_> so, um, wtf.
[02:40:05] <kdavy_> good thing i restarted the Splunk syslog service after hours and not in the middle of the day
[02:40:09] <tabularasa> what version of code is the damn thing on?
[02:40:11] <tabularasa> thats retarded
[02:40:23] <kdavy_> tabularasa, 6.3(3)
[02:40:28] <tabularasa> oh ffs...
[02:40:44] <tabularasa> thats like 15 versions old.  :)
[02:41:05] <kdavy_> tabularasa, it's a PIX 520 - 6.3(4) is the last supported version
[02:41:10] <tabularasa> 7.0 7.1 7.11 7.2 8.0 8.04 8.1 .82 .823 .83 .84
[02:41:12] <kdavy_> it has a floppy drive :)
[02:41:15] <tabularasa> hurtin...
[02:41:22] <tabularasa> yeah, no need for TCP though... screw that
[02:42:06] <tabularasa> makson: this script is sweet
[02:42:42] <kdavy_> and i can KINDA understand _why_ it went down, but still that's fucking stupid
[02:42:55] <tabularasa> agreed
[02:43:28] <kdavy_> probably to prevent it from going out of memory or having a buffer overflow of some sort, since it'd just keep retransmitting the damn TCP packets
[02:45:38] <tabularasa> makson: i hope this works, it will round out my reboot schedule perfectly
[02:46:38] <kdavy_> tabularasa, what is your reboot schedule?
[02:46:53] <tabularasa> per management once a week... saturday at 2am
[02:47:01] <tabularasa> if i had my way, i'd do it every night
[02:47:05] <kdavy_> nightly @1:15 am here
[02:47:17] <tabularasa> but we are selling the solution that you can goto sleep and reconnect the same place you were last night!!
[02:47:24] <tabularasa> and i can see the point, but blah
[02:47:41] <tabularasa> i'd love to do nightly
[02:47:45] <tabularasa> i know it would run better
[02:47:51] <tabularasa> do you do any sort of profile cleaning upon reboot?
[02:47:56] <kdavy_> ha, yeah in my case we are selling a solution where if you don't log out in the evening, your profile changes won't be saved
[02:48:03] <kdavy_> yea, profile cleaning after reboot
[02:48:08] <tabularasa> how are you doing it?
[02:49:15] <kdavy_> just a script to delete old profiles from xenapp servers based on two variables: amount of free space on server and number of profiles currently on the C: Drive
[02:49:33] <kdavy_> nothing more in-depth
[02:49:49] <tabularasa> how do you delete them without deleting important shit?
[02:50:25] <kdavy_> tabularasa, roaming profiles. 80% of users are still not on UPM
[02:51:07] <kdavy_> or is that not what you meant?
[02:56:03] <makson> kdavy_: your still running XA5 eh?
[02:56:18] <kdavy_> makson, yep
[02:56:22] <makson> did you ever get your XA6 test farm going?
[02:56:30] <makson> kdavy_: smart, they worked out a lot of the bugs
[02:56:33] <makson> for XA6.
[02:56:40] <kdavy_> i have 3 servers in the test farm, but no production users
[02:57:02] <kdavy_> and with my pipeline, it'll probably be 2012 before i start any migration efforts
[02:57:53] <tabularasa> sure, but still, if a roaming profile fails to delete upon logoff, i want to make sure its destroyed before anyone else logs in
[02:58:00] <tabularasa> i'm gonna go hang out with my girlfriend... see you guys later
[02:58:00] <makson> kdavy_: at the end of the day, it's what features can be offered to the end user's that would better there expereince. for you, I don't see how XA6 would help with that besides the UPM part, and you can do that with XA5.
[02:58:16] <makson> tabularasa: haah, i got 15 till mine comes home trying to get as much done on the comp as possible.
[02:58:22] <kdavy_> tabularasa, roaming profile deletion happens right after reboot
[02:58:33] <makson> tabularasa: maybe disable logins on boot, until profiles deleted?
[02:58:40] <tabularasa> but if shit happens, and the ntuser.dat isn't unlocked, it won't delete
[02:59:06] <tabularasa> unless your script has logic to unload the profile hive
[02:59:10] <tabularasa> which sometimes locks also
[02:59:17] <kdavy_> deletion script also deletes based on last modified date, so it will not touch anything newer than 24 hours
[02:59:25] <tabularasa> ah, gotcha
[02:59:37] <kdavy_> this is an awesome cartoon... http://www.youtube.com/watch?v=8CRYmcun3tc
[02:59:41] <tabularasa> good discussion...  this is the kind of crap we need to have our "scheduled talks" about
[02:59:50] <kdavy_> yeah :)
[02:59:52] <tabularasa> have a good night/morning guys
[03:01:55] <makson> tabularasa: yah, we have to get back on that... we should circle back on that maybe begining of next month, this month is slammed for me.
