From johan.vermeulen7 at telenet.be Thu Apr 3 20:33:21 2014 From: johan.vermeulen7 at telenet.be (johan.vermeulen7 at telenet.be) Date: Thu, 3 Apr 2014 22:33:21 +0200 (CEST) Subject: [K12OSN] usb not working In-Reply-To: <1396286979.16858.2.camel@alpaga.bursztynowski.waw.pl> References: <29109766.21396252576661.JavaMail.root@poczta.bursztynowski.waw.pl> <483894773.47742194.1396282960523.JavaMail.root@telenet.be> <1396286979.16858.2.camel@alpaga.bursztynowski.waw.pl> Message-ID: <745214964.53255792.1396557201175.JavaMail.root@telenet.be> hello Radek, ah yes. The thing with the chroot? I always forget to check that. I will do that as soon as possible. greetings, J. ----- Oorspronkelijk bericht ----- Van: "Radek Bursztynowski" Aan: "Support list for open source software in schools." Verzonden: Maandag 31 maart 2014 19:29:39 Onderwerp: Re: [K12OSN] usb not working Hello, Can you see any davice connected to USB port on thin client terminal? You can check it by tail -f /var/log/messages on local xterm or shell runs on thin client terminal. Regards, Radek ----- > hallo, > > thanks for the reaction. > > Yes, I added the users to the fuse group. > > Regards, j. > > ----- Oorspronkelijk bericht ----- > Van: "Radek Bursztynowski" > Aan: "Support list for open source software in schools." , "Support list for open source software in schools." > Verzonden: Maandag 31 maart 2014 09:56:16 > Onderwerp: Re: [K12OSN] usb not working > > Hi, > > Did you remebmer to add users to fuse group on the server? > > Regards, > Radek > > ------ > Dear All, > > since the last major change in k12Linux, I'm doing an alternative install because with the newest image, our thinclients do not work. > > So I configure dhcpd, tftp and so on, and copy over /opt/ltsp and /var/lib/tftpboot/ltsp > This works great, except I can't get usb to work. > > So how does usb on k12Linux actually work behind the scene ? > > Thanks for any advise. > > greetings, J. > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see ; > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see From radek at bursztynowski.waw.pl Thu Apr 3 22:12:04 2014 From: radek at bursztynowski.waw.pl (Radek Bursztynowski) Date: Fri, 4 Apr 2014 00:12:04 +0200 Subject: [K12OSN] usb not working In-Reply-To: <745214964.53255792.1396557201175.JavaMail.root@telenet.be> References: <29109766.21396252576661.JavaMail.root@poczta.bursztynowski.waw.pl> Message-ID: <1396563124.18775.0.camel@alpaga.bursztynowski.waw.pl> Johan, No think with chroot. I think about checking on thin client terminal, not on the server. If you define in lts.conf one of the free thin client terminal screen (let say 5) by SCREEN_05 = shell record, you can switch thin client using Ctrl+Alt+F5 to the shell on the thin client terminal. Then you can check is the usb device mounted. In my opinion it is important because within the first step thin client mounts usb device locally and within the next step there is the redirection to the server. If usb device is not mounted on the thin client terminal - check the module on thin client image (chroot) responsible for usb device file system. If there is no this module - install it using chroot mechanizm. If usb device is mounted on thin client terminal and you don't see this device on the server - there is the problem. Check rpm -q ltspfs on the server. If there is no ltspfs - install this package. You can find this package in LTSP repo. I think that it could be the reason of mounting troubles. Best regards, Radek > hello Radek, > > ah yes. The thing with the chroot? I always forget to check that. > I will do that as soon as possible. > > greetings, J. > > ----- Oorspronkelijk bericht ----- > Van: "Radek Bursztynowski" > Aan: "Support list for open source software in schools." > Verzonden: Maandag 31 maart 2014 19:29:39 > Onderwerp: Re: [K12OSN] usb not working > > Hello, > > Can you see any davice connected to USB port on thin client terminal? > You can check it by tail -f /var/log/messages on local xterm or shell > runs on thin client terminal. > > Regards, > Radek > > ----- > > > > hallo, > > > > thanks for the reaction. > > > > Yes, I added the users to the fuse group. > > > > Regards, j. > > > > ----- Oorspronkelijk bericht ----- > > Van: "Radek Bursztynowski" > > Aan: "Support list for open source software in schools." , "Support list for open source software in schools." > > Verzonden: Maandag 31 maart 2014 09:56:16 > > Onderwerp: Re: [K12OSN] usb not working > > > > Hi, > > > > Did you remebmer to add users to fuse group on the server? > > > > Regards, > > Radek > > > > ------ > > Dear All, > > > > since the last major change in k12Linux, I'm doing an alternative install because with the newest image, our thinclients do not work. > > > > So I configure dhcpd, tftp and so on, and copy over /opt/ltsp and /var/lib/tftpboot/ltsp > > This works great, except I can't get usb to work. > > > > So how does usb on k12Linux actually work behind the scene ? > > > > Thanks for any advise. > > > > greetings, J. > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see ; > > > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see > > > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From brcisna at eazylivin.net Sat Apr 5 11:01:41 2014 From: brcisna at eazylivin.net (Barry R Cisna) Date: Sat, 05 Apr 2014 06:01:41 -0500 Subject: [K12OSN] TC wifi initializtion - concept Message-ID: <1396695701.6574.46.camel@server1.eazylivin.net> Hello All, I asked this quite some time ago and didn't get any response so just throwing this idea out again. (Maybe there is a project doing what I would like to achieve with wifi on tc's and I do not know of it,also?) Situation,and hopefully a remedy: 1) At this point it seems that a big downfall of using tc's is no bootable wifi initialization & association possibility with any surrounding AP's in the boot process. 1a) Has anyone ever thought of incorporating an "option" by adding the code in the system bios to add a parameter for wifi association to "an" access point along with password if needed . This would be very much the same scenario as when we used to add an etherboot chip to the nics on tc's back in the day (before pxe became prevelant), only using the "native" bios chip rather than an add on chip. In other words to get away from having to actually have an add on,bootable media,onto any given TC system,,such as having to hang an usb stick or slap a cd in to achieve wifi initialization and association to a given AP. It just seems very odd with the length of time,that wifi has been a 'must have' on any network device that a TC vendor has not even come up with an spec for just this? Thanks, Barry From burke at thealmquists.net Sat Apr 5 16:39:34 2014 From: burke at thealmquists.net (Burke Almquist) Date: Sat, 5 Apr 2014 11:39:34 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: <1396695701.6574.46.camel@server1.eazylivin.net> References: <1396695701.6574.46.camel@server1.eazylivin.net> Message-ID: <9E46F517-EC8A-4975-9F78-4F731FA705D5@thealmquists.net> On Apr 5, 2014, at 6:01 AM, Barry R Cisna wrote: > Situation,and hopefully a remedy: > > 1) At this point it seems that a big downfall of using tc's is no > bootable wifi initialization & association possibility with any > surrounding AP's in the boot process. > > 1a) Has anyone ever thought of incorporating an "option" by adding the > code in the system bios to add a parameter for wifi association to "an" > access point along with password if needed . > > This would be very much the same scenario as when we used to add an > etherboot chip to the nics on tc's back in the day (before pxe became > prevelant), only using the "native" bios chip rather than an add on > chip. > > In other words to get away from having to actually have an add > on,bootable media,onto any given TC system,,such as having to hang an > usb stick or slap a cd in to achieve wifi initialization and > association to a given AP. > > It just seems very odd with the length of time,that wifi has been a > 'must have' on any network device that a TC vendor has not even come up > with an spec for just this? > > Thanks, > Barry > I think there are two reasons that something like PXE/etherboot for wireless hasn?t happened. 1. It?s more complicated to do than ether boot/PXE because of the additional security/authentication requirements, and also less likely to be used since portablity means it might not have a boot server available to boot from. 2. The nature of wireless connections means that the bandwidth is shared and performance can vary enormously, unlike a properly wired connection which is switched and almost always runs at the rated speed. Thus you are extremely limited in the number of thin clients you can run over such a connection. What I HAVE seen people do that don?t want to build large wired networks, is to use wireless to ethernet dongles for a small number of individual clients, or to use a cheep wifi router as a sort of bridge for multiple clients that can just plug into it wherever they set up. Using a router solves some signal strength and packet collision issues, though still not a good as having gigabit ethernet obviously. From microman at cmosnetworks.com Tue Apr 8 02:54:30 2014 From: microman at cmosnetworks.com (Terrell Prude' Jr.) Date: Mon, 07 Apr 2014 22:54:30 -0400 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: <1396695701.6574.46.camel@server1.eazylivin.net> References: <1396695701.6574.46.camel@server1.eazylivin.net> Message-ID: <534364E6.7020903@cmosnetworks.com> No way in heck I would want TC's on wireless. It's a hub. X11 screen updates can suck up to 73 Mbit/sec when playing TuxType (I did this measurement about 10 years ago and re-confirmed it about 5 years ago). As much as the wireless hype may be going on now, there's just no way an LTSP architecture can be successful on a hub, be that hub wireless or wired. Well, not if you intend to use the full complement of educational apps, that is. And don't even *think* about anything that uses Flash, for the same reason. LibreOffice'll work OK. So will Gedit or even Firefox, provided there aren't a lot of screen updates. But relatively static (as far as screen updates