planetlab-users Mailing List for PlanetLab (Page 2)
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: Steve M. <sm...@cs...> - 2004-01-21 15:18:26
|
Emil, 1) Every PlanetLab node has a single name in the database. Periodically (a couple of times a day maybe) I generate a table showing me various status items for each node. When I see that a node is not behaving as expected I go investigate - since that is driven by the list of nodes derived from the database my mental map is from database names to unexpected behaviour. I generally don't remember DNS aliases or IP addresses very well, so if you refer to a node by a name other than what's in the database I have to go look it up in my table. 2) I am familiar with the 'host' command. steve On Wed, 21 Jan 2004, Emil Sit wrote: > > 2) when you tell us the nodes that you cannot login to please tell us the > > DNS names e.g., planetlab-1.cs.princeton.edu, rather than the IP > > addresses. You will often get a faster response because we can tell you > > right away, for example, that planetlab11.millennium.berkeley.edu does not > > have its CD-ROM mounted, but matching 169.229.50.89 makes us go search a > > list of nodes. > > Just for clarification, will any equivalent DNS name suffice? > For example, berkeley3.plab.lcs.mit.edu? Or berkeley3.lists.planet-lab.org? > Does it have to be planetlab3.millennium.berkeley.edu? I want to make > sure that I'll get the fastest response possible if I ever have a > problem; sometimes I use these aliases in my scripts and they don't > report the CNAME of the host. > > Incidentally, you might find the 'host' command useful: > > $ host 169.229.50.89 > 89.50.229.169.IN-ADDR.ARPA domain name pointer planetlab11.Millennium.Berkeley.EDU > > And if you have a whole big list of IP addresses, you can relatively > easily reverse-resolve them using adnsresfilter [*]. > > [*] http://rpmfind.net/linux/rpm2html/search.php?query=adns > |
|
From: Emil S. <si...@MI...> - 2004-01-21 14:50:12
|
> 2) when you tell us the nodes that you cannot login to please tell us the > DNS names e.g., planetlab-1.cs.princeton.edu, rather than the IP > addresses. You will often get a faster response because we can tell you > right away, for example, that planetlab11.millennium.berkeley.edu does not > have its CD-ROM mounted, but matching 169.229.50.89 makes us go search a > list of nodes. Just for clarification, will any equivalent DNS name suffice? For example, berkeley3.plab.lcs.mit.edu? Or berkeley3.lists.planet-lab.org? Does it have to be planetlab3.millennium.berkeley.edu? I want to make sure that I'll get the fastest response possible if I ever have a problem; sometimes I use these aliases in my scripts and they don't report the CNAME of the host. Incidentally, you might find the 'host' command useful: $ host 169.229.50.89 89.50.229.169.IN-ADDR.ARPA domain name pointer planetlab11.Millennium.Berkeley.EDU And if you have a whole big list of IP addresses, you can relatively easily reverse-resolve them using adnsresfilter [*]. [*] http://rpmfind.net/linux/rpm2html/search.php?query=adns |
|
From: Steve M. <sm...@CS...> - 2004-01-21 14:29:16
|
Dear Users, When reporting to us that you are unable to login to your dynamic slice, please observe the following couple of guidelines: 1) tell us your slice name and the list of nodes that you have attempted to login to. 2) when you tell us the nodes that you cannot login to please tell us the DNS names e.g., planetlab-1.cs.princeton.edu, rather than the IP addresses. You will often get a faster response because we can tell you right away, for example, that planetlab11.millennium.berkeley.edu does not have its CD-ROM mounted, but matching 169.229.50.89 makes us go search a list of nodes. cheers, steve |
|
From: Paul B. <pau...@pl...> - 2004-01-19 18:27:03
|
The Canarie nodes are moving from their current (lab) location out to the data centers, and will be unavailable for most of this week. If you have any queries, please do not hesitate to contact me. Paul Brett PlanetLab Support Email: pau...@pl... Tel No: +1 503 712 4520 |
|
From: Larry P. <ll...@CS...> - 2004-01-19 13:01:36
|
