planetlab-users Mailing List for PlanetLab (Page 6)
Brought to you by:
alklinga
Menu
▾
▴
planetlab-users — Discussion among PlanetLab users; but send trouble reports to planetlab-support
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(3) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
(2) |
Feb
(2) |
Mar
(7) |
Apr
(6) |
May
(5) |
Jun
(8) |
Jul
|
Aug
(9) |
Sep
(18) |
Oct
(24) |
Nov
(46) |
Dec
(37) |
| 2004 |
Jan
(65) |
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
(2) |
Jul
(1) |
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
|
| 2005 |
Jan
(2) |
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Justin C. <jus...@ho...> - 2003-11-25 18:36:03
|
Has anyone implemented the mechanism for /etc/rc.vinit on the PlanetLab nodes (i.e. will such a script in my slice be executed upon node reboot)? Also, it seems like it could be useful to have a node (or small set of nodes) that reboot periodically so that researchers could test how well fail-overs, etc. are handled by their slices. Is there any interest in this? Thanks, Justin Cappos (University of Arizona Grad Student) _________________________________________________________________ Say goodbye to busy signals and slow downloads with a high-speed Internet connection! Prices start at less than $1 a day average. https://broadband.msn.com (Prices may vary by service area.) |
|
From: Adams, R. <rob...@in...> - 2003-11-24 17:38:02
|
PlanetLab nodes require a static IP address. Could you check with your IT department and see what you must do to get a static IP assignment. PlanetLab nodes won't work with a changing IP address. =20 See if that solution is possible. =20 -- RA -----Original Message----- From: pla...@li... [mailto:pla...@li...] On Behalf Of Ashok Janardhanan Sent: Monday, November 24, 2003 9:15 AM To: pla...@li... Subject: RE: [Planetlab-users] error trying to setup node I tried using a configuration floppy and that didn't help either. Please let me know if there is anything else that i can try while i wait for a solution. =20 Thanks Ashok "Adams, Robert" <rob...@in...> wrote: The problem must be that the DHCP is giving a set of addresses different than what you specified joining and you are not using a configuration floppy which complicates things. =20 Let us get back to you with a solution. =20 -- RA -----Original Message----- From: pla...@li... [mailto:pla...@li...] On Behalf Of Ashok Janardhanan Sent: Thursday, November 20, 2003 1:06 PM To: pla...@li... Subject: [Planetlab-users] error trying to setup node I was trying to set up a node at our site using the PlanetLab BootCD. I didn't prepare a configuration floppy since DHCP is used for the nodes at our site. However it gives me an error, while trying to install saying that the site is not authorized. We got our application authorized yesterday. Please let me know if i am doing something wrong. I have written to the support mailing list (yesterday), and haven't heard back yet. =20 Thanks Ashok _____ =20 Do you Yahoo!? Free <http://us.rd.yahoo.com/slv/mailtag/*http://companion.yahoo.com/> Pop-Up Blocker - Get it now _____ =20 Do you Yahoo!? Free <http://us.rd.yahoo.com/slv/mailtag/*http://companion.yahoo.com/> Pop-Up Blocker - Get it now |
|
From: Ashok J. <pla...@ya...> - 2003-11-24 17:15:22
|
I tried using a configuration floppy and that didn't help either. Please let me know if there is anything else that i can try while i wait for a solution. Thanks Ashok "Adams, Robert" <rob...@in...> wrote: The problem must be that the DHCP is giving a set of addresses different than what you specified joining and you are not using a configuration floppy which complicates things. Let us get back to you with a solution. -- RA -----Original Message----- From: pla...@li... [mailto:pla...@li...] On Behalf Of Ashok Janardhanan Sent: Thursday, November 20, 2003 1:06 PM To: pla...@li... Subject: [Planetlab-users] error trying to setup node I was trying to set up a node at our site using the PlanetLab BootCD. I didn't prepare a configuration floppy since DHCP is used for the nodes at our site. However it gives me an error, while trying to install saying that the site is not authorized. We got our application authorized yesterday. Please let me know if i am doing something wrong. I have written to the support mailing list (yesterday), and haven't heard back yet. Thanks Ashok --------------------------------- Do you Yahoo!? Free Pop-Up Blocker - Get it now --------------------------------- Do you Yahoo!? Free Pop-Up Blocker - Get it now |