[03:11:38] *** ScottCochran has joined #Citrix
[03:11:47] <JarianGibson> chop chop
[03:12:04] <kdavy_> pork chop
[03:12:11] <JarianGibson> and apple sauce
[03:12:30] <kdavy_> and a sledgehammer to the fucking PIX firewall :)
[03:12:46] <JarianGibson> nice
[03:13:38] <JarianGibson> i feel like that guy in the video somedays
[03:13:53] <kdavy_> oh yeah
[03:14:14] <JarianGibson> haha
[03:14:29] <JarianGibson> a lot lately
[03:16:00] <kdavy_> ok i gotta go... still at work, been doing damage control
[03:19:57] *** robust has quit IRC
[03:37:08] *** zentron has joined #Citrix
[03:41:15] *** Meson has quit IRC
[03:53:31] *** jamesd2 has joined #Citrix
[03:53:57] *** jamesd2 has quit IRC
[03:54:38] *** jamesd2 has joined #Citrix
[03:56:38] <kdavy> ping
[03:57:39] <JarianGibson> pong
[03:59:19] <makson> JarianGibson:
[03:59:27] <JarianGibson> makson: it didn't beep
[03:59:30] <JarianGibson> wtf
[03:59:38] <makson> might because of ssh?
[03:59:44] <makson> i think i looked into this a while ago
[03:59:45] <JarianGibson> goo point
[04:00:00] <JarianGibson> trying to get a better notifier
[04:00:13] <makson> there is an irssi channel
[04:00:30] <JarianGibson> libnotify
[04:01:53] <makson> growl same concept
[04:02:00] <kdavy> beep
[04:02:11] <makson> http://code.google.com/p/irssi-libnotify/
[04:02:15] <makson> bloop bloop
[04:02:16] <makson> <><
[04:02:19] <kdavy> Colloquy ftw
[04:03:50] <makson> be back in a bit. kdavy VPS ftw
[04:05:06] <kdavy> makson: 5 racks in a colo ftw :)
[04:05:23] <JarianGibson> i don't use mac
[04:07:07] <kdavy> JarianGibson: to each their own. i like mac and windows equally
[04:07:17] <JarianGibson> mac to $$
[04:07:25] <kdavy> and have opensolaris on my netbook :)
[04:07:47] <JarianGibson> nice
[04:07:57] <kdavy> JarianGibson: you have $$, how much is your time worth?
[04:07:59] *** zentron has quit IRC
[04:08:23] <kdavy> i get things done on mac slightly faster because it's less annoying, i think that's a good enough payoff
[04:08:49] <JarianGibson> i run my windows 7 machine in a vm that is on hyper-v on my laptop
[04:09:04] <JarianGibson> remotefx
[04:09:39] <JarianGibson> makson
[04:09:55] <kdavy> JarianGibson: now we have a topic. what do you think of remotefx when it comes to wan connections?
[04:10:05] <JarianGibson> haven't tested it yet. need to
[04:10:08] <kdavy> let's say plenty of bandwidth, but 30-50ms latency
[04:10:34] <JarianGibson> rdp over wan is ok but with remotefx don't know probably less than ok
[04:10:48] <JarianGibson> setting up test server for this
[04:11:00] <kdavy> i've been accessing my management server at the colo via remotefx since the day W7/R2 SP1 came out
[04:11:09] <JarianGibson> and
[04:11:35] <kdavy> and it is much more usable than via RDP, mostly because i do a lot of double-hop things
[04:11:37] <JarianGibson> are you just using grpahics or doing usb redirection as well?