Perhaps... Are you still using your static slice (gatech7 or whatever)? You need to follow the "Version 2.0" link on the home page. All slices are now dynamically created. Larry On Jan 19, 2004, at 3:24 AM, Ramesh Viswanath wrote: > Hello folks > > I just restarted using Planetlab. And I couldn't log in. > So as suggested, I went in and regenerated and uploaded my public key, > but > am still denied access. > It says Permission denied ..yada yada..public key doesn't match et al > > I waited for two days, I tried on multiple nodes, I re-regenerated my > public key..nothing worked :( > Is there anything I am supposed to do extra..I used this last year, and > many changes have been made and I may have missed something. > > Ramesh > > > > > > > ------------------------------------------------------- > The SF.Net email is sponsored by EclipseCon 2004 > Premiere Conference on Open Tools Development and Integration > See the breadth of Eclipse activity. February 3-5 in Anaheim, CA. > http://www.eclipsecon.org/osdn > _______________________________________________ > Planetlab-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/planetlab-users > |
|
From: Steven H. <Ste...@cl...> - 2004-01-19 10:15:17
|
Ramesh Viswanath <ra...@cc...> writes: > > I just restarted using Planetlab. And I couldn't log in. > So as suggested, I went in and regenerated and uploaded my public key, but > am still denied access. > It says Permission denied ..yada yada..public key doesn't match et al > > I waited for two days, I tried on multiple nodes, I re-regenerated my > public key..nothing worked :( > Is there anything I am supposed to do extra..I used this last year, and > many changes have been made and I may have missed something. The world has changed to version 2.0; you will not be able to log in to any node unless a) there is a slice of which you are a member, and b) the node is question is part of that slice To effect this, an authorised PI for gatech needs to perform (a); at present afaik any member of a slice can perform (b). These actions can be performed on the web (https://www.planet-lab.org/db/slices/) or or programatically (see http://www.planet-lab.org/doc/InfraGuide.php for more information and example scripts) hth, S. |
|
From: Ramesh V. <ra...@cc...> - 2004-01-19 08:24:39
|
Hello folks I just restarted using Planetlab. And I couldn't log in. So as suggested, I went in and regenerated and uploaded my public key, but am still denied access. It says Permission denied ..yada yada..public key doesn't match et al I waited for two days, I tried on multiple nodes, I re-regenerated my public key..nothing worked :( Is there anything I am supposed to do extra..I used this last year, and many changes have been made and I may have missed something. Ramesh |
|
From: Bowman, M. <mic...@in...> - 2004-01-16 17:33:21
|
multicast depends on the site.
--mic
-----Original Message-----
From: pla...@li... =
[mailto:pla...@li...] On Behalf Of =
Manuel Carro
Sent: Friday, January 16, 2004 03:30 AM
To: pla...@li...
Cc: pla...@cl...; jcb...@fi...
Subject: [Planetlab-support] Multicast traffic
Hello. I have a questino related to multicast traffic. Is it =
supposed to be a facility permanently available in PlanetLab nodes =
(because, e.g., it is needed for core services of the PlanetLab =
architecture), or it is just something that might be available at =
certain sites, but not at others?
=20
The reason for my question is that, for bandwidth reasons, multicast =
is to be enabled under demand at our University (it is usually done so =
for teleteaching), and, if it is required, I'd have to send in a formal =
request including a rationale and some specific data.
--=20
+------------------------------------------------------------------------=
--+
|Manuel Carro -- Facultad de Inform=E1tica -- U. Polit=E9cnica de Madrid =
|(UPM) | mcarro_at_fi_dot_upm_dot_es --- Ph: +34-91336-7455 --- FAX:=20
|+34-91336-7412| http://lml.ls.fi.upm.es/~boris ---=20
|http://clip.dia.fi.upm.es/Software/Ciao|
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration See the =
breadth of Eclipse activity. February 3-5 in Anaheim, CA. =
http://www.eclipsecon.org/osdn =
_______________________________________________
Planetlab-support mailing list Pla...@li...