|
From: Michal S. <mi...@cs...> - 2003-11-21 16:19:17
|
Hi, > planetlab.h: No such file or directory :-) you have to download planetlab.h from the web site: http://www.planet-lab.org/raw_sockets/planetlab.h Yes, I find it funny, too :-) Enjoy, M. -- Michal Szymaniak | mailto:mi...@cs... | http://www.cs.vu.nl/~michal |
|
From: Thomas G. <tgi...@gm...> - 2003-11-21 12:51:35
|
hi, i've tried to make a sniffer socket. Therefore i had do include planetlab.h. I tried to compile my code on two different nodes. Everytime i got this error-message: planetlab.h: No such file or directory do anybody know what is wrong? thanks for helping gingold tomasz -- GMX Weihnachts-Special: Seychellen-Traumreise zu gewinnen! Rentier entlaufen. Finden Sie Rudolph! Als Belohnung winken tolle Preise. http://www.gmx.net/de/cgi/specialmail/ +++ GMX - die erste Adresse für Mail, Message, More! +++ |
|
From: Adams, R. <rob...@in...> - 2003-11-20 21:21:23
|
The problem must be that the DHCP is giving a set of addresses different than what you specified joining and you are not using a configuration floppy which complicates things. =20 Let us get back to you with a solution. =20 -- RA -----Original Message----- From: pla...@li... [mailto:pla...@li...] On Behalf Of Ashok Janardhanan Sent: Thursday, November 20, 2003 1:06 PM To: pla...@li... Subject: [Planetlab-users] error trying to setup node I was trying to set up a node at our site using the PlanetLab BootCD. I didn't prepare a configuration floppy since DHCP is used for the nodes at our site. However it gives me an error, while trying to install saying that the site is not authorized. We got our application authorized yesterday. Please let me know if i am doing something wrong. I have written to the support mailing list (yesterday), and haven't heard back yet. =20 Thanks Ashok _____ =20 Do you Yahoo!? Free <http://us.rd.yahoo.com/slv/mailtag/*http://companion.yahoo.com/> Pop-Up Blocker - Get it now |
|
From: Ashok J. <pla...@ya...> - 2003-11-20 21:06:21
|
I was trying to set up a node at our site using the PlanetLab BootCD. I didn't prepare a configuration floppy since DHCP is used for the nodes at our site. However it gives me an error, while trying to install saying that the site is not authorized. We got our application authorized yesterday. Please let me know if i am doing something wrong. I have written to the support mailing list (yesterday), and haven't heard back yet. Thanks Ashok --------------------------------- Do you Yahoo!? Free Pop-Up Blocker - Get it now |
|
From: Adams, R. <rob...@in...> - 2003-11-20 16:17:31
|
PlanetLab is experiencing a problem with the accumulation of zombie processes on nodes. The problem is being worked by both PlanetLab support and several experimenters, but, in the meantime, I will be rebooting some PlanetLab nodes to clean out the accumulated zombies. This will bring the nodes back into full operation. The following nodes have more than 150 zombie processes and will be rebooted in the next four days: csplanetlab3.kaist.ac.kr (zombies =3D 150) csplanetlab2.kaist.ac.kr (zombies =3D 151) csplanetlab4.kaist.ac.kr (zombies =3D 151) csplanetlab1.kaist.ac.kr (zombies =3D 153) arari.snu.ac.kr (zombies =3D 153) lefthand.eecs.harvard.edu (zombies =3D 154) planetlab1.cs.cornell.edu (zombies =3D 161) planetlab1.ias.csusb.edu (zombies =3D 162) planetlab0.ias.csusb.edu (zombies =3D 164) planetlab1.cs.umass.edu (zombies =3D 170) planetlab2.rn.informatics.scitech.susx.ac.uk (zombies =3D 181) planetlab1.it.uts.edu.au (zombies =3D 188) planetlab2.cs.wayne.edu (zombies =3D 195) planetlab2.cs.cornell.edu (zombies =3D 196) planet1.ecse.rpi.edu (zombies =3D 197) recall.snu.ac.kr (zombies =3D 198) planetlab2.it.uts.edu.au (zombies =3D 200) righthand.eecs.harvard.edu (zombies =3D 200) planetlab1.ashburn.equinix.planet-lab.org (zombies =3D 279) planet3.berkeley.intel-research.net (zombies =3D 397) planetlab1.cse.msu.edu (zombies =3D 401) -- Robert Adams 2111 NE 25th; Hillsboro, Oregon 97124 (v) +1-503-264-2682 (f) +1-503-264-1578 |
|
From: Shih-Yen G. <ca...@cs...> - 2003-11-19 05:00:59
|
Dear Mr.
