From pilani at usc.edu Sun Feb 1 12:10:58 2004 From: pilani at usc.edu (rahul pilani) Date: Sun Feb 1 12:11:00 2004 Subject: [Csci551-talk] Re: Csci551-talk Digest, Vol 2, Issue 1 Message-ID: <4b2fc4d500.4d5004b2fc@usc.edu> Wouldnt it be better to use learn.usc.edu's discussion board .. rather than a mailing list like this one?.. i know this is more direct.. but that is more sophisticated and much more convenient.. just a suggestion.. Rahul ----- Original Message ----- From: csci551-talk-request@mailman.isi.edu Date: Sunday, February 1, 2004 12:00 pm Subject: Csci551-talk Digest, Vol 2, Issue 1 > Send Csci551-talk mailing list submissions to > csci551-talk@mailman.isi.edu > > To subscribe or unsubscribe via the World Wide Web, visit > http://mailman.isi.edu/mailman/listinfo/csci551-talk > or, via email, send a message with subject or body 'help' to > csci551-talk-request@mailman.isi.edu > > You can reach the person managing the list at > csci551-talk-owner@mailman.isi.edu > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Csci551-talk digest..." > > > Today's Topics: > > 1. RE: Socket Program (Aaron Tu) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sat, 31 Jan 2004 17:37:44 -0800 > From: "Aaron Tu" > Subject: [Csci551-talk] RE: Socket Program > To: > Message-ID: <001001c3e864$02f489e0$6701a8c0@LATUALP> > Content-Type: text/plain; charset="us-ascii" > > Is there a homework assignment (socket program) that I'm not aware of? > > Aaron > > -----Original Message----- > From: csci551-talk-bounces@mailman.isi.edu > [csci551-talk-bounces@mailman.isi.edu] On Behalf Of > csci551-talk-request@mailman.isi.edu > Sent: Friday, January 30, 2004 12:00 PM > To: csci551-talk@mailman.isi.edu > Subject: Csci551-talk Digest, Vol 1, Issue 154 > > Send Csci551-talk mailing list submissions to > csci551-talk@mailman.isi.edu > > To subscribe or unsubscribe via the World Wide Web, visit > http://mailman.isi.edu/mailman/listinfo/csci551-talk > or, via email, send a message with subject or body 'help' to > csci551-talk-request@mailman.isi.edu > > You can reach the person managing the list at > csci551-talk-owner@mailman.isi.edu > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Csci551-talk digest..." > > > Today's Topics: > > 1. LandMark Hierarchy Example (Varun Goel) > 2. Re: LandMark Hierarchy Example (Yuan Li) > 3. Re: LandMark Hierarchy Example (Varun Goel) > 4. socket programs (nishant agarwal) > 5. Re: LandMark Hierarchy Example (Yuan Li) > 6. RE: socket programs (Devarshi Shah) > 7. Re: socket programs (John Heidemann) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 29 Jan 2004 15:45:12 -0500 > From: "Varun Goel" > Subject: [Csci551-talk] LandMark Hierarchy Example > To: csci551-talk@mailman.isi.edu > Message-ID: <20040129204512.32399.qmail@usa.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hi, > > In the Landmark Hierarchy Example given in the research paper they have > derived resulting path as 5 hops 1 greater than shortest path but it > happened due to the fact that router having no entry for destination > forwarded packet to level 2 router which was three hops away but if it > had routed the packet to level 1 router which was two hops away and had > an entry for it too, it could have resulted in shortest path. > > I am unable to understand why the packet wasnt routed to level 1 router > instead? > > Please Clarify if anybody know..... > > Varun > -- > __________________________________________________________ > Sign-up for your own personalized E-mail at Mail.com > http://www.mail.com/?sr=signup > > Search Smarter - get the new eXact Search Bar for free! > http://www.exactsearchbar.com/ > > > > ------------------------------ > > Message: 2 > Date: Thu, 29 Jan 2004 13:40:57 -0800 (PST) > From: Yuan Li > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > To: Varun Goel > Cc: csci551-talk@mailman.isi.edu > Message-ID: > Content-Type: TEXT/PLAIN; charset=US-ASCII > > > In the example(Fig 4), d.i.g's routing table does not have an entry for > level 1 router n, so it needs to forward the packet to Level 2 router d > instead. And in its routing table, f(not k) is the next hop to reach d. > > 2.2.5 could help you to better understand routing in Landmark Hierarchy. > > Thanks > > Yuan > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > Hi, > > > > In the Landmark Hierarchy Example given in the research paper they > have derived resulting path as 5 hops 1 greater than shortest path but > it happened due to the fact that router having no entry for destination > forwarded packet to level 2 router which was three hops away but if it > had routed the packet to level 1 router which was two hops away and had > an entry for it too, it could have resulted in shortest path. > > > > I am unable to understand why the packet wasnt routed to level 1 > router instead? > > > > Please Clarify if anybody know..... > > > > Varun > > -- > > __________________________________________________________ > > Sign-up for your own personalized E-mail at Mail.com > > http://www.mail.com/?sr=signup > > > > Search Smarter - get the new eXact Search Bar for free! > > http://www.exactsearchbar.com/ > > > > > > ------------------------------ > > Message: 3 > Date: Thu, 29 Jan 2004 17:15:45 -0500 > From: "Varun Goel" > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > To: "Yuan Li" > Cc: csci551-talk@mailman.isi.edu > Message-ID: <20040129221545.7728.qmail@usa.com> > Content-Type: text/plain; charset="iso-8859-1" > > But Yuan if u will carefully see u will notice that router in question > do have an entry for