https://lists.sourceforge.net/lists/listinfo/planetlab-support
|
|
From: Krishna G. <gu...@cs...> - 2004-01-16 17:00:19
|
Here's a simple trick to fix this. Looking over the source code at: http://cvs.sourceforge.net/viewcvs.py/planetlab/plkmodutil/ It appears that the PlanetLab traceroute program tries to bind to a default port (33433) for all users. So there is no way two traceroute programs can be simultaneously run by the same or even different users ! I added a few lines to "traceroute.c" file to retry random ports when the bind fails and so far it works fine for me. The modified source code (binary) are at: www.cs.washington.edu/homes/gummadi/temp/my_pl_traceroute-1.4a12 --Krishna On Mon, 12 Jan 2004, Jeremy Stribling wrote: > Hi all, > > Does anyone have experience with the traceroute included in the > plkmodutil package? It seems to work all right some of the time, but > when I try to use more than one copy at once, I always get a: > > traceroute: bind: Address already in use > > I also get this sometimes even if I'm only doing a single traceroute. > Has anyone else encountered this? Thanks, > > Jeremy > > > > ------------------------------------------------------- > This SF.net email is sponsored by: Perforce Software. > Perforce is the Fast Software Configuration Management System offering > advanced branching capabilities and atomic changes on 50+ platforms. > Free Eval! http://www.perforce.com/perforce/loadprog.html > _______________________________________________ > Planetlab-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/planetlab-users > |
|
From: Aaron K. <al...@CS...> - 2004-01-16 15:42:25
|
If you want to set up yum and python in your slice, you can simply run this line: curl --silent http://www.cs.princeton.edu/~alk/setup_yum.sh | /bin/bash Aaron |
|
From: Aaron K. <al...@CS...> - 2004-01-15 16:00:03
|
Hi, could you give us a quick update on the status of the two UC Davis nodes? They are currently in a rins state, however, if both have been replaced with new hardware, they will need a user to type 'yes' for the initial install attempt because our system keys off of the mac address and they will appear to be new machines. Aaron On Monday 05 January 2004 04:32 pm, Chen-Nee Chuah wrote: > Planetlab1.ece.ucdavis.edu will be re-installed on Jan 7, 2004 > as part of hardware swap (replacing Dell PC with HP Proliant server). > > Sorry for any inconvenience it may cause. > > > Regards, > Chen-Nee |
|
From: Sedayao, J. <jef...@in...> - 2004-01-13 22:19:51
|
Robert Adams has created some Python scripts for creating and managing = dynamic slices. I have modified the addnode utility so that it will = take a list of nodes to add to a slice from standard input. You can read about this software and other contributions at = http://www.planet-lab.org/php/contrib/contrib.php Jeff Sedayao PlanetLab Support |
|
From: Minseok K. <kw...@cs...> - 2004-01-13 13:33:11
|
Thanks Yun and Neil, I should have checked previous messages. Now everything works. Thanks. I have one more question. I cannot get g++ using a similar way to get gcc. Any reason for this? Also, I am not that familar with the linux environment since I have mostly worked on Solaris. What would be good points covering from the basics (e.g., handling these managers like rpm, yum, etc.) to advanced topics? Thanks. Best regards, Minseok Kwon On Mon, 12 Jan 2004, Neil Spring wrote: > follow Aaron's helpful instructions to install yum from earlier posts > to planetlab-users, or invoke > > curl -O http://www.scriptroute.org/planetlab/planetlab-install > sh ./planetlab-install > yum -y install make > > you may also want > yum -y install man gzip less gcc flex bison bc rpm-build > > (which has the added benefit of installing a traceroute that can run 50 > copies at a time...) > > the planetlab-install script is still a work-in-progress, but it's > close. > > -neil > > On Jan 12, 2004, at 12:43 PM, Minseok Kwon wrote: > > > It seems that we do not have make and many other utilities on planetlab > > nodes. Are there any tools or ways to get similar functions? Thanks. > > > > Best regards, > > > > Minseok Kwon > > > > > > ------------------------------------------------------- > > This SF.net email is sponsored by: Perforce Software. > > Perforce is the Fast Software Configuration Management System offering > > advanced branching capabilities and atomic changes on 50+ platforms. > > Free Eval! http://www.perforce.com/perforce/loadprog.html > > _______________________________________________ > > Planetlab-users mailing list > > Pla...@li... > > https://lists.sourceforge.net/lists/listinfo/planetlab-users > > > |