I am from National Tsing-Hua University. I have some problems about =
log-in PlanetLab.
1. How can I log-in all machines on the PlanetLab...?
I followed the steps below , but it seems to fail =A1G
(1). I can make sure I do all the right thing on =
registering a new account.
Especially, the "ssh-keygen ......" and upload the =
identity.pub
(2). I log-in one Linux station , in which I did the =
"ssh-keygen -t rsa -f ~/.ssh/identity" instruction.
(3). And do "ssh -l nthu1 pads21.cs.nthu.edu.tw "
(nthu1 is our slice, and pads21.cs is our =
PlanetLab machine)
[carengao@pads29 .ssh]$ ssh -l nthu1 =
pads21.cs.nthu.edu.tw
Warning: Permanently added =
'pads21.cs.nthu.edu.tw,140.114.79.231' (RSA) to the list of known hosts.
Permission denied =
(publickey,keyboard-interactive).
=20
As above, it showed a "Permission denied" message. Why can't I =
log-in ~? Did I make any mistake ~?
How can I log-in a PlanetLab machine ~?
2. Does PlanetLab have some job dispatching tools for experiments? If =
my experiment needs 50 nodes, how do I submit my
job??Do I need to login 50 nodes that I choose one by one?? Or =
there are another ways to help this? How do I monitor=20
the jobs in every nodes?
Appreciate your kindness responses.
Best Regards
S.Y Gau
--
Shih-Yen Gao
Parallel and Distributed System Laboratory
National Tsing-Hua University
Hsinchu, Taiwan 300
Tel: 886+3+5715131 ~ 3553
http://pads1.cs.nthu.edu.tw/
|
|
From: Jeremy S. <st...@uc...> - 2003-11-19 02:37:24
|
Jay Lepreau wrote: >>Thanks for the tip. It looks like I can use this to get commodity >>internet only nodes, but not internet2-only nodes. >> >> > >No, you can ask for any of the four subtypes: inet2, inet, intl, dsl, >or just "plab" if you don't care. > > Great, thanks for the list, I appreciate it. Just to clarify though, this does not contain the list of internet2-only nodes (that is, nodes with which the inet nodes cannot directly communicate). I'm guessing this is because it's based on data from Brent's machine at Intel (just like infospect can't handle inet2-only nodes because it's run from Mothy's machine). Of course, subtracting this list from the list of production nodes on the planet-lab website gives me an idea, but that would really just be the union of the set of inet2 nodes and the set of nodes not running dslice. Jeremy |
|
From: Jay L. <le...@cs...> - 2003-11-19 02:07:20
|
> Thanks for the tip. It looks like I can use this to get commodity > internet only nodes, but not internet2-only nodes. No, you can ask for any of the four subtypes: inet2, inet, intl, dsl, or just "plab" if you don't care. > Even so, all I want > is the list of these nodes, or the technique used to get those lists. > Any way to do this short of going through them one by one? Thanks, Rob did a DB dump and we temporarily stuck it here: http://www.emulab.net/tutorial/docwrapper.php3?docname=plab-nodetypes.txt The beginning of it is appended. Note the caveat about missing nodes! As for the technique, there are a lot of heuristics based on the name and routes, followed by a manual check. Errors might exist. --------------------------------------------------------------------------------------- PlanetLab nodes known to Emulab via the bnc's dslice service, 11/18/03. Nodes that aren't sucessfully running dslice do not appear. Not all of these 140 nodes are "up" at any given time. (First column is Emulab's arbitrary internal/external name for the node, which also appears in DNS, eg. plab58.emulab.net.) SUMMARY TYPE SITES HOSTS dsl 2 3 inet 8 16 inet2 44 96 intl 13 25 --- --- 67 140 NODE TYPE SITE HOSTNAME plab123 dsl accretive-dsl.nodes.planet-lab.org uw1.accretive-dsl.nodes.planet-lab.org plab58 dsl gti-dsl.nodes.planet-lab.org planetlab2.gti-dsl.nodes.planet-lab.org plab59 dsl gti-dsl.nodes.planet-lab.org planetlab1.gti-dsl.nodes.planet-lab.org plab138 inet ashburn.equinix.planet-lab.org planetlab2.ashburn.equinix.planet-lab.org #Colo plab21 inet ashburn.equinix.planet-lab.org planetlab1.ashburn.equinix.planet-lab.org #Colo plab22 inet sanjose.equinix.planet-lab.org planetlab3.sanjose.equinix.planet-lab.org #Colo plab31 inet berkeley.intel-research.net planet1.berkeley.intel-research.net plab32 inet berkeley.intel-research.net planet2.berkeley.intel-research.net plab25 inet hpl.hp.com pli1-pa-1.hpl.hp.com .....etc.... |