[04:12:04] <kdavy> no usb redirection, i'm using RDS remotefx, not the hardware accelerated guest VDI stuff
[04:12:27] <JarianGibson> oh ok, that i can see
[04:12:32] <JarianGibson> just cpu offload
[04:12:51] <JarianGibson> vdi a little differen.t
[04:12:56] <kdavy> but RemoteFX->VNC (for example XenCenter or vSphere Client or IBM Bladecenter management) is orders of magnitude faster than via pure RDP
[04:13:08] <JarianGibson> nice
[04:13:10] <kdavy> so is multimedia stuff like flash
[04:13:30] <JarianGibson> i am getting readty to setup this up at the office and test from home
[04:13:36] <JarianGibson> going to test both rds and vdi
[04:13:48] <kdavy> the only thing i don't like so far is, at 30+ms your mouse cursor starts lagging because it's displayed on the host and encoded into RemoteFX, instead of being a GDI sprite on top of RDP
[04:15:04] <kdavy> but that may also be related to CPU performance... my management server is a P4-EM64T based dual single core Xeon, pretty much the slowest Xeon you can run a 64bit OS on
[04:15:39] <kdavy> its an old HS20 64bit IBM blade
[04:18:05] <JarianGibson> i have some video issues in my testing but using gpu of laptop ati m7820
[04:18:36] <kdavy> JarianGibson: what kinda video issues?
[04:22:50] <JarianGibson> just sometimes vdieo recovering in win 7 vm
[04:23:16] <JarianGibson> makson: nned script uploaded
[04:39:37] <ScottCochran> Any AppSense guys around>>
[04:40:40] <ScottCochran> JarianGibson You familiar with AppSense?
[04:40:52] <JarianGibson> just getting into it
[04:42:12] <ScottCochran> yea I took the new Professional and Specialist class two months ago but forgot a lot bc my last project didnt use it... Now this one is...
[04:42:25] <JarianGibson> taking them next month
[04:42:50] <ScottCochran> You know why a CCA client deployment would just be sitting at Initializing the Installation?
[04:45:00] <JarianGibson> hmm, what verison?
[04:45:19] <makson> JarianGibson:
[04:45:26] <JarianGibson> yes
[04:45:49] <makson> ScottCochran: meson is big into appsence although he isn't around right now it appears.
[04:46:31] <JarianGibson> makson: i don't thinki can do lib notify
[04:46:37] <ScottCochran> 8.1.1x...  Here is a pic  https://docs.google.com/leaf?id=0B95u7324eu91MWNhMWY4YTctNjVmNC00YWY2LTg0OTgtM2NiMTJjNWExMWE5&hl=en
[04:46:41] <makson> JarianGibson: does it have perl dependencies
[04:47:00] <JarianGibson> perl loaded. i think it i html piece
[04:47:07] <ScottCochran> makson: thanks
[05:41:59] *** rocky439 has quit IRC
[05:44:58] *** rocky566 has joined #Citrix
[06:45:35] *** rocky566 has quit IRC
[07:59:58] *** kdavy_ has quit IRC
[09:38:23] *** Jenius has joined #Citrix
[10:00:50] *** Grimdin has joined #Citrix
[10:31:53] *** Kevin` has quit IRC
[10:32:00] *** Kevin` has joined #Citrix
[10:45:59] *** Jenius has quit IRC
[11:31:02] *** derwayne has joined #Citrix
[11:43:36] *** derwayne has quit IRC
[11:46:59] *** Quietly_Confiden has joined #Citrix
[11:51:03] <Quietly_Confiden> Is there any kind of 'getting started with citrix xenapp' document around? Just something that I can use in conjunction with the support docs to read up on it.
[11:51:13] *** Quietly_Confiden is now known as QuietlyConfident
[11:52:51] *** Trixboxer has joined #Citrix
[12:31:52] *** BWMerlin has joined #Citrix
[13:05:46] *** QuietlyConfident has quit IRC
[13:29:05] *** Tollar has joined #Citrix
[13:29:27] *** brad[] has quit IRC
[13:30:07] *** brad[] has joined #Citrix
[13:40:30] *** ScottCochran has quit IRC
[13:53:46] *** derwayne has joined #Citrix
[14:08:33] <Tollar> I am trying to use powershell cmdlets from c#, but not using the powershell SDK, anyone know of any site with good examples how to do this?
[14:09:30] <Tollar> Forgot to say that it is XenApp 6.5 I am on now, and trying to test the remoting powershell commands
[14:14:22] *** houseofcurrie has joined #Citrix
[14:36:50] *** ScottCochran has joined #Citrix
[14:38:16] *** Meson has joined #Citrix
[14:49:27] *** BWMerlin has quit IRC
[15:10:09] <makson> Morning everyone
[15:11:38] <Meson> Good Morning
[15:22:19] <ScottCochran> Morning all!
[15:23:22] <makson> So ThinPC, that's like winFLP?
[15:23:26] <makson> but windows 7 version >
[15:23:52] <makson> I'll have to see if it's on tech net
[15:24:28] *** pyrofallout has quit IRC
[15:28:45] <tabularasa> I want to check out thinpc... it's beta right now...