|
From: Neil S. <ns...@cs...> - 2004-01-12 20:58:59
|
follow Aaron's helpful instructions to install yum from earlier posts to planetlab-users, or invoke curl -O http://www.scriptroute.org/planetlab/planetlab-install sh ./planetlab-install yum -y install make you may also want yum -y install man gzip less gcc flex bison bc rpm-build (which has the added benefit of installing a traceroute that can run 50 copies at a time...) the planetlab-install script is still a work-in-progress, but it's close. -neil On Jan 12, 2004, at 12:43 PM, Minseok Kwon wrote: > It seems that we do not have make and many other utilities on planetlab > nodes. Are there any tools or ways to get similar functions? Thanks. > > Best regards, > > Minseok Kwon > > > ------------------------------------------------------- > This SF.net email is sponsored by: Perforce Software. > Perforce is the Fast Software Configuration Management System offering > advanced branching capabilities and atomic changes on 50+ platforms. > Free Eval! http://www.perforce.com/perforce/loadprog.html > _______________________________________________ > Planetlab-users mailing list > Pla...@li... > https://lists.sourceforge.net/lists/listinfo/planetlab-users > |
|
From: Yun M. <ma...@ci...> - 2004-01-12 20:57:45
|
Hi Minseok, just treat it like a redhat 9 distribution with minimal install. (maybe even less?) You can install any rpms you want. Use apt-get (freshrpms.net) or yum may help the installation process. E.g., if you want "make", do: su - rpm -Uvh http://ftp.freshrpms.net/pub/freshrpms/redhat/9/apt/apt-get-xxx.rpm apt-get update apt-get install make HTH. Yun On Mon, 12 Jan 2004, Minseok Kwon wrote: > It seems that we do not have make and many other utilities on planetlab > nodes. Are there any tools or ways to get similar functions? Thanks. > > Best regards, > > Minseok Kwon > |
|
From: Minseok K. <kw...@cs...> - 2004-01-12 20:44:04
|
It seems that we do not have make and many other utilities on planetlab nodes. Are there any tools or ways to get similar functions? Thanks. Best regards, Minseok Kwon |
|
From: Jeremy S. <st...@am...> - 2004-01-12 18:28:06
|
Hi all, Does anyone have experience with the traceroute included in the plkmodutil package? It seems to work all right some of the time, but when I try to use more than one copy at once, I always get a: traceroute: bind: Address already in use I also get this sometimes even if I'm only doing a single traceroute. Has anyone else encountered this? Thanks, Jeremy |
|
From: Sedayao, J. <jef...@in...> - 2004-01-12 18:00:32
|
Yun,
It is intended as a metric of what proportion of the resources on a =
slice is granted. It isn't used yet, though, so setting it doesn't =
reall do anything.
Jeff
-----Original Message-----
From: Yun Mao [mailto:ma...@ci...]
Sent: Friday, January 09, 2004 1:58 PM
To: Sedayao, Jeff
Cc: pla...@li...
Subject: RE: [Planetlab-users] dynamic slices doc
Thanks Jeff.
I still have one question left:
=
=20
In "AssignShares" call, I understand what is renewTime, but what is
"share" for? I can't match it with any parameters in the GUI interface.
=
=20
Thanks!
Yun
On Fri, 9 Jan 2004, Sedayao, Jeff wrote:
> Yun,
>=20
> This is back now.
>=20
> Jeff
>=20
> -----Original Message-----
> From: pla...@li...
> [mailto:pla...@li...]On Behalf Of Yun =
Mao
> Sent: Friday, January 09, 2004 11:29 AM
> To: pla...@li...
> Subject: [Planetlab-users] dynamic slices doc
>=20
>=20
> Hi, where can I find dynamic slice programming interface doc? The link =
at
> planet-lab.org
> (http://www.planet-lab.org/doc/InfraGuide.php#DynamicSlicesAPI) is =
broken.
> Thank you!
>=20
> Yun
>=20
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> Planetlab-users mailing list
> Pla...@li...
> https://lists.sourceforge.net/lists/listinfo/planetlab-users
>=20
>=20
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> Planetlab-users mailing list
> Pla...@li...