|
From: Jeremy S. <st...@uc...> - 2003-11-19 01:50:06
|
Thanks for the tip. It looks like I can use this to get commodity internet only nodes, but not internet2-only nodes. Even so, all I want is the list of these nodes, or the technique used to get those lists. Any way to do this short of going through them one by one? Thanks, Jeremy Manpreet Singh wrote: >Hi, > > In case you have access to Emulab network, you can just submit a file, >and request Planetlab nodes of specific type (e.g. Internet2-only nodes, >nodes in North America, etc). > >For details, please refer to: > >http://www.emulab.net/tutorial/planetlab.ns >http://www.emulab.net/tutorial/docwrapper.php3?docname=plab-announce.txt > >Cheers, Manpreet. > >On Tue, 18 Nov 2003, Jeremy Stribling wrote: > > > >>Is there a definitive list somewhere that names all of the >>internet2-only nodes on PlanetLab? And the same question for >>internet1-only nodes. There seems to be a psuedo naming convention for >>internet2-only nodes, but it doesn't hold for the HP I2 node, for >>example. If anyone knows this data, or knows of an easy way for me to >>figure it out (which doesn't involves having all the nodes ping each >>other -- I already do that and am trying to make sense of what I'm >>seeing), I'd be much obliged. I suppose I could have them all ping some >>well-known internet1-only or -2-only node, but I'd like something more >>official if possible. Thanks, >> >>Jeremy >> >> >> > > > |
|
From: Manpreet S. <man...@cs...> - 2003-11-19 00:30:24
|
Hi, In case you have access to Emulab network, you can just submit a file, and request Planetlab nodes of specific type (e.g. Internet2-only nodes, nodes in North America, etc). For details, please refer to: http://www.emulab.net/tutorial/planetlab.ns http://www.emulab.net/tutorial/docwrapper.php3?docname=plab-announce.txt Cheers, Manpreet. On Tue, 18 Nov 2003, Jeremy Stribling wrote: > Is there a definitive list somewhere that names all of the > internet2-only nodes on PlanetLab? And the same question for > internet1-only nodes. There seems to be a psuedo naming convention for > internet2-only nodes, but it doesn't hold for the HP I2 node, for > example. If anyone knows this data, or knows of an easy way for me to > figure it out (which doesn't involves having all the nodes ping each > other -- I already do that and am trying to make sense of what I'm > seeing), I'd be much obliged. I suppose I could have them all ping some > well-known internet1-only or -2-only node, but I'd like something more > official if possible. Thanks, > > Jeremy > |
|
From: Jeremy S. <st...@am...> - 2003-11-18 23:28:38
|
Is there a definitive list somewhere that names all of the internet2-only nodes on PlanetLab? And the same question for internet1-only nodes. There seems to be a psuedo naming convention for internet2-only nodes, but it doesn't hold for the HP I2 node, for example. If anyone knows this data, or knows of an easy way for me to figure it out (which doesn't involves having all the nodes ping each other -- I already do that and am trying to make sense of what I'm seeing), I'd be much obliged. I suppose I could have them all ping some well-known internet1-only or -2-only node, but I'd like something more official if possible. Thanks, Jeremy |
|
From: Andy B. <ac...@CS...> - 2003-11-17 15:22:16
|