go) apps like that are about it. Putting LTSP on wireless is a recipe for failure, given current wireless technology. --TP On 04/05/2014 07:01 AM, Barry R Cisna wrote: > Hello All, > > I asked this quite some time ago and didn't get any response so just > throwing this idea out again. > (Maybe there is a project doing what I would like to achieve with wifi > on tc's and I do not know of it,also?) > > Situation,and hopefully a remedy: > > 1) At this point it seems that a big downfall of using tc's is no > bootable wifi initialization & association possibility with any > surrounding AP's in the boot process. > > 1a) Has anyone ever thought of incorporating an "option" by adding the > code in the system bios to add a parameter for wifi association to "an" > access point along with password if needed . > > This would be very much the same scenario as when we used to add an > etherboot chip to the nics on tc's back in the day (before pxe became > prevelant), only using the "native" bios chip rather than an add on > chip. > > In other words to get away from having to actually have an add > on,bootable media,onto any given TC system,,such as having to hang an > usb stick or slap a cd in to achieve wifi initialization and > association to a given AP. > > It just seems very odd with the length of time,that wifi has been a > 'must have' on any network device that a TC vendor has not even come up > with an spec for just this? > > Thanks, > Barry > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From news at siddall.name Tue Apr 8 14:43:09 2014 From: news at siddall.name (Jeff Siddall) Date: Tue, 08 Apr 2014 10:43:09 -0400 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: <534364E6.7020903@cmosnetworks.com> References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: <53440AFD.6020506@siddall.name> On 04/07/2014 10:54 PM, Terrell Prude' Jr. wrote: > No way in heck I would want TC's on wireless. It's a hub. X11 screen > updates can suck up to 73 Mbit/sec when playing TuxType (I did this > measurement about 10 years ago and re-confirmed it about 5 years ago). > As much as the wireless hype may be going on now, there's just no way an > LTSP architecture can be successful on a hub, be that hub wireless or > wired. Well, not if you intend to use the full complement of > educational apps, that is. And don't even *think* about anything that > uses Flash, for the same reason. > > LibreOffice'll work OK. So will Gedit or even Firefox, provided there > aren't a lot of screen updates. But relatively static (as far as screen > updates go) apps like that are about it. Putting LTSP on wireless is a > recipe for failure, given current wireless technology. I agree. Wireless is just not cut out for _general_ LTSP use. You might find some limited success for things like kiosks running everything as a localapp but the use cases are relatively niche. I have moved most of my clients to switched wired 1000 Mbps connections because even dedicated 100 M connections were too slow for some things. Also keep in mind that the apps that use a ton of bandwidth might not be obvious. LibreOffice Impress was one of the applications that was frequently unusable at 100 Mbps. Jeff From jim.kinney at gmail.com Tue Apr 8 14:52:53 2014 From: jim.kinney at gmail.com (Jim Kinney) Date: Tue, 8 Apr 2014 10:52:53 -0400 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: <53440AFD.6020506@siddall.name> References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> <53440AFD.6020506@siddall.name> Message-ID: Anything that uses the SDL toolkit for graphics is barely usable at 100Mbps and less than 1GHz on the client cpu. Tuxmath is barely usable with a 1Gbps network and a 1.5GHz client cpu. On Tue, Apr 8, 2014 at 10:43 AM, Jeff Siddall wrote: > On 04/07/2014 10:54 PM, Terrell Prude' Jr. wrote: > >> No way in heck I would want TC's on wireless. It's a hub. X11 screen >> updates can suck up to 73 Mbit/sec when playing TuxType (I did this >> measurement about 10 years ago and re-confirmed it about 5 years ago). >> As much as the wireless hype may be going on now, there's just no way an >> LTSP architecture can be successful on a hub, be that hub wireless or >> wired. Well, not if you intend to use the full complement of >> educational apps, that is. And don't even *think* about anything that >> uses Flash, for the same reason. >> >> LibreOffice'll work OK. So will Gedit or even Firefox, provided there >> aren't a lot of screen updates. But relatively static (as far as screen >> updates go) apps like that are about it. Putting LTSP on wireless is a >> recipe for failure, given current wireless technology. >> > > I agree. > > Wireless is just not cut out for _general_ LTSP use. You might find some > limited success for things like kiosks running everything as a localapp but > the use cases are relatively niche. > > I have moved most of my clients to switched wired 1000 Mbps connections > because even dedicated 100 M connections were too slow for some things. > Also keep in mind that the apps that use a ton of bandwidth might not be > obvious. LibreOffice Impress was one of the applications that was > frequently unusable at 100 Mbps. > > Jeff > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- -- James P. Kinney III Every time you stop a school, you will have to build a jail. What you gain at one end you lose at the other. It's like feeding a dog on his own tail. It won't fatten the dog. - Speech 11/23/1900 Mark Twain *http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From lesmikesell at gmail.com Tue Apr 8 15:13:44 2014 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 8 Apr 2014 10:13:44 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: <534364E6.7020903@cmosnetworks.com> References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: On Mon, Apr 7, 2014 at 9:54 PM, Terrell Prude' Jr. wrote: > No way in heck I would want TC's on wireless. It's a hub. X11 screen > updates can suck up to 73 Mbit/sec when playing TuxType (I did this > measurement about 10 years ago and re-confirmed it about 5 years ago). As > much as the wireless hype may be going on now, there's just no way an LTSP > architecture can be successful on a hub, be that hub wireless or wired. > Well, not if you intend to use the full complement of educational apps, that > is. And don't even *think* about anything that uses Flash, for the same > reason. > > LibreOffice'll work OK. So will Gedit or even Firefox, provided there > aren't a lot of screen updates. But relatively static (as far as screen > updates go) apps like that are about it. Putting LTSP on wireless is a > recipe for failure, given current wireless technology. I don't think you'd want to network-boot over wifi, but has anyone tried more than a few sessions of NX/freenx or x2go over wifi simultaneously? I use it myself from a laptop frequently and think it is great for everything short of streaming video but don't know how well it scales. If you aren't familiar with the NX or x2go technology, it adds proxy/caching/compression on both the server and client sides to greatly reduce latency and the overall bandwidth needed, and it has cross platform clients. Otherwise it appears as a normal X desktop, adding the option to disconnect from running sessions and reconnecting later with everything still running (even from a different client). -- Les Mikesell lesmikesell at gmail.com From lesmikesell at gmail.com Tue Apr 8 15:15:36 2014 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 8 Apr 2014 10:15:36 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> <53440AFD.6020506@siddall.name> Message-ID: On Tue, Apr 8, 2014 at 9:52 AM, Jim Kinney wrote: > Anything that uses the SDL toolkit for graphics is barely usable at 100Mbps > and less than 1GHz on the client cpu. Tuxmath is barely usable with a 1Gbps > network and a 1.5GHz client cpu. It would be interesting to test this with x2go to see if the screen refreshes end up being usefully cached. -- Les Mikesell lesmikesell at gmail.com From gnutbeam at gmail.com Tue Apr 8 16:08:19 2014 From: gnutbeam at gmail.com (Roger Nutbeam) Date: Tue, 8 Apr 2014 11:08:19 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: <534364E6.7020903@cmosnetworks.com> References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: I concur. Latency is just terrible on wireless. On Mon, Apr 7, 2014 at 9:54 PM, Terrell Prude' Jr. < microman at cmosnetworks.com> wrote: > No way in heck I would want TC's on wireless. It's a hub. X11 screen > updates can suck up to 73 Mbit/sec when playing TuxType (I did this > measurement about 10 years ago and re-confirmed it about 5 years ago). As > much as the wireless hype may be going on now, there's just no way an LTSP > architecture can be successful on a hub, be that hub wireless or wired. > Well, not if you intend to use the full complement of educational apps, > that is. And don't even *think* about anything that uses Flash, for the > same reason. > > LibreOffice'll work OK. So will Gedit or even Firefox, provided there > aren't a lot of screen updates. But relatively static (as far as screen > updates go) apps like that are about it. Putting LTSP on wireless is a > recipe for failure, given current wireless technology. > > --TP > > > On 04/05/2014 07:01 AM, Barry R Cisna wrote: > >> Hello All, >> >> I asked this quite some time ago and didn't get any response so just >> throwing this idea out again. >> (Maybe there is a project doing what I would like to achieve with wifi >> on tc's and I do not know of it,also?) >> >> Situation,and hopefully a remedy: >> >> 1) At this point it seems that a big downfall of using tc's is no >> bootable wifi initialization & association possibility with any >> surrounding AP's in the boot process. >> >> 1a) Has anyone ever thought of incorporating an "option" by adding the >> code in the system bios to add a parameter for wifi association to "an" >> access point along with password if needed . >> >> This would be very much the same scenario as when we used to add an >> etherboot chip to the nics on tc's back in the day (before pxe became >> prevelant), only using the "native" bios chip rather than an add on >> chip. >> >> In