> https://lists.sourceforge.net/lists/listinfo/planetlab-users
>=20
|
|
From: Paul B. <pau...@pl...> - 2004-01-11 03:48:03
|
The latest planetlab kernel, along with packages listed below have just
been deployed to all working planetlab nodes.
initscripts-7.14-3_planetlab.i386.rpm
linux-2.4.22-r3_planetlab.i386.rpm
linux-headers-2.4.22-r3_planetlab.i386.rpm
lkcdutils-4.1-3_planetlab.i386.rpm
plkmod-2.0.1-2.4.22_r3_planetlab.i386.rpm
vr-tools-0.14-1_planetlab.i386.rpm
vserver-0.29-2_planetlab.i386.rpm
vsh-2.05.2-3_planetlab.i386.rpm
vtune_driver-918-2.4.22_r3_planetlab.i386.rpm
The upgrade provides the following capabilities:
1) Linux Kernel
Upgraded vserver patch to 1.22 including:
* bugfix tasklist_lock/uts_sem issue
* bugfix in tcp_ipv4_addr_conflict()
* critical bugfix SMP deadlock, dyn.
* bugfix dyn. wraparound lockup
* bugfix in vroot device
* small bugfix in switch_user_struct()
* fixed compat quota capability
* changed vmalloc to kmalloc
* added syscall switch
* added send signal to context
* included vroot device vr0.14
* included reboot userspace helper
* limit/kill syscall templates
* dynamic id range moved to 49152-
* block static use of dyn. ctx id
* three levels of cleanup and restructuring
* x86_64, parisc, and ppc64 support
Bug fix enabling read only bind mounts
Enabled panic reboot after 15 minutes, when kernel panics
during initialization.
Security fix for the do_remap exploit.
2) Linux Kernel Crash Dump
Fixed error in packaging which could prevent crash dumps being
generated.
3) Plkmod
Ported to vserver 1.22 and gcc 3.2
4) vtune_driver
Re-compilation for kernel - no changes
5) vserver tools
Upgraded to 0.29 including the following:
* support for new syscall
* bug fix in ULIMIT handling
* security fix for capability handling
6) VR tools
Upgraded to 0.14 to match vroot device now included as standard
part of vserver kernel patch
7) vsh
Upgraded to vservers 1.22 syscall
Re-built as a static binary to remove dependencies on libraries
installed within the vserver
8) vserver-init
Bug fix which prevented rc.vinit scripts from being executed on
startup.
Regards
Paul Brett
PlanetLab Support
Email: pau...@pl...
Tel No: +1 503 712 4520
|
|
From: Minseok K. <kw...@cs...> - 2004-01-10 17:36:47
|
Hi, I still cannot login a significant number of nodes though I have assigned them to my slice purdue_15 (I can login many other nodes, btw). For example, I cannot access planetlab1.cnds.jhu.edu, planetlab-2.stanford.edu (planetlab-1.stanford.edu is OK), planetlab1.eecs.umich.edu, etc. Any reason for this? Thanks. Best Regards, Minseok Kwon |
|
From: Yun M. <ma...@ci...> - 2004-01-09 21:58:04
|
Thanks Jeff.
I still have one question left:
In "AssignShares" call, I understand what is renewTime, but what is
"share" for? I can't match it with any parameters in the GUI interface.
Thanks!
Yun
On Fri, 9 Jan 2004, Sedayao, Jeff wrote:
> Yun,
>
> This is back now.
>
> Jeff
>
> -----Original Message-----
> From: pla...@li...
> [mailto:pla...@li...]On Behalf Of Yun Mao
> Sent: Friday, January 09, 2004 11:29 AM
> To: pla...@li...
> Subject: [Planetlab-users] dynamic slices doc
>
>
> Hi, where can I find dynamic slice programming interface doc? The link at
> planet-lab.org
> (http://www.planet-lab.org/doc/InfraGuide.php#DynamicSlicesAPI) is broken.
> Thank you!
>
> Yun
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> Planetlab-users mailing list
> Pla...@li...
> https://lists.sourceforge.net/lists/listinfo/planetlab-users
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Perforce Software.
> Perforce is the Fast Software Configuration Management System offering
> advanced branching capabilities and atomic changes on 50+ platforms.
> Free Eval! http://www.perforce.com/perforce/loadprog.html
> _______________________________________________
> Planetlab-users mailing list
> Pla...@li...
> https://lists.sourceforge.net/lists/listinfo/planetlab-users
>
|