> From: Neil Spring <ns...@cs...> > Subject: Re: [Planetlab-users] sysctl variables > Date: Sat, 15 Nov 2003 14:42:02 -0800 > To: pla...@li... > > > On Nov 14, 2003, at 6:12 AM, Larry Peterson wrote: > > In fact, as a generalization of this issue, some people would > > like to run alternative versions of TCP that are tuned for > > high-speed networks (e.g., FAST). One way of doing this, which > > you do as well, is run a user-level version of TCP. Then you > > can tweak it as much as you want. Of course, there's still going > > to be a limit on how much shared bandwidth is available, but if > > you selectively use nodes connected by fat pipes (e.g., the > > Abilene nodes) you might be able to do what you want. > > Larry, (or anyone well-versed in the raw sockets code) > > As Manpreet's last mail suggests he's trying this approach, do you > think a user-level TCP is sufficient to inflate buffer sizes? I > understand that keeping retransmissions in user memory removes the need > for a send buffer, but I'm worried about the fake receive buffer. If > the raw socket kernel->application buffer is limited to the same size > as the tcp buffer, then advertising a large receive window (now > possible with the user-level TCP) without being able to provide it > would lead to packet loss when the machine is busy, exactly what flow > control is designed to avoid. That's right, the size of the raw socket receive buffer is currently limited by rmem_max as well. Andy |
|
From: Neil S. <ns...@cs...> - 2003-11-15 22:42:02
|
On Nov 14, 2003, at 6:12 AM, Larry Peterson wrote: > In fact, as a generalization of this issue, some people would > like to run alternative versions of TCP that are tuned for > high-speed networks (e.g., FAST). One way of doing this, which > you do as well, is run a user-level version of TCP. Then you > can tweak it as much as you want. Of course, there's still going > to be a limit on how much shared bandwidth is available, but if > you selectively use nodes connected by fat pipes (e.g., the > Abilene nodes) you might be able to do what you want. Larry, (or anyone well-versed in the raw sockets code) As Manpreet's last mail suggests he's trying this approach, do you think a user-level TCP is sufficient to inflate buffer sizes? I understand that keeping retransmissions in user memory removes the need for a send buffer, but I'm worried about the fake receive buffer. If the raw socket kernel->application buffer is limited to the same size as the tcp buffer, then advertising a large receive window (now possible with the user-level TCP) without being able to provide it would lead to packet loss when the machine is busy, exactly what flow control is designed to avoid. Assuming I'm right about how raw socket buffers work, I'm wondering if a hybrid approach, inspired a bit by Semke, Mahdavi, and Mathis's sigcomm 98 paper on buffer autotuning, would work well -- raise the receive buffer size through the roof, and let the user tcp manipulate the send window. (In their paper, they focus on tuning the send buffer so as many connections as possible make good progress, but they leave the receive buffers as large as possible, expecting that congestion control will limit actual buffer consumption.) I don't know if a large kernel->user buffer is easy to configure for the raw socket code (or if it is already there and my concern is unwarranted). -neil musing that HZ=1000 might help user-level TCPs stay responsive on java-cpu-bound machines. |
|
From: Manpreet S. <man...@cs...> - 2003-11-15 20:47:09
|
Hello, Does anyone have a user-level implementation of TCP that works on the planetlab nodes? If yes, are you willing to share it with other users? I am sure there are many more people who would benefit from this. Thanks, Manpreet. |
|
From: Larry P. <ll...@CS...> - 2003-11-14 14:11:23
|