other words to get away from having to actually have an add >> on,bootable media,onto any given TC system,,such as having to hang an >> usb stick or slap a cd in to achieve wifi initialization and >> association to a given AP. >> >> It just seems very odd with the length of time,that wifi has been a >> 'must have' on any network device that a TC vendor has not even come up >> with an spec for just this? >> >> Thanks, >> Barry >> >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see >> > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gnutbeam at gmail.com Tue Apr 8 16:10:48 2014 From: gnutbeam at gmail.com (Roger Nutbeam) Date: Tue, 8 Apr 2014 11:10:48 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: I've have some clients on regular desktops that have to run X over wireless, and it just sucks. And this is only with a couple of desktops using wireless. On Tue, Apr 8, 2014 at 11:08 AM, Roger Nutbeam wrote: > I concur. Latency is just terrible on wireless. > > > On Mon, Apr 7, 2014 at 9:54 PM, Terrell Prude' Jr. < > microman at cmosnetworks.com> wrote: > >> No way in heck I would want TC's on wireless. It's a hub. X11 screen >> updates can suck up to 73 Mbit/sec when playing TuxType (I did this >> measurement about 10 years ago and re-confirmed it about 5 years ago). As >> much as the wireless hype may be going on now, there's just no way an LTSP >> architecture can be successful on a hub, be that hub wireless or wired. >> Well, not if you intend to use the full complement of educational apps, >> that is. And don't even *think* about anything that uses Flash, for the >> same reason. >> >> LibreOffice'll work OK. So will Gedit or even Firefox, provided there >> aren't a lot of screen updates. But relatively static (as far as screen >> updates go) apps like that are about it. Putting LTSP on wireless is a >> recipe for failure, given current wireless technology. >> >> --TP >> >> >> On 04/05/2014 07:01 AM, Barry R Cisna wrote: >> >>> Hello All, >>> >>> I asked this quite some time ago and didn't get any response so just >>> throwing this idea out again. >>> (Maybe there is a project doing what I would like to achieve with wifi >>> on tc's and I do not know of it,also?) >>> >>> Situation,and hopefully a remedy: >>> >>> 1) At this point it seems that a big downfall of using tc's is no >>> bootable wifi initialization & association possibility with any >>> surrounding AP's in the boot process. >>> >>> 1a) Has anyone ever thought of incorporating an "option" by adding the >>> code in the system bios to add a parameter for wifi association to "an" >>> access point along with password if needed . >>> >>> This would be very much the same scenario as when we used to add an >>> etherboot chip to the nics on tc's back in the day (before pxe became >>> prevelant), only using the "native" bios chip rather than an add on >>> chip. >>> >>> In other words to get away from having to actually have an add >>> on,bootable media,onto any given TC system,,such as having to hang an >>> usb stick or slap a cd in to achieve wifi initialization and >>> association to a given AP. >>> >>> It just seems very odd with the length of time,that wifi has been a >>> 'must have' on any network device that a TC vendor has not even come up >>> with an spec for just this? >>> >>> Thanks, >>> Barry >>> >>> >>> _______________________________________________ >>> K12OSN mailing list >>> K12OSN at redhat.com >>> https://www.redhat.com/mailman/listinfo/k12osn >>> For more info see >>> >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lesmikesell at gmail.com Tue Apr 8 16:22:53 2014 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 8 Apr 2014 11:22:53 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: On Tue, Apr 8, 2014 at 11:10 AM, Roger Nutbeam wrote: > I've have some clients on regular desktops that have to run X over wireless, > and it just sucks. And this is only with a couple of desktops using > wireless. > Have you ever tried x2go or the NX client (either with their server or freenx server-side? -- Les Mikesell lesmikesell at gmail.com From joseph.bishay at gmail.com Tue Apr 8 16:35:00 2014 From: joseph.bishay at gmail.com (Joseph Bishay) Date: Tue, 8 Apr 2014 12:35:00 -0400 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: Hello, I regularly use NX to connect to the LTSP server. I needed a way to allow teachers to connect from home / outside school that would work with high-latency wireless networks that they'd probably have at home. They install the NX client at home and I'm running FreeNX on the server and it works amazingly well. It's just as if they're sitting at school. It is fantastic. Joseph On Tue, Apr 8, 2014 at 12:22 PM, Les Mikesell wrote: > On Tue, Apr 8, 2014 at 11:10 AM, Roger Nutbeam wrote: >> I've have some clients on regular desktops that have to run X over wireless, >> and it just sucks. And this is only with a couple of desktops using >> wireless. >> > > Have you ever tried x2go or the NX client (either with their server or > freenx server-side? > > -- > Les Mikesell > lesmikesell at gmail.com > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From gnutbeam at gmail.com Tue Apr 8 16:30:11 2014 From: gnutbeam at gmail.com (Roger Nutbeam) Date: Tue, 8 Apr 2014 11:30:11 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: Freenx is a lot better performing than X over wireless. I haven't had the chance to test x2go but from what I've looked it so far, I like what I see. On Tue, Apr 8, 2014 at 11:22 AM, Les Mikesell wrote: > On Tue, Apr 8, 2014 at 11:10 AM, Roger Nutbeam wrote: > > I've have some clients on regular desktops that have to run X over > wireless, > > and it just sucks. And this is only with a couple of desktops using > > wireless. > > > > Have you ever tried x2go or the NX client (either with their server or > freenx server-side? > > -- > Les Mikesell > lesmikesell at gmail.com > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lesmikesell at gmail.com Tue Apr 8 16:50:50 2014 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 8 Apr 2014 11:50:50 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: On Tue, Apr 8, 2014 at 11:30 AM, Roger Nutbeam wrote: > Freenx is a lot better performing than X over wireless. I haven't had the > chance to test x2go but from what I've looked it so far, I like what I see. X2go uses some of the same nx libraries as freenx and unlike NX the clients are also open source. At least on CentOS6 freenx and x2go packages can co-exist if you want to try both, although they don't interoperate. X2go seems to have more active development now and seems pretty usable. I think a USB-boot into something that just opened an x2go window might be a practical thin client - but I haven't run a large number of clients against the same server. There is probably some overhead for the virtual sceen/cache/compression for each session. -- Les Mikesell lesmikesell at gmail.com From DLWillson at TheGeek.NU Tue Apr 8 17:13:01 2014 From: DLWillson at TheGeek.NU (David L. Willson) Date: Tue, 08 Apr 2014 11:13:01 -0600 (MDT) Subject: [K12OSN] auto-trailer In-Reply-To: Message-ID: The link at the very end of the K12OSN emails seems to be mis-pointed. Maybe it should be k12osn.org, not k12os.org? _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From gnutbeam at gmail.com Tue Apr 8 17:43:03 2014 From: gnutbeam at gmail.com (Roger Nutbeam) Date: Tue, 8 Apr 2014 12:43:03 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: You'd definitely want to run your outside<-in connections over a VPN too. I use OpenVPN and it works well. On Tue, Apr 8, 2014 at 11:50 AM, Les Mikesell wrote: > On Tue, Apr 8, 2014 at 11:30 AM, Roger Nutbeam wrote: > > Freenx is a lot better performing than X over wireless. I haven't had the > > chance to test x2go but from what I've looked it so far, I like what I > see. > > X2go uses some of the same nx libraries as freenx and unlike NX the > clients are also open source. At least on CentOS6 freenx and x2go > packages can co-exist if you want to try both, although they don't > interoperate. X2go seems to have more active development now and > seems pretty usable. I think a USB-boot into something that just > opened an x2go window might be a practical thin client - but I haven't > run a large number of clients against the same server. There is > probably some overhead for the virtual sceen/cache/compression for > each session. > > -- > Les Mikesell > lesmikesell at gmail.com > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From lesmikesell at gmail.com Tue Apr 8 17:52:37 2014 From: lesmikesell at gmail.com (Les Mikesell) Date: Tue, 8 Apr 2014 12:52:37 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: On Tue, Apr 8, 2014 at 12:43 PM, Roger Nutbeam wrote: > You'd definitely want to run your outside<-in connections over a VPN too. I > use OpenVPN and it works well. That's always a good idea - but both NX and x2go will use ssh for their connection and tunnel everything through it anyway. X2go will also transparently fire up a pass-through connection using rdp to a windows desktop if you need remote access and want the encryption/compression/caching features. -- Les Mikesell lesmikesell at gmail.com From gnutbeam at gmail.com Tue Apr 8 18:06:36 2014 From: gnutbeam at gmail.com (Roger Nutbeam) Date: Tue, 8 Apr 2014 13:06:36 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: I know both will use ssh tunneling, but I find a VPN easier to manage on a larger scale than ssh clients, keys and tunnels. It also doesn't give me the warm fuzzies forwarding ssh through my firewall to internal machines. On Tue, Apr 8, 2014 at 12:52 PM, Les Mikesell wrote: > On Tue, Apr 8, 2014 at 12:43 PM, Roger Nutbeam wrote: > > You'd definitely want to run your outside<-in connections over a VPN > too. I > > use OpenVPN and it works well. > > That's always a good idea - but both NX and x2go will use ssh for > their connection and tunnel everything through it anyway. X2go will > also transparently fire up a pass-through connection using rdp to a > windows desktop if you need remote access and want the > encryption/compression/caching features. > > -- > Les Mikesell > lesmikesell at gmail.com > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jim.kinney at gmail.com Tue Apr 8 19:05:57 2014 From: jim.kinney at gmail.com (Jim Kinney) Date: Tue, 8 Apr 2014 15:05:57 -0400 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: Same encryption algorithms on ssh and openvpn. Ssh through VPN is bloated and overkill for k12 needs. Can manage ssh keys with freeIPA on centos6 as ssh knows through Pam to check LDAP for user keys (slick!). Big issue is requiring password on ssh keys for users. On Apr 8, 2014 2:10 PM, "Roger Nutbeam" wrote: > I know both will use ssh tunneling, but I find a VPN easier to manage on a > larger scale than ssh clients, keys and tunnels. It also doesn't give me > the warm fuzzies forwarding ssh through my firewall to internal machines. > > > On Tue, Apr 8, 2014 at 12:52 PM, Les Mikesell wrote: > >> On Tue, Apr 8, 2014 at 12:43 PM, Roger Nutbeam >> wrote: >> > You'd definitely want to run your outside<-in connections over a VPN >> too. I >> > use OpenVPN and it works well. >> >> That's always a good idea - but both NX and x2go will use ssh for >> their connection and tunnel everything through it anyway. X2go will >> also transparently fire up a pass-through connection using rdp to a >> windows desktop if you need remote access and want the >> encryption/compression/caching features. >> >> -- >> Les Mikesell >> lesmikesell at gmail.com >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see >> > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From gnutbeam at gmail.com Tue Apr 8 19:41:22 2014 From: gnutbeam at gmail.com (Roger Nutbeam) Date: Tue, 8 Apr 2014 14:41:22 -0500 Subject: [K12OSN] TC wifi initializtion - concept In-Reply-To: References: <1396695701.6574.46.camel@server1.eazylivin.net> <534364E6.7020903@cmosnetworks.com> Message-ID: Well you can turn passwords off using ssh and just use the keys to authenticate. OpenSSH and OpenVPN use different ciphers. I don't think arcfour is even listed as a cipher option for OpenVPN. The protocols are different too, so you get different network behavior using them. You could choose not to use ssh tunneling on your clients if you're using a VPN, so that gets rid of the "messy". No point in an encrypted tunnel in an encrypted tunnel. Unless of course one is clinically paranoid. Of course using either ssh or a VPN is pretty secure as long as they're well implemented. I still don't like encrypted tunnels through my firewall that I can't inspect. On Tue, Apr 8, 2014 at 2:05 PM, Jim Kinney wrote: > Same encryption algorithms on ssh and openvpn. Ssh through VPN is bloated > and overkill for k12 needs. Can manage ssh keys with freeIPA on centos6 as > ssh knows through Pam to check LDAP for user keys (slick!). Big issue is > requiring password on ssh keys for users. > On Apr 8, 2014 2:10 PM, "Roger Nutbeam" wrote: > >> I know both will use ssh tunneling, but I find a VPN easier to manage on >> a larger scale than ssh clients, keys and tunnels. It also doesn't give me >> the warm fuzzies forwarding ssh through my firewall to internal machines. >> >> >> On Tue, Apr 8, 2014 at 12:52 PM, Les Mikesell wrote: >> >>> On Tue, Apr 8, 2014 at 12:43 PM, Roger Nutbeam >>> wrote: >>> > You'd definitely want to run your outside<-in connections over a VPN >>> too. I >>> > use OpenVPN and it works well. >>> >>> That's always a good idea - but both NX and x2go will use ssh for >>> their connection and tunnel everything through it anyway. X2go will >>> also transparently fire up a pass-through connection using rdp to a >>> windows desktop if you need remote access and want the >>> encryption/compression/caching features. >>> >>> -- >>> Les Mikesell >>> lesmikesell at gmail.com >>> >>> _______________________________________________ >>> K12OSN mailing list >>> K12OSN at redhat.com >>> https://www.redhat.com/mailman/listinfo/k12osn >>> For more info see >>> >> >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see >> > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -------------- next part -------------- An HTML attachment was scrubbed... URL: From radek at bursztynowski.waw.pl Wed Apr 16 11:50:14 2014 From: radek at bursztynowski.waw.pl (Radek Bursztynowski) Date: Wed, 16 Apr 2014 13:50:14 +0200 Subject: [K12OSN] Log off doesn end all user processes Message-ID: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> Hello, I have a problem with ending on server thin client LTSP session. When I log out from the server session I return to the LDM, but there are not closed my user processes on the server. After GNOME log out there are: [root at tryandbuy pxelinux.cfg]# ps aux | grep radek radek 10790 0.0 0.0 20040 680 ? S 13:22 0:00 dbus-launch --sh-syntax --exit-with-session radek 10791 0.0 0.0 21816 1236 ? Ss 13:22 0:00 /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session radek 10858 0.0 0.0 232924 6588 ? Ss 13:22 0:00 /usr/bin/seahorse-agent --variables radek 10862 0.0 0.0 137360 2152 ? S 13:22 0:00 /usr/libexec/gvfsd radek 10876 0.0 0.0 133864 5808 ? S 13:22 0:00 /usr/libexec/gconfd-2 radek 10906 0.0 0.0 24752 2032 ? S 13:22 0:00 /usr/lib64/xfce4/xfconf/xfconfd radek 10929 0.0 0.0 238448 5640 ? S 13:22 0:00 Thunar --daemon radek 10936 0.0 0.0 242892 2900 ? Ss 13:22 0:00 gnome-screensaver radek 10943 0.0 0.0 328644 10784 ? Sl 13:22 0:00 /usr/libexec/polkit-gnome-authentication-agent-1 radek 10946 0.0 0.0 508884 10908 ? S 13:22 0:00 gnome-volume-control-applet radek 10952 0.0 0.0 115096 4052 ? S 13:22 0:00 /usr/libexec/im-settings-daemon radek 10975 0.0 0.0 251984 6792 ? S 13:22 0:00 gnome-power-manager radek 10978 0.0 0.1 303344 18736 ? S 13:22 0:00 python /usr/share/system-config-printer/applet.py radek 10986 0.0 0.0 362372 9840 ? S 13:22 0:00 gpk-update-icon radek 10997 0.0 0.0 358308 6872 ? S 13:22 0:00 /usr/libexec/evolution/2.32/evolution-alarm-notify radek 11119 0.0 0.0 40900 2232 ? S 13:22 0:00 /usr/libexec/gconf-im-settings-daemon radek 11120 0.0 0.0 201512 2528 ? S 13:22 0:00 xfsettingsd radek 11140 0.0 0.0 148776 3532 ? S 13:22 0:00 /usr/libexec/gvfs-gdu-volume-monitor radek 11144 0.0 0.0 232436 2064 ? Sl 13:22 0:00 /usr/libexec/gvfs-afc-volume-monitor radek 11147 0.0 0.0 150892 2196 ? S 13:22 0:00 /usr/libexec/gvfs-gphoto2-volume-monitor radek 11149 0.0 0.0 146304 3228 ? S 13:22 0:00 /usr/libexec/gvfsd-trash --spawner :1.2 /org/gtk/gvfs/exec_spaw/0 radek 11297 0.0 0.0 356628 10964 ? S 13:28 0:00 /usr/libexec/notification-daemon After KDE log out: [root at tryandbuy pxelinux.cfg]# ps aux | grep radek [root at tryandbuy pxelinux.cfg]# ps aux | grep radek radek 11832 0.0 0.0 20040 692 ? S 13:40 0:00 dbus-launch --sh-syntax --exit-with-session radek 11833 0.0 0.0 21668 1208 ? Ss 13:40 0:00 /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session radek 11900 0.0 0.0 235144 6708 ? Ss 13:40 0:00 /usr/bin/seahorse-agent --variables radek 11908 0.0 0.0 137360 2104 ? S 13:40 0:00 /usr/libexec/gvfsd radek 11914 0.0 0.0 133848 5768 ? S 13:40 0:00 /usr/libexec/gconfd-2 radek 12102 0.0 0.0 115096 4052 ? S 13:40 0:00 /usr/libexec/im-settings-daemon radek 12258 0.0 0.0 40900 2232 ? S 13:40 0:00 /usr/libexec/gconf-im-settings-daemon Could anybody advice me how to close all processes within log out? Best regards, Radek -------------- next part -------------- An HTML attachment was scrubbed... URL: From jim.kinney at gmail.com Wed Apr 16 12:58:42 2014 From: jim.kinney at gmail.com (Jim Kinney) Date: Wed, 16 Apr 2014 08:58:42 -0400 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> Message-ID: All of those _SHOULD_ close when you log out. You can add a script that does a kill -9 on all the PIDs with your UID as owner. However, those processes can hang around if a session crashes. They may be stale. I also noticed you have a Thunar process. That's not part of normal gnome so it may be holding open the gnome stuff. run this: ps aux | grep radek | awk '{print $2}' | xargs kill -9 That will kill all the processes. Now launch a TC session and log right back out, wait a minute and then ps aux | grep radek Which base server OS are you running? On Wed, Apr 16, 2014 at 7:50 AM, Radek Bursztynowski < radek at bursztynowski.waw.pl> wrote: > Hello, > > I have a problem with ending on server thin client LTSP session. When I > log out from the server session I return to the LDM, but there are not > closed my user processes on the server. After GNOME log out there are: > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > radek 10790 0.0 0.0 20040 680 ? S 13:22 0:00 > dbus-launch --sh-syntax --exit-with-session > radek 10791 0.0 0.0 21816 1236 ? Ss 13:22 0:00 > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > radek 10858 0.0 0.0 232924 6588 ? Ss 13:22 0:00 > /usr/bin/seahorse-agent --variables > radek 10862 0.0 0.0 137360 2152 ? S 13:22 0:00 > /usr/libexec/gvfsd > radek 10876 0.0 0.0 133864 5808 ? S 13:22 0:00 > /usr/libexec/gconfd-2 > radek 10906 0.0 0.0 24752 2032 ? S 13:22 0:00 > /usr/lib64/xfce4/xfconf/xfconfd > radek 10929 0.0 0.0 238448 5640 ? S 13:22 0:00 Thunar > --daemon > radek 10936 0.0 0.0 242892 2900 ? Ss 13:22 0:00 > gnome-screensaver > radek 10943 0.0 0.0 328644 10784 ? Sl 13:22 0:00 > /usr/libexec/polkit-gnome-authentication-agent-1 > radek 10946 0.0 0.0 508884 10908 ? S 13:22 0:00 > gnome-volume-control-applet > radek 10952 0.0 0.0 115096 4052 ? S 13:22 0:00 > /usr/libexec/im-settings-daemon > radek 10975 0.0 0.0 251984 6792 ? S 13:22 0:00 > gnome-power-manager > radek 10978 0.0 0.1 303344 18736 ? S 13:22 0:00 python > /usr/share/system-config-printer/applet.py > radek 10986 0.0 0.0 362372 9840 ? S 13:22 0:00 > gpk-update-icon > radek 10997 0.0 0.0 358308 6872 ? S 13:22 0:00 > /usr/libexec/evolution/2.32/evolution-alarm-notify > radek 11119 0.0 0.0 40900 2232 ? S 13:22 0:00 > /usr/libexec/gconf-im-settings-daemon > radek 11120 0.0 0.0 201512 2528 ? S 13:22 0:00 > xfsettingsd > radek 11140 0.0 0.0 148776 3532 ? S 13:22 0:00 > /usr/libexec/gvfs-gdu-volume-monitor > radek 11144 0.0 0.0 232436 2064 ? Sl 13:22 0:00 > /usr/libexec/gvfs-afc-volume-monitor > radek 11147 0.0 0.0 150892 2196 ? S 13:22 0:00 > /usr/libexec/gvfs-gphoto2-volume-monitor > radek 11149 0.0 0.0 146304 3228 ? S 13:22 0:00 > /usr/libexec/gvfsd-trash --spawner :1.2 /org/gtk/gvfs/exec_spaw/0 > radek 11297 0.0 0.0 356628 10964 ? S 13:28 0:00 > /usr/libexec/notification-daemon > > After KDE log out: > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > radek 11832 0.0 0.0 20040 692 ? S 13:40 0:00 > dbus-launch --sh-syntax --exit-with-session > radek 11833 0.0 0.0 21668 1208 ? Ss 13:40 0:00 > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > radek 11900 0.0 0.0 235144 6708 ? Ss 13:40 0:00 > /usr/bin/seahorse-agent --variables > radek 11908 0.0 0.0 137360 2104 ? S 13:40 0:00 > /usr/libexec/gvfsd > radek 11914 0.0 0.0 133848 5768 ? S 13:40 0:00 > /usr/libexec/gconfd-2 > radek 12102 0.0 0.0 115096 4052 ? S 13:40 0:00 > /usr/libexec/im-settings-daemon > radek 12258 0.0 0.0 40900 2232 ? S 13:40 0:00 > /usr/libexec/gconf-im-settings-daemon > > Could anybody advice me how to close all processes within log out? > > Best regards, > Radek > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- -- James P. Kinney III Every time you stop a school, you will have to build a jail. What you gain at one end you lose at the other. It's like feeding a dog on his own tail. It won't fatten the dog. - Speech 11/23/1900 Mark Twain *http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From jomegat at jomegat.com Wed Apr 16 13:06:26 2014 From: jomegat at jomegat.com (jomegat) Date: Wed, 16 Apr 2014 09:06:26 -0400 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> Message-ID: <534E8052.2030803@jomegat.com> On 04/16/2014 08:58 AM, Jim Kinney wrote: > run this: > > ps aux | grep radek | awk '{print $2}' | xargs kill -9 Here's an easier option: pkill -u radek Also, pkill -u radek --signal 9 -- Jim Thomas (j omegat t) It's easier to outsmart people than it is to outdumb them. - Scott Meyers From jim.kinney at gmail.com Wed Apr 16 13:10:41 2014 From: jim.kinney at gmail.com (Jim Kinney) Date: Wed, 16 Apr 2014 09:10:41 -0400 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: <534E8052.2030803@jomegat.com> References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> <534E8052.2030803@jomegat.com> Message-ID: nice! New tool. On Wed, Apr 16, 2014 at 9:06 AM, jomegat wrote: > On 04/16/2014 08:58 AM, Jim Kinney wrote: > >> run this: >> >> ps aux | grep radek | awk '{print $2}' | xargs kill -9 >> > > Here's an easier option: > > pkill -u radek > > Also, > > pkill -u radek --signal 9 > > -- > Jim Thomas (j omegat t) > It's easier to outsmart people than it is to outdumb them. > - Scott Meyers > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- -- James P. Kinney III Every time you stop a school, you will have to build a jail. What you gain at one end you lose at the other. It's like feeding a dog on his own tail. It won't fatten the dog. - Speech 11/23/1900 Mark Twain *http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From veli-matti.lintu at opinsys.fi Wed Apr 16 13:18:07 2014 From: veli-matti.lintu at opinsys.fi (Veli-Matti Lintu) Date: Wed, 16 Apr 2014 13:18:07 +0000 (UTC) Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> Message-ID: <215249773.5557.1397654287308.JavaMail.zimbra@opinsys.fi> Hello, There's a tool called xexit that you could try. It triggers a script to clean up user's processes when the session has died for some reason or X does not respond anymore. There's more here: http://labs.opinsys.com/blog/2011/03/22/killing-user-processes-with-xexit-on-ltsp-servers/ Happy hacking, Veli-Matti ----- Original Message ----- > From: "Radek Bursztynowski" > To: k12osn at redhat.com > Sent: Wednesday, April 16, 2014 2:50:14 PM > Subject: [K12OSN] Log off doesn end all user processes > > Hello, > > I have a problem with ending on server thin client LTSP session. When I log > out from the server session I return to the LDM, but there are not closed my > user processes on the server. After GNOME log out there are: > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > radek 10790 0.0 0.0 20040 680 ? S 13:22 0:00 dbus-launch --sh-syntax > --exit-with-session > radek 10791 0.0 0.0 21816 1236 ? Ss 13:22 0:00 /bin/dbus-daemon --fork > --print-pid 5 --print-address 7 --session > radek 10858 0.0 0.0 232924 6588 ? Ss 13:22 0:00 /usr/bin/seahorse-agent > --variables > radek 10862 0.0 0.0 137360 2152 ? S 13:22 0:00 /usr/libexec/gvfsd > radek 10876 0.0 0.0 133864 5808 ? S 13:22 0:00 /usr/libexec/gconfd-2 > radek 10906 0.0 0.0 24752 2032 ? S 13:22 0:00 /usr/lib64/xfce4/xfconf/xfconfd > radek 10929 0.0 0.0 238448 5640 ? S 13:22 0:00 Thunar --daemon > radek 10936 0.0 0.0 242892 2900 ? Ss 13:22 0:00 gnome-screensaver > radek 10943 0.0 0.0 328644 10784 ? Sl 13:22 0:00 > /usr/libexec/polkit-gnome-authentication-agent-1 > radek 10946 0.0 0.0 508884 10908 ? S 13:22 0:00 gnome-volume-control-applet > radek 10952 0.0 0.0 115096 4052 ? S 13:22 0:00 > /usr/libexec/im-settings-daemon > radek 10975 0.0 0.0 251984 6792 ? S 13:22 0:00 gnome-power-manager > radek 10978 0.0 0.1 303344 18736 ? S 13:22 0:00 python > /usr/share/system-config-printer/applet.py > radek 10986 0.0 0.0 362372 9840 ? S 13:22 0:00 gpk-update-icon > radek 10997 0.0 0.0 358308 6872 ? S 13:22 0:00 > /usr/libexec/evolution/2.32/evolution-alarm-notify > radek 11119 0.0 0.0 40900 2232 ? S 13:22 0:00 > /usr/libexec/gconf-im-settings-daemon > radek 11120 0.0 0.0 201512 2528 ? S 13:22 0:00 xfsettingsd > radek 11140 0.0 0.0 148776 3532 ? S 13:22 0:00 > /usr/libexec/gvfs-gdu-volume-monitor > radek 11144 0.0 0.0 232436 2064 ? Sl 13:22 0:00 > /usr/libexec/gvfs-afc-volume-monitor > radek 11147 0.0 0.0 150892 2196 ? S 13:22 0:00 > /usr/libexec/gvfs-gphoto2-volume-monitor > radek 11149 0.0 0.0 146304 3228 ? S 13:22 0:00 /usr/libexec/gvfsd-trash > --spawner :1.2 /org/gtk/gvfs/exec_spaw/0 > radek 11297 0.0 0.0 356628 10964 ? S 13:28 0:00 > /usr/libexec/notification-daemon > > After KDE log out: > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > radek 11832 0.0 0.0 20040 692 ? S 13:40 0:00 dbus-launch --sh-syntax > --exit-with-session > radek 11833 0.0 0.0 21668 1208 ? Ss 13:40 0:00 /bin/dbus-daemon --fork > --print-pid 5 --print-address 7 --session > radek 11900 0.0 0.0 235144 6708 ? Ss 13:40 0:00 /usr/bin/seahorse-agent > --variables > radek 11908 0.0 0.0 137360 2104 ? S 13:40 0:00 /usr/libexec/gvfsd > radek 11914 0.0 0.0 133848 5768 ? S 13:40 0:00 /usr/libexec/gconfd-2 > radek 12102 0.0 0.0 115096 4052 ? S 13:40 0:00 > /usr/libexec/im-settings-daemon > radek 12258 0.0 0.0 40900 2232 ? S 13:40 0:00 > /usr/libexec/gconf-im-settings-daemon > > Could anybody advice me how to close all processes within log out? > > Best regards, > Radek > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From jomegat at jomegat.com Wed Apr 16 13:20:27 2014 From: jomegat at jomegat.com (jomegat) Date: Wed, 16 Apr 2014 09:20:27 -0400 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> <534E8052.2030803@jomegat.com> Message-ID: <534E839B.7050807@jomegat.com> On 04/16/2014 09:10 AM, Jim Kinney wrote: > nice! New tool. It is a nice tool. Unfortunately, using it in this case treats the symptom rather than the disease. But sometimes that's what you need to do. > > > On Wed, Apr 16, 2014 at 9:06 AM, jomegat > wrote: > > On 04/16/2014 08:58 AM, Jim Kinney wrote: > > run this: > > ps aux | grep radek | awk '{print $2}' | xargs kill -9 > > > Here's an easier option: > > pkill -u radek > > Also, > > pkill -u radek --signal 9 > > -- > Jim Thomas (j omegat t) > It's easier to outsmart people than it is to outdumb them. > - Scott Meyers > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > > > > -- > -- > James P. Kinney III > //// > ////Every time you stop a school, you will have to build a jail. What > you gain at one end you lose at the other. It's like feeding a dog on > his own tail. It won't fatten the dog. > - Speech 11/23/1900 Mark Twain > //// > http://heretothereideas.blogspot.com/ > //// > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see -- Jim Thomas (j omegat t) It's easier to outsmart people than it is to outdumb them. - Scott Meyers -------------- next part -------------- An HTML attachment was scrubbed... URL: From radek at bursztynowski.waw.pl Wed Apr 16 15:07:19 2014 From: radek at bursztynowski.waw.pl (Radek Bursztynowski) Date: Wed, 16 Apr 2014 17:07:19 +0200 Subject: [K12OSN] Log off doesn end all user processes Message-ID: <20042002.381397660839420.JavaMail.root@poczta.bursztynowski.waw.pl> Many thanks to all for advices. Yes, I use pkill -U $USER for loging out LTSP client from the server. Every time I check is the user is loged in from thin client or from the server directly and for LTSP session I use pkill, for directed session standard log out GNOME, XFCE or KDE program. It works fine with Scientific Linux 6.1 or CentOS 6.4 thin clent image. But I have several old machines which have to use old thin client image (Fedora 11 - i586). And pkill -U $USER with Fedora 11 thin client image kills all user processes but LDM on thin client couldn't refresh and I can see black screen only. When I use Ctrl+Alt+Backspace keys everything back to the norm and LDM reloads. It isn't clear solution for the users who use Fedora 11 thin client image with on old machines. Let me add that when I log out using standard GNOME/XFCE/KDE logging out program on Fedora 11 image, LDM reloads well, but few user processes still run. This is my real problem. On the margin I noticed that pkill -U $USER doesn't umount breaks link with /home/$USER/.gvf. It breajs tar user home directory for simple bacakup. So, the first I umount /home/$USER/gvfs and the next is pkill. Many thanks again for help, perhaps something else? Radek ----- Original Message ----- From: jomegat Sent: Wed, 4/16/2014 3:20pm To: "Support list for open source software in schools." Subject: Re: [K12OSN] Log off doesn end all user processes On 04/16/2014 09:10 AM, Jim Kinney wrote: nice! New tool. It is a nice tool. Unfortunately, using it in this case treats the symptom rather than the disease. But sometimes that's what you need to do. On Wed, Apr 16, 2014 at 9:06 AM, jomegat wrote: On 04/16/2014 08:58 AM, Jim Kinney wrote: run this: ps aux | grep radek | awk '{print $2}' | xargs kill -9 Here's an easier option: pkill -u radek Also, pkill -u radek --signal 9 -- Jim Thomas (j omegat t) It's easier to outsmart people than it is to outdumb