[15:28:52] *** Tollar has quit IRC
[15:29:28] <tabularasa> I saw a link for it... it must have been getting hammered because it failed all day yesterday.
[15:49:16] *** Pierre_Marmignon has quit IRC
[15:50:08] *** pyrofallout has joined #Citrix
[15:53:47] *** The_Machine has joined #Citrix
[16:21:38] *** bubba has joined #Citrix
[16:23:38] *** bubba has quit IRC
[16:38:13] <gblfxt> when deleting profiles on logout how do you preserve specific programs setting/logs that are kept in that profile?
[16:41:39] *** pyrofallout has quit IRC
[16:43:02] *** pyrofallout has joined #Citrix
[16:46:41] *** waynerr__ has joined #Citrix
[16:50:21] *** waynerr has quit IRC
[17:00:51] *** waynerr has joined #Citrix
[17:03:22] *** derwayne has quit IRC
[17:19:40] *** The_Machine has quit IRC
[17:36:10] *** CrazyGir is now known as Gir
[17:36:36] *** Gir is now known as CrazyGir
[17:39:39] *** unop has joined #Citrix
[17:47:12] *** Zed` has quit IRC
[17:50:19] *** Zed` has joined #Citrix
[18:37:20] *** kdavy[1] has joined #Citrix
[18:41:04] *** Zed` has quit IRC
[18:41:39] *** Zed` has joined #Citrix
[19:26:29] *** waynerr__ has quit IRC
[21:03:08] *** cathederal_ has joined #Citrix
[21:05:15] *** Trixboxer has quit IRC
[21:27:39] *** Meson has quit IRC
[21:33:40] *** cathederal_ has quit IRC
[21:37:46] *** cathederal_ has joined #Citrix
[22:05:55] *** cathederal_ has quit IRC
[22:09:59] *** pyrofallout has left #Citrix
[22:13:48] *** cathederal_ has joined #Citrix
[22:16:16] <kdavy[1]> it
[22:16:22] <kdavy[1]> it's quiet in here...
[22:16:26] *** kdavy[1] is now known as kdavy_
[22:25:52] <nikade> sure is
[22:28:21] <kdavy_> i'm pretending to be multitasking
[22:28:45] <kdavy_> have four different vendors remoted into four remote desktops via my machine, watching them work
[22:29:10] <kdavy_> making sure they dont do anything stupid
[22:32:34] <nikade> lol, yeah you better watch them
[22:34:38] <kdavy_> yep, they tend to do stupid shit
[22:34:45] <kdavy_> i'm recording everything they do, as well
[22:35:47] <nikade> yeah, that might be wise incase you need to "check on em" afterwards
[22:56:00] <zaf> anybody gotten phd virtual to work in a satisfactory way yet?
[22:56:11] <kdavy_> zaf, i thought it works well
[22:56:25] <kdavy_> i havent tried it but heard others say good things about it
[22:56:30] <zaf> taking FOREVER for backup seed
[22:56:49] <kdavy_> backup seed? whats that
[22:56:54] <zaf> initial backup
[22:57:10] <kdavy_> ah. well there's a lot of data for initial backup, no?
[22:58:13] <kdavy_> it'll go faster over 10GbE
[22:58:16] <zaf> i've got around 8tb
[23:00:31] <zaf> just not getting a reliable throughput. there are 4 concurrent jobs/appliances running. they keep fluctuating between a nice steady 50MB/s and a stalled state
[23:08:28] <kdavy_> 4 concurrent? off of how many spindles?
[23:08:40] <kdavy_> your dicks are probably saturated
[23:08:43] <kdavy_> *disks
[23:08:45] <zaf> 32
[23:09:19] <kdavy_> hmm yea 32 should be more than plenty
[23:09:40] <kdavy_> is it going over http/https or does phd use a proprietary protocol?
[23:10:09] <zaf> it uses VMs that take xen snapshots and then mount those snapshots, saves data over cifs
[23:10:24] <kdavy_> ew, cifs
[23:10:53] <kdavy_> increasing # of cores allocated to dom0 may help
[23:11:12] <kdavy_> check dom0 load during the export
[23:14:27] <zaf> according to PHD it wont help, heh
[23:14:41] <zaf> just hoping that after the seed, the nightlies are much quicker
[23:15:38] *** ScottCochran has quit IRC
[23:55:38] *** cathederal_ has quit IRC

top