Neil is right that you currently have no way of adjusting the TCP window sizes. The question is should you be able to. I'd be interested in hearing what kind of bandwidth you're getting now and what level you'd hope to get with a larger window. At some point, you're going to hit a per-slice limit simply because you're sharing the outgoing link with a bunch of other users trying to do similar things. It seems unlikely that we're going to be able to support applications that are trying to achieve heroic TCP numbers in general. In fact, as a generalization of this issue, some people would like to run alternative versions of TCP that are tuned for high-speed networks (e.g., FAST). One way of doing this, which you do as well, is run a user-level version of TCP. Then you can tweak it as much as you want. Of course, there's still going to be a limit on how much shared bandwidth is available, but if you selectively use nodes connected by fat pipes (e.g., the Abilene nodes) you might be able to do what you want. Larry On Thursday, November 13, 2003, at 11:38 PM, Manpreet Singh wrote: > > Hello, > > Sorry to bother everyone, but I am also new on planetlab. Am I > allowed > to change some sysctl variables? I tried 'su' on some nodes, but that > also does not permit me to do this. > > In particular, I want to change the default send/receive window > sizes. > I know that I can use setsockopt, but that is ultimately limited by the > sysctl variables. > > Thanks and Regards, > Manpreet Singh. > > > ------------------------------------------------------- > This SF.Net email sponsored by: ApacheCon 2003, > 16-19 November in Las Vegas. Learn firsthand the latest > developments in Apache, PHP, Perl, XML, Java, MySQL, > WebDAV, and more! http://www.apachecon.com/ > _______________________________________________ > Planetlab-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/planetlab-users > |
|
From: Manpreet S. <man...@cs...> - 2003-11-14 07:05:07
|
Thanks a lot for your help! > It might be possible to increase [rw]mem_max, but that is a discussion > for planetlab-arch or planetlab-support; how large did you intend to > configure? and can you tell that 132KB is too small? I only want the [rw]mem_max variables to be increased, so that my application is not constrained by send/receive buffers. Maybe 512KB would suffice. Cheers, Manpreet. |
|
From: Neil S. <ns...@cs...> - 2003-11-14 06:35:14
|
Manpreet, On Nov 13, 2003, at 8:38 PM, Manpreet Singh wrote: > Sorry to bother everyone, but I am also new on planetlab. I was told last month that this is the forum for new users to bother people and feel comfortable doing it. > Am I allowed > to change some sysctl variables? I tried 'su' on some nodes, but that > also does not permit me to do this. Being root in your vserver just lets you manipulate the filesystem, it doesn't let you reconfigure the kernel. :( > In particular, I want to change the default send/receive window > sizes. > I know that I can use setsockopt, but that is ultimately limited by the > sysctl variables. Your application should use setsockopt, because I don't expect any change to vservers will let you frob wmem_default or tcp_wmem. there's still just one kernel and a large socket buffer would probably do my application harm. (hmm, maybe I should even shrink it.) It might be possible to increase [rw]mem_max, but that is a discussion for planetlab-arch or planetlab-support; how large did you intend to configure? and can you tell that 132KB is too small? -neil (no stranger to the joys of tcp window frobbing) |
|
From: Manpreet S. <man...@cs...> - 2003-11-14 04:39:08
|
Hello,
Sorry to bother everyone, but I am also new on planetlab. Am I allowed
to change some sysctl variables? I tried 'su' on some nodes, but that
also does not permit me to do this.
In particular, I want to change the default send/receive window sizes.
I know that I can use setsockopt, but that is ultimately limited by the
sysctl variables.