them. - Scott Meyers _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see -- -- James P. Kinney III Every time you stop a school, you will have to build a jail. What you gain at one end you lose at the other. It's like feeding a dog on his own tail. It won't fatten the dog. - Speech 11/23/1900 Mark Twain http://heretothereideas.blogspot.com/ _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see -- Jim Thomas (j omegat t) It's easier to outsmart people than it is to outdumb them. - Scott Meyers _______________________________________________ K12OSN mailing list K12OSN at redhat.com https://www.redhat.com/mailman/listinfo/k12osn For more info see ; From emil.krotki at ekrotech.com Wed Apr 16 18:33:36 2014 From: emil.krotki at ekrotech.com (emil.krotki at ekrotech.com) Date: Wed, 16 Apr 2014 11:33:36 -0700 Subject: [K12OSN] Log off doesn end all user processes Message-ID: <20140416113335.05d394f8f702b6ef793b0ec8fcf7c622.dba887164e.wbe@email07.europe.secureserver.net> An HTML attachment was scrubbed... URL: From radek at bursztynowski.waw.pl Wed Apr 16 20:36:43 2014 From: radek at bursztynowski.waw.pl (Radek Bursztynowski) Date: Wed, 16 Apr 2014 22:36:43 +0200 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: <20140416113335.05d394f8f702b6ef793b0ec8fcf7c622.dba887164e.wbe@email07.europe.secureserver.net> References: <20140416113335.05d394f8f702b6ef793b0ec8fcf7c622.dba887164e.wbe@email07.europe.secureserver.net> Message-ID: <1397680603.28170.7.camel@alpaga.bursztynowski.waw.pl> Emil, Thanks for explanation. Let me add my feelings. In my opinion newer thin client images (Scientific Linux 6.1, CentOS 6.4) reload LDM spontaneously when active session between thin client and the server is broken. Owing to that pkill -U $USER works fine. Fedora 11 thin image doesn't reload LDM spontaneously. So, I think the solution is on the thin client image, not on the server site. Pozdrawiam, Radek > Killing processes using user name is just part of the solution. > > > Any Linux, process has a parent process, and if the parent process > dies, child process should get terminated too. > Key word: "should". > Problem arises if client process does not intercept the termination of > the parent process and stll runs, or is already dis-joined from the > group with the parent while going background, or created as parallel. > > > In case parent process dies and child does not end, "init" overtakes > parentship of the "orphaned" process, and you may need to search for > that processes to kill explicitly by name, but you need to know their > names (PID is not the good criteria). > > > IMHO all LDM created processes should NOT dis-join that group and > should react on the signal about parent death, ceasing to execute any > longer. > Again: "should". > > > Well, it may mean, that LTSP (LDM and childs) may need some > code-review focusing on process parentship handling at > creation/termination... > > > Kind regards > Emil Krotki > EkroTech > Mobile: +420 775 314 445 > > > > -------- Original Message -------- > Subject: Re: [K12OSN] Log off doesn end all user processes > From: Radek Bursztynowski > Date: Wed, April 16, 2014 4:07 pm > To: "Support list for open source software in schools." > , "Support list for open source software in > schools." > > Many thanks to all for advices. > Yes, I use pkill -U $USER for loging out LTSP client from the > server. > Every time I check is the user is loged in from thin client or > from the server directly and for LTSP session I use pkill, for > directed session standard log out GNOME, XFCE or KDE program. > It works fine with Scientific Linux 6.1 or CentOS 6.4 thin > clent image. But I have several old machines which have to use > old thin client image (Fedora 11 - i586). And pkill -U $USER > with Fedora 11 thin client image kills all user processes but > LDM on thin client couldn't refresh and I can see black screen > only. When I use Ctrl+Alt+Backspace keys everything back to > the norm and LDM reloads. It isn't clear solution for the > users who use Fedora 11 thin client image with on old > machines. > > Let me add that when I log out using standard GNOME/XFCE/KDE > logging out program on Fedora 11 image, LDM reloads well, but > few user processes still run. > > This is my real problem. > > On the margin I noticed that pkill -U $USER doesn't umount > breaks link with /home/$USER/.gvf. It breajs tar user home > directory for simple bacakup. So, the first I > umount /home/$USER/gvfs and the next is pkill. > > Many thanks again for help, perhaps something else? > > Radek > > > > ----- Original Message ----- > From: jomegat > Sent: Wed, 4/16/2014 3:20pm > To: "Support list for open source software in schools." > > Subject: Re: [K12OSN] Log off doesn end all user processes > > > On 04/16/2014 09:10 AM, Jim Kinney > wrote: > > > > nice! New tool. > > > > > > It is a nice tool. Unfortunately, using it in this case treats > the > symptom rather than the disease. But sometimes that's what you > need > to do. > > > > > > > > > On Wed, Apr 16, 2014 at 9:06 AM, > jomegat > wrote: > > > On 04/16/2014 08:58 AM, Jim Kinney wrote: > > > run this: > > > > ps aux | grep radek | awk '{print $2}' | xargs kill -9 > > > > > > Here's an easier option: > > > > pkill -u radek > > > > Also, > > > > pkill -u radek --signal 9 > > > > -- > > Jim Thomas (j omegat t) > > It's easier to outsmart people than it is to outdumb > them. > > - Scott > Meyers > > > > > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see > > > > > > > > > > > -- > > -- > > James P. Kinney III > > > > Every time you stop a school, you will > have to build a jail. What you gain at one end you lose at the > other. It's like feeding a dog on his own tail. It won't > fatten the dog. > > - Speech 11/23/1900 Mark Twain > > > > http://heretothereideas.blogspot.com/ > > > > > > > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > > > > > -- > Jim Thomas (j omegat t) > It's easier to outsmart people than it is to outdumb them. > - Scott Meyers > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see ; > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From joseph.bishay at gmail.com Thu Apr 17 17:41:52 2014 From: joseph.bishay at gmail.com (Joseph Bishay) Date: Thu, 17 Apr 2014 13:41:52 -0400 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> Message-ID: Hello, I have this same problem and I found the 'slay' command works perfectly to resolve the leftover processes. # slay username Plus it's fun to tell the others that I had to 'slay' so-and-so :) Info: http://linuxpoison.blogspot.ca/2011/01/kills-all-of-users-processes-slay.html Joseph On Wed, Apr 16, 2014 at 7:50 AM, Radek Bursztynowski wrote: > Hello, > > I have a problem with ending on server thin client LTSP session. When I log > out from the server session I return to the LDM, but there are not closed my > user processes on the server. After GNOME log out there are: > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > radek 10790 0.0 0.0 20040 680 ? S 13:22 0:00 dbus-launch > --sh-syntax --exit-with-session > radek 10791 0.0 0.0 21816 1236 ? Ss 13:22 0:00 > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > radek 10858 0.0 0.0 232924 6588 ? Ss 13:22 0:00 > /usr/bin/seahorse-agent --variables > radek 10862 0.0 0.0 137360 2152 ? S 13:22 0:00 > /usr/libexec/gvfsd > radek 10876 0.0 0.0 133864 5808 ? S 13:22 0:00 > /usr/libexec/gconfd-2 > radek 10906 0.0 0.0 24752 2032 ? S 13:22 0:00 > /usr/lib64/xfce4/xfconf/xfconfd > radek 10929 0.0 0.0 238448 5640 ? S 13:22 0:00 Thunar > --daemon > radek 10936 0.0 0.0 242892 2900 ? Ss 13:22 0:00 > gnome-screensaver > radek 10943 0.0 0.0 328644 10784 ? Sl 13:22 0:00 > /usr/libexec/polkit-gnome-authentication-agent-1 > radek 10946 0.0 0.0 508884 10908 ? S 13:22 0:00 > gnome-volume-control-applet > radek 10952 0.0 0.0 115096 4052 ? S 13:22 0:00 > /usr/libexec/im-settings-daemon > radek 10975 0.0 0.0 251984 6792 ? S 13:22 0:00 > gnome-power-manager > radek 10978 0.0 0.1 303344 18736 ? S 13:22 0:00 python > /usr/share/system-config-printer/applet.py > radek 10986 0.0 0.0 362372 9840 ? S 13:22 0:00 > gpk-update-icon > radek 10997 0.0 0.0 358308 6872 ? S 13:22 0:00 > /usr/libexec/evolution/2.32/evolution-alarm-notify > radek 11119 0.0 0.0 40900 2232 ? S 13:22 0:00 > /usr/libexec/gconf-im-settings-daemon > radek 11120 0.0 0.0 201512 2528 ? S 13:22 0:00 xfsettingsd > radek 11140 0.0 0.0 148776 3532 ? S 13:22 0:00 > /usr/libexec/gvfs-gdu-volume-monitor > radek 11144 0.0 0.0 232436 2064 ? Sl 13:22 0:00 > /usr/libexec/gvfs-afc-volume-monitor > radek 11147 0.0 0.0 150892 2196 ? S 13:22 0:00 > /usr/libexec/gvfs-gphoto2-volume-monitor > radek 11149 0.0 0.0 146304 3228 ? S 13:22 0:00 > /usr/libexec/gvfsd-trash --spawner :1.2 /org/gtk/gvfs/exec_spaw/0 > radek 11297 0.0 0.0 356628 10964 ? S 13:28 0:00 > /usr/libexec/notification-daemon > > After KDE log out: > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > radek 11832 0.0 0.0 20040 692 ? S 13:40 0:00 dbus-launch > --sh-syntax --exit-with-session > radek 11833 0.0 0.0 21668 1208 ? Ss 13:40 0:00 > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > radek 11900 0.0 0.0 235144 6708 ? Ss 13:40 0:00 > /usr/bin/seahorse-agent --variables > radek 11908 0.0 0.0 137360 2104 ? S 13:40 0:00 > /usr/libexec/gvfsd > radek 11914 0.0 0.0 133848 5768 ? S 13:40 0:00 > /usr/libexec/gconfd-2 > radek 12102 0.0 0.0 115096 4052 ? S 13:40 0:00 > /usr/libexec/im-settings-daemon > radek 12258 0.0 0.0 40900 2232 ? S 13:40 0:00 > /usr/libexec/gconf-im-settings-daemon > > Could anybody advice me how to close all processes within log out? > > Best regards, > Radek > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From radek at bursztynowski.waw.pl Thu Apr 17 17:56:26 2014 From: radek at bursztynowski.waw.pl (Radek Bursztynowski) Date: Thu, 17 Apr 2014 19:56:26 +0200 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> Message-ID: <1397757386.31997.6.camel@alpaga.bursztynowski.waw.pl> Hello, Many thanks for advice. I will test it tomorrow . Coming back to this problem with Fedora 11 thin client image I keep conviction that I have to do something on thin client site. After killing all user processes on server site I can see on thin client working one user process (socket username at server_ip_address. When I kill this proces on the terminal LDM reloads and everything comes back to the norm. So I must to do something with Fedora 11 thin client image. Best regards, Radek > Hello, > > I have this same problem and I found the 'slay' command works > perfectly to resolve the leftover processes. > > # slay username > > Plus it's fun to tell the others that I had to 'slay' so-and-so :) > > Info: > http://linuxpoison.blogspot.ca/2011/01/kills-all-of-users-processes-slay.html > > Joseph > > On Wed, Apr 16, 2014 at 7:50 AM, Radek Bursztynowski > wrote: > > Hello, > > > > I have a problem with ending on server thin client LTSP session. When I log > > out from the server session I return to the LDM, but there are not closed my > > user processes on the server. After GNOME log out there are: > > > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > radek 10790 0.0 0.0 20040 680 ? S 13:22 0:00 dbus-launch > > --sh-syntax --exit-with-session > > radek 10791 0.0 0.0 21816 1236 ? Ss 13:22 0:00 > > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > > radek 10858 0.0 0.0 232924 6588 ? Ss 13:22 0:00 > > /usr/bin/seahorse-agent --variables > > radek 10862 0.0 0.0 137360 2152 ? S 13:22 0:00 > > /usr/libexec/gvfsd > > radek 10876 0.0 0.0 133864 5808 ? S 13:22 0:00 > > /usr/libexec/gconfd-2 > > radek 10906 0.0 0.0 24752 2032 ? S 13:22 0:00 > > /usr/lib64/xfce4/xfconf/xfconfd > > radek 10929 0.0 0.0 238448 5640 ? S 13:22 0:00 Thunar > > --daemon > > radek 10936 0.0 0.0 242892 2900 ? Ss 13:22 0:00 > > gnome-screensaver > > radek 10943 0.0 0.0 328644 10784 ? Sl 13:22 0:00 > > /usr/libexec/polkit-gnome-authentication-agent-1 > > radek 10946 0.0 0.0 508884 10908 ? S 13:22 0:00 > > gnome-volume-control-applet > > radek 10952 0.0 0.0 115096 4052 ? S 13:22 0:00 > > /usr/libexec/im-settings-daemon > > radek 10975 0.0 0.0 251984 6792 ? S 13:22 0:00 > > gnome-power-manager > > radek 10978 0.0 0.1 303344 18736 ? S 13:22 0:00 python > > /usr/share/system-config-printer/applet.py > > radek 10986 0.0 0.0 362372 9840 ? S 13:22 0:00 > > gpk-update-icon > > radek 10997 0.0 0.0 358308 6872 ? S 13:22 0:00 > > /usr/libexec/evolution/2.32/evolution-alarm-notify > > radek 11119 0.0 0.0 40900 2232 ? S 13:22 0:00 > > /usr/libexec/gconf-im-settings-daemon > > radek 11120 0.0 0.0 201512 2528 ? S 13:22 0:00 xfsettingsd > > radek 11140 0.0 0.0 148776 3532 ? S 13:22 0:00 > > /usr/libexec/gvfs-gdu-volume-monitor > > radek 11144 0.0 0.0 232436 2064 ? Sl 13:22 0:00 > > /usr/libexec/gvfs-afc-volume-monitor > > radek 11147 0.0 0.0 150892 2196 ? S 13:22 0:00 > > /usr/libexec/gvfs-gphoto2-volume-monitor > > radek 11149 0.0 0.0 146304 3228 ? S 13:22 0:00 > > /usr/libexec/gvfsd-trash --spawner :1.2 /org/gtk/gvfs/exec_spaw/0 > > radek 11297 0.0 0.0 356628 10964 ? S 13:28 0:00 > > /usr/libexec/notification-daemon > > > > After KDE log out: > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > radek 11832 0.0 0.0 20040 692 ? S 13:40 0:00 dbus-launch > > --sh-syntax --exit-with-session > > radek 11833 0.0 0.0 21668 1208 ? Ss 13:40 0:00 > > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > > radek 11900 0.0 0.0 235144 6708 ? Ss 13:40 0:00 > > /usr/bin/seahorse-agent --variables > > radek 11908 0.0 0.0 137360 2104 ? S 13:40 0:00 > > /usr/libexec/gvfsd > > radek 11914 0.0 0.0 133848 5768 ? S 13:40 0:00 > > /usr/libexec/gconfd-2 > > radek 12102 0.0 0.0 115096 4052 ? S 13:40 0:00 > > /usr/libexec/im-settings-daemon > > radek 12258 0.0 0.0 40900 2232 ? S 13:40 0:00 > > /usr/libexec/gconf-im-settings-daemon > > > > Could anybody advice me how to close all processes within log out? > > > > Best regards, > > Radek > > > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From accessys at smart.net Thu Apr 17 18:47:18 2014 From: accessys at smart.net (accessys at smart.net) Date: Thu, 17 Apr 2014 14:47:18 -0400 (EDT) Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: <1397757386.31997.6.camel@alpaga.bursztynowski.waw.pl> References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> <1397757386.31997.6.camel@alpaga.bursztynowski.waw.pl> Message-ID: yes I had to do that a couple times but it gets old having to slay them one at a time Bob On Thu, 17 Apr 2014, Radek Bursztynowski wrote: > Date: Thu, 17 Apr 2014 19:56:26 +0200 > From: Radek Bursztynowski > To: Support list for open source software in schools. > Subject: Re: [K12OSN] Log off doesn end all user processes > > Hello, > > Many thanks for advice. I will test it tomorrow > . > Coming back to this problem with Fedora 11 thin client image I keep > conviction that I have to do something on thin client site. After > killing all user processes on server site I can see on thin client > working one user process (socket username at server_ip_address. When I kill > this proces on the terminal LDM reloads and everything comes back to the > norm. So I must to do something with Fedora 11 thin client image. > > Best regards, > Radek > > > >> Hello, >> >> I have this same problem and I found the 'slay' command works >> perfectly to resolve the leftover processes. >> >> # slay username >> >> Plus it's fun to tell the others that I had to 'slay' so-and-so :) >> >> Info: >> http://linuxpoison.blogspot.ca/2011/01/kills-all-of-users-processes-slay.html >> >> Joseph >> >> On Wed, Apr 16, 2014 at 7:50 AM, Radek Bursztynowski >> wrote: >>> Hello, >>> >>> I have a problem with ending on server thin client LTSP session. When I log >>> out from the server session I return to the LDM, but there are not closed my >>> user processes on the server. After GNOME log out there are: >>> >>> [root at tryandbuy pxelinux.cfg]# ps aux | grep radek >>> radek 10790 0.0 0.0 20040 680 ? S 13:22 0:00 dbus-launch >>> --sh-syntax --exit-with-session >>> radek 10791 0.0 0.0 21816 1236 ? Ss 13:22 0:00 >>> /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session >>> radek 10858 0.0 0.0 232924 6588 ? Ss 13:22 0:00 >>> /usr/bin/seahorse-agent --variables >>> radek 10862 0.0 0.0 137360 2152 ? S 13:22 0:00 >>> /usr/libexec/gvfsd >>> radek 10876 0.0 0.0 133864 5808 ? S 13:22 0:00 >>> /usr/libexec/gconfd-2 >>> radek 10906 0.0 0.0 24752 2032 ? S 13:22 0:00 >>> /usr/lib64/xfce4/xfconf/xfconfd >>> radek 10929 0.0 0.0 238448 5640 ? S 13:22 0:00 Thunar >>> --daemon >>> radek 10936 0.0 0.0 242892 2900 ? Ss 13:22 0:00 >>> gnome-screensaver >>> radek 10943 0.0 0.0 328644 10784 ? Sl 13:22 0:00 >>> /usr/libexec/polkit-gnome-authentication-agent-1 >>> radek 10946 0.0 0.0 508884 10908 ? S 13:22 0:00 >>> gnome-volume-control-applet >>> radek 10952 0.0 0.0 115096 4052 ? S 13:22 0:00 >>> /usr/libexec/im-settings-daemon >>> radek 10975 0.0 0.0 251984 6792 ? S 13:22 0:00 >>> gnome-power-manager >>> radek 10978 0.0 0.1 303344 18736 ? S 13:22 0:00 python >>> /usr/share/system-config-printer/applet.py >>> radek 10986 0.0 0.0 362372 9840 ? S 13:22 0:00 >>> gpk-update-icon >>> radek 10997 0.0 0.0 358308 6872 ? S 13:22 0:00 >>> /usr/libexec/evolution/2.32/evolution-alarm-notify >>> radek 11119 0.0 0.0 40900 2232 ? S 13:22 0:00 >>> /usr/libexec/gconf-im-settings-daemon >>> radek 11120 0.0 0.0 201512 2528 ? S 13:22 0:00 xfsettingsd >>> radek 11140 0.0 0.0 148776 3532 ? S 13:22 0:00 >>> /usr/libexec/gvfs-gdu-volume-monitor >>> radek 11144 0.0 0.0 232436 2064 ? Sl 13:22 0:00 >>> /usr/libexec/gvfs-afc-volume-monitor >>> radek 11147 0.0 0.0 150892 2196 ? S 13:22 0:00 >>> /usr/libexec/gvfs-gphoto2-volume-monitor >>> radek 11149 0.0 0.0 146304 3228 ? S 13:22 0:00 >>> /usr/libexec/gvfsd-trash --spawner :1.2 /org/gtk/gvfs/exec_spaw/0 >>> radek 11297 0.0 0.0 356628 10964 ? S 13:28 0:00 >>> /usr/libexec/notification-daemon >>> >>> After KDE log out: >>> [root at tryandbuy pxelinux.cfg]# ps aux | grep radek >>> >>> [root at tryandbuy pxelinux.cfg]# ps aux | grep radek >>> radek 11832 0.0 0.0 20040 692 ? S 13:40 0:00 dbus-launch >>> --sh-syntax --exit-with-session >>> radek 11833 0.0 0.0 21668 1208 ? Ss 13:40 0:00 >>> /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session >>> radek 11900 0.0 0.0 235144 6708 ? Ss 13:40 0:00 >>> /usr/bin/seahorse-agent --variables >>> radek 11908 0.0 0.0 137360 2104 ? S 13:40 0:00 >>> /usr/libexec/gvfsd >>> radek 11914 0.0 0.0 133848 5768 ? S 13:40 0:00 >>> /usr/libexec/gconfd-2 >>> radek 12102 0.0 0.0 115096 4052 ? S 13:40 0:00 >>> /usr/libexec/im-settings-daemon >>> radek 12258 0.0 0.0 40900 2232 ? S 13:40 0:00 >>> /usr/libexec/gconf-im-settings-daemon >>> >>> Could anybody advice me how to close all processes within log out? >>> >>> Best regards, >>> Radek >>> >>> _______________________________________________ >>> K12OSN mailing list >>> K12OSN at redhat.com >>> https://www.redhat.com/mailman/listinfo/k12osn >>> For more info see >> >> _______________________________________________ >> K12OSN mailing list >> K12OSN at redhat.com >> https://www.redhat.com/mailman/listinfo/k12osn >> For more info see > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > From cisna-barry at wc235.k12.il.us Fri Apr 18 14:52:08 2014 From: cisna-barry at wc235.k12.il.us (Barry Cisna) Date: Fri, 18 Apr 2014 09:52:08 -0500 Subject: [K12OSN] HP T5740 boot panic Message-ID: <1397832729.31327.7.camel@localhost> Hello All, Just purchased a new HP T5740 TC to test with,to use with about a one year old install of (LTSP 6.0 SL client). The unit kernel panics at boot,so I updated the bios to the latest available bios from HP, with fingers crossed hoping possibly the later bios would remedy the boot panic.Of course this didn't change the situation. After doing some searching it appears the Broadcom nic in these units is the snag on the boot process. I have seen several workarounds to get these units to boot,mostly being building an Ubuntu boot image to drop into the tftp dir on the server. We have had several of the HP T5730's that would boot up fine out of the box. Anyone tried the HP T5740 Atom 1.6 ghz ,Intel i915 video,broadcom nic,atheros wireless.? Thank You, Barry From jim.kinney at gmail.com Fri Apr 18 15:02:20 2014 From: jim.kinney at gmail.com (Jim Kinney) Date: Fri, 18 Apr 2014 11:02:20 -0400 Subject: [K12OSN] HP T5740 boot panic In-Reply-To: <1397832729.31327.7.camel@localhost> References: <1397832729.31327.7.camel@localhost> Message-ID: Some Broadcom nics are closed source driver issues (Grr!). You may have one of these. You'll need to use a module from rpmfusion-nonfree to get the nic to work on the TC. That will require a boot kernel rebuild and a rebuild of the initrd. The Ubuntu image works because they are not based in the US and often ship code that is not freely available fro distribution. On Fri, Apr 18, 2014 at 10:52 AM, Barry Cisna wrote: > Hello All, > > Just purchased a new HP T5740 TC to test with,to use with about a one > year old install of (LTSP 6.0 SL client). > > The unit kernel panics at boot,so I updated the bios to the latest > available bios from HP, with fingers crossed hoping possibly the later > bios would remedy the boot panic.Of course this didn't change the > situation. > > After doing some searching it appears the Broadcom nic in these units is > the snag on the boot process. > I have seen several workarounds to get these units to boot,mostly being > building an Ubuntu boot image to drop into the tftp dir on the server. > > We have had several of the HP T5730's that would boot up fine out of the > box. > > Anyone tried the HP T5740 Atom 1.6 ghz ,Intel i915 video,broadcom > nic,atheros wireless.? > > Thank You, > Barry > > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see > -- -- James P. Kinney III Every time you stop a school, you will have to build a jail. What you gain at one end you lose at the other. It's like feeding a dog on his own tail. It won't fatten the dog. - Speech 11/23/1900 Mark Twain *http://heretothereideas.blogspot.com/ * -------------- next part -------------- An HTML attachment was scrubbed... URL: From radek at bursztynowski.waw.pl Fri Apr 18 15:56:13 2014 From: radek at bursztynowski.waw.pl (Radek Bursztynowski) Date: Fri, 18 Apr 2014 17:56:13 +0200 Subject: [K12OSN] Log off doesn end all user processes In-Reply-To: References: <18941283.311397649014863.JavaMail.root@poczta.bursztynowski.waw.pl> Message-ID: <1397836573.4547.4.camel@alpaga.bursztynowski.waw.pl> Hello, Thanks again for advice, but unfortunately 'slay' doesn't solve my problem with Fedora 11 thin client image. So, I should come back to my previous idea - to do something with thin client image. Best regards, Radek -- > Hello, > > I have this same problem and I found the 'slay' command works > perfectly to resolve the leftover processes. > > # slay username > > Plus it's fun to tell the others that I had to 'slay' so-and-so :) > > Info: > http://linuxpoison.blogspot.ca/2011/01/kills-all-of-users-processes-slay.html > > Joseph > > On Wed, Apr 16, 2014 at 7:50 AM, Radek Bursztynowski > wrote: > > Hello, > > > > I have a problem with ending on server thin client LTSP session. When I log > > out from the server session I return to the LDM, but there are not closed my > > user processes on the server. After GNOME log out there are: > > > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > radek 10790 0.0 0.0 20040 680 ? S 13:22 0:00 dbus-launch > > --sh-syntax --exit-with-session > > radek 10791 0.0 0.0 21816 1236 ? Ss 13:22 0:00 > > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > > radek 10858 0.0 0.0 232924 6588 ? Ss 13:22 0:00 > > /usr/bin/seahorse-agent --variables > > radek 10862 0.0 0.0 137360 2152 ? S 13:22 0:00 > > /usr/libexec/gvfsd > > radek 10876 0.0 0.0 133864 5808 ? S 13:22 0:00 > > /usr/libexec/gconfd-2 > > radek 10906 0.0 0.0 24752 2032 ? S 13:22 0:00 > > /usr/lib64/xfce4/xfconf/xfconfd > > radek 10929 0.0 0.0 238448 5640 ? S 13:22 0:00 Thunar > > --daemon > > radek 10936 0.0 0.0 242892 2900 ? Ss 13:22 0:00 > > gnome-screensaver > > radek 10943 0.0 0.0 328644 10784 ? Sl 13:22 0:00 > > /usr/libexec/polkit-gnome-authentication-agent-1 > > radek 10946 0.0 0.0 508884 10908 ? S 13:22 0:00 > > gnome-volume-control-applet > > radek 10952 0.0 0.0 115096 4052 ? S 13:22 0:00 > > /usr/libexec/im-settings-daemon > > radek 10975 0.0 0.0 251984 6792 ? S 13:22 0:00 > > gnome-power-manager > > radek 10978 0.0 0.1 303344 18736 ? S 13:22 0:00 python > > /usr/share/system-config-printer/applet.py > > radek 10986 0.0 0.0 362372 9840 ? S 13:22 0:00 > > gpk-update-icon > > radek 10997 0.0 0.0 358308 6872 ? S 13:22 0:00 > > /usr/libexec/evolution/2.32/evolution-alarm-notify > > radek 11119 0.0 0.0 40900 2232 ? S 13:22 0:00 > > /usr/libexec/gconf-im-settings-daemon > > radek 11120 0.0 0.0 201512 2528 ? S 13:22 0:00 xfsettingsd > > radek 11140 0.0 0.0 148776 3532 ? S 13:22 0:00 > > /usr/libexec/gvfs-gdu-volume-monitor > > radek 11144 0.0 0.0 232436 2064 ? Sl 13:22 0:00 > > /usr/libexec/gvfs-afc-volume-monitor > > radek 11147 0.0 0.0 150892 2196 ? S 13:22 0:00 > > /usr/libexec/gvfs-gphoto2-volume-monitor > > radek 11149 0.0 0.0 146304 3228 ? S 13:22 0:00 > > /usr/libexec/gvfsd-trash --spawner :1.2 /org/gtk/gvfs/exec_spaw/0 > > radek 11297 0.0 0.0 356628 10964 ? S 13:28 0:00 > > /usr/libexec/notification-daemon > > > > After KDE log out: > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > > > [root at tryandbuy pxelinux.cfg]# ps aux | grep radek > > radek 11832 0.0 0.0 20040 692 ? S 13:40 0:00 dbus-launch > > --sh-syntax --exit-with-session > > radek 11833 0.0 0.0 21668 1208 ? Ss 13:40 0:00 > > /bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session > > radek 11900 0.0 0.0 235144 6708 ? Ss 13:40 0:00 > > /usr/bin/seahorse-agent --variables > > radek 11908 0.0 0.0 137360 2104 ? S 13:40 0:00 > > /usr/libexec/gvfsd > > radek 11914 0.0 0.0 133848 5768 ? S 13:40 0:00 > > /usr/libexec/gconfd-2 > > radek 12102 0.0 0.0 115096 4052 ? S 13:40 0:00 > > /usr/libexec/im-settings-daemon > > radek 12258 0.0 0.0 40900 2232 ? S 13:40 0:00 > > /usr/libexec/gconf-im-settings-daemon > > > > Could anybody advice me how to close all processes within log out? > > > > Best regards, > > Radek > > > > _______________________________________________ > > K12OSN mailing list > > K12OSN at redhat.com > > https://www.redhat.com/mailman/listinfo/k12osn > > For more info see > > _______________________________________________ > K12OSN mailing list > K12OSN at redhat.com > https://www.redhat.com/mailman/listinfo/k12osn > For more info see From william at fragakis.com Fri Apr 18 16:41:36 2014 From: william at fragakis.com (William Fragakis) Date: Fri, 18 Apr 2014 12:41:36 -0400 Subject: [K12OSN] HP T5740 boot panic In-Reply-To: References: Message-ID: <1397839296.3201.58.camel@server.ltsp> Not knowing what Jim just told you, I ended up using a Debian boot image. Installed LTSP in a Debian VM and then copied it over. If you need the image/files, etc, email me off list. Additional quirk is that you have to make it ignore what it thinks is its built in screen and use the vga connection instead. Otherwise, you end up on a "second" screen. Add video=LVDS-1:d to the boot parameters in your files in /var/lib/tftpboot/ltsp/your_image/pxelinux.cfg e.g. label ltsp-NFS menu label LTSP, using NFS kernel vmlinuz-3.10-3-486 append ro initrd=initrd.img-3.10-3-486 init=/sbin/init-ltsp quiet root=/dev/nfs ip=dhcp boot=nfs video=LVDS-1:d ipappend 2 (append is all one line until ipappend in case your email reader wraps it) They are solid units. Typing on one presently. Regards, William On Fri, 2014-04-18 at 12:00 -0400, k12osn-request at redhat.com wrote: > > Message: 5 > Date: Fri, 18 Apr 2014 11:02:20 -0400 > From: Jim Kinney > To: "Support list for open source software in schools." > > Subject: Re: [K12OSN] HP T5740 boot panic > Message-ID: > +mEo5yyeWpjz1SsmCnhSNQ at mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Some Broadcom nics are closed source driver issues (Grr!). You may > have one > of these. You'll need to use a module from rpmfusion-nonfree to get > the nic > to work on the TC. That will require a boot kernel rebuild and a > rebuild of > the initrd. The Ubuntu image works because they are not based in the > US and > often ship code that is not freely available fro distribution. > > > On Fri, Apr 18, 2014 at 10:52 AM, Barry Cisna > wrote: > > > Hello All, > > > > Just purchased a new HP T5740 TC to test with,to use with about a > one > > year old install of (LTSP 6.0 SL client). > > > > The unit kernel panics at boot,so I updated the bios to the latest > > available bios from HP, with fingers crossed hoping possibly the > later > > bios would remedy the boot panic.Of course this didn't change the > > situation. > > > > After doing some searching it appears the Broadcom nic in these > units is > > the snag on the boot process. > > I have seen several workarounds to get these units to boot,mostly > being > > building an Ubuntu boot image to drop into the tftp dir on the > server. > > > > We have had several of the HP T5730's that would boot up fine out of > the > > box. > > > > Anyone tried the HP T5740 Atom 1.6 ghz ,Intel i915 video,broadcom > > nic,atheros wireless.? > > > > Thank You, > > Barry > > > > From brcisna at eazylivin.net Mon Apr 21 20:40:36 2014 From: brcisna at eazylivin.net (Barry R Cisna) Date: Mon, 21 Apr 2014 15:40:36 -0500 Subject: [K12OSN] HP T5740 boot panic Message-ID: <1398112837.6827.10.camel@server1.eazylivin.net> Hello All, A big Thank You to William Fragakis for providing an easy fix for the HP T5740 not booting properly. It was a simple copying of an Debian based boot kernel in an extra directory on the K12Linux/LTSP server and the TC booted without a hitch. Barry From brcisna at eazylivin.net Mon Apr 21 20:46:39 2014 From: brcisna at eazylivin.net (Barry R Cisna) Date: Mon, 21 Apr 2014 15:46:39 -0500 Subject: [K12OSN] New K12Linux/LTSP install , client boots slow Message-ID: <1398113199.6827.16.camel@server1.eazylivin.net> Hello All, Curious if anyone has done a very recent install of LTSP server & client build? I am not honestly sure what the status is of K12Linux as all the pages I have bookmarked for a long time for L12Linux points to LTSP? I done an default install in a VM of LTSP and in turn installed Virtualbox in it,to run an Virtual PXE client to test boot this setup. The Virtual PXE client boots miserably slow,and ends up just "stopping" and not going any further. Tried appending a few various switches to the kernel command line for the chroot boot image but any changes made has not made any difference. Thanks, Barry