Thanks and Regards,
Manpreet Singh.
|
|
From: Neil S. <ns...@cs...> - 2003-11-13 17:07:42
|
On Nov 12, 2003, at 12:04 PM, tomas wrote: > i have read this paper. My problem is not how to implement a raw > socket. I > want to know how to to EXECUTE my allready written code an a node. > Here are some concrete questions. > I mean, is it enough just to compile my code on my pc and upload it to > a > node? unlikely, unless you build a statically linked executable. > Or do i have to compile it after logging into a node? yes, but you can then copy the binary you build to other nodes, which can save time. If your program involves several files, you might build an rpm (redhat package), then install that rpm elsewhere. > Do i have to > compile it as a *.o-file or as a shared library *.so? I don't understand. shared libraries aren't executable. neither are object files. > Are they default names > like for example "index.html" (concerning web-pages)? no. > I mean the node must > know what file execute first. you execute what you want on a node. the node handles no execution for you (except, I suppose, for rc.vinit, a script run when the machine boots). scp mycode.c me...@pl...: ssh -n -l me1 planetlab-555.university.edu "gcc -W -Wall mycode.c -o myprogram && ./myprogram" -neil |
|
From: tomas <tgi...@gm...> - 2003-11-12 20:04:59
|
>Hi Thomas, >For networking, there are 'safe raw sockets' >(http://www.planet-lab.org/raw_sockets) which make many raw >socket operations available. >Chears, Manpreet. Hey Manpreet, i have read this paper. My problem is not how to implement a raw socket. I want to know how to to EXECUTE my allready written code an a node. Here are some concrete questions. I mean, is it enough just to compile my code on my pc and upload it to a node? Or do i have to compile it after logging into a node? Do i have to compile it as a *.o-file or as a shared library *.so? Are they default names like for example "index.html" (concerning web-pages)? I mean the node must know what file execute first. I know that this are basic questions, but how can i know when nobody tells me. please help me regards Tomasz Gingold tkn-berlin |
|
From: Manpreet S. <man...@cs...> - 2003-11-12 19:19:34
|
Hi Thomas,=20 For networking, there are 'safe raw sockets' (http://www.planet-lab.org/raw_sockets) which make many raw socket operations available. Chears, Manpreet.=20 On Wed, 12 Nov 2003, Thomas Gingold wrote: > hi, >=20 > my PI has allready created a slice, and i am now able to log into any nod= e. > But how can i start a capsule (service) on this node??? I first want to w= rite > a script which only tries to open a raw socket and then execute this scri= pt > on one node. >=20 > I have not found any related docs. >=20 >=20 > Thanks in advance. > Tomasz Gingold > tkn.berlin >=20 > --=20 > NEU F=DCR ALLE - GMX MediaCenter - f=FCr Fotos, Musik, Dateien... > Fotoalbum, File Sharing, MMS, Multimedia-Gru=DF, GMX FotoService >=20 > Jetzt kostenlos anmelden unter http://www.gmx.net >=20 > +++ GMX - die erste Adresse f=FCr Mail, Message, More! +++ >=20 >=20 >=20 > ------------------------------------------------------- > This SF.Net email sponsored by: ApacheCon 2003, > 16-19 November in Las Vegas. Learn firsthand the latest > developments in Apache, PHP, Perl, XML, Java, MySQL, > WebDAV, and more! http://www.apachecon.com/ > _______________________________________________ > Planetlab-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/planetlab-users >=20 |
|
From: Thomas G. <tgi...@gm...> - 2003-11-12 16:38:57
|
hi, my PI has allready created a slice, and i am now able to log into any node. But how can i start a capsule (service) on this node??? I first want to write a script which only tries to open a raw socket and then execute this script on one node. I have not found any related docs. Thanks in advance. Tomasz Gingold tkn.berlin -- NEU FÜR ALLE - GMX MediaCenter - für Fotos, Musik, Dateien... Fotoalbum, File Sharing, MMS, Multimedia-Gruß, GMX FotoService Jetzt kostenlos anmelden unter http://www.gmx.net +++ GMX - die erste Adresse für Mail, Message, More! +++ |