level 1 router d.i.i through k so it could have > routed to d.i.i instead of d.d.d which seems more appropriate as it is > just two hops away i.e in the range of d.i.g (r0=2 given) > > > Please clarify... > > Thanks > > Varun > > ----- Original Message ----- > > > From: Yuan Li > Date: Thu, 29 Jan 2004 13:40:57 -0800 (PST) > To: Varun Goel > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > > > > > In the example(Fig 4), d.i.g's routing table does not have an entry > for > > level 1 router n, so it needs to forward the packet to Level 2 router > d > > instead. And in its routing table, f(not k) is the next hop to reach > d. > > > > 2.2.5 could help you to better understand routing in Landmark > Hierarchy. > > > > Thanks > > > > Yuan > > > > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > > > Hi, > > > > > > In the Landmark Hierarchy Example given in the research paper they > have derived resulting path as 5 hops 1 greater than shortest path but > it happened due to the fact that router having no entry for destination > forwarded packet to level 2 router which was three hops away but if it > had routed the packet to level 1 router which was two hops away and had > an entry for it too, it could have resulted in shortest path. > > > > > > I am unable to understand why the packet wasnt routed to level 1 > router instead? > > > > > > Please Clarify if anybody know..... > > > > > > Varun > > > -- > > > __________________________________________________________ > > > Sign-up for your own personalized E-mail at Mail.com > > > http://www.mail.com/?sr=signup > > > > > > Search Smarter - get the new eXact Search Bar for free! > > > http://www.exactsearchbar.com/ > > > > > > > -- > __________________________________________________________ > Sign-up for your own personalized E-mail at Mail.com > http://www.mail.com/?sr=signup > > Search Smarter - get the new eXact Search Bar for free! > http://www.exactsearchbar.com/ > > > > ------------------------------ > > Message: 4 > Date: Thu, 29 Jan 2004 20:11:14 -0800 > From: nishant agarwal > Subject: [Csci551-talk] socket programs > To: csci551-talk@ISI.EDU > Message-ID: > Content-Type: text/plain; charset=us-ascii > > I have done: > #include and so on for sys/types.h , netinet/in.h, > arpa/inet.h and all the files suggested in the local man pages. But yet > I get symbol referencing errors for socket, bind and inet_addr . what > cud be the problem. > > Also, what is the way to examine the status of my ports using netstat. > How do I identify my localhost local address there so that I can grep > and see the status of my machine ports. > > cheers > Nishant > > > > ------------------------------ > > Message: 5 > Date: Thu, 29 Jan 2004 20:36:54 -0800 (PST) > From: Yuan Li > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > To: Varun Goel > Cc: csci551-talk@mailman.isi.edu > Message-ID: > Content-Type: TEXT/PLAIN; charset=US-ASCII > > > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > But Yuan if u will carefully see u will notice that router in question > > do have an entry for level 1 router d.i.i through k so it could have > > routed to d.i.i instead of d.d.d which seems more appropriate as it > is > > just two hops away i.e in the range of d.i.g (r0=2 given) > > But g can not say for sure that d.i.i will find a way to t, though in > this > case it happens that i has an entry for t. > > So in general when g does not have an entry for n, it needs to forward > the > packets to upper level router, d in this case. d for sure has an entry > for n, and the packets can thus be forwarded. > > > Thanks > > Yuan > > > > > > > > Please clarify... > > > > Thanks > > > > Varun > > > > ----- Original Message ----- > > > > > > From: Yuan Li > > Date: Thu, 29 Jan 2004 13:40:57 -0800 (PST) > > To: Varun Goel > > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > > > > > > > > In the example(Fig 4), d.i.g's routing table does not have an entry > for > > > level 1 router n, so it needs to forward the packet to Level 2 > router d > > > instead. And in its routing table, f(not k) is the next hop to > reach d. > > > > > > 2.2.5 could help you to better understand routing in Landmark > Hierarchy. > > > > > > Thanks > > > > > > Yuan > > > > > > > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > > > > > Hi, > > > > > > > > In the Landmark Hierarchy Example given in the research paper they > have derived resulting path as 5 hops 1 greater than shortest path but > it happened due to the fact that router having no entry for destination > forwarded packet to level 2 router which was three hops away but if it > had routed the packet to level 1 router which was two hops away and had > an entry for it too, it could have resulted in shortest path. > > > > > > > > I am unable to understand why the packet wasnt routed to level 1 > router instead? > > > > > > > > Please Clarify if anybody know..... > > > > > > > > Varun > > > > -- > > > > __________________________________________________________ > > > > Sign-up for your own personalized E-mail at Mail.com > > > > http://www.mail.com/?sr=signup > > > > > > > > Search Smarter - get the new eXact Search Bar for free! > > > > http://www.exactsearchbar.com/ > > > > > > > > > > > -- > > __________________________________________________________ > > Sign-up for your own personalized E-mail at Mail.com > > http://www.mail.com/?sr=signup > > > > Search Smarter - get the new eXact Search Bar for free! > > http://www.exactsearchbar.com/ > > > > > > ------------------------------ > > Message: 6 > Date: Thu, 29 Jan 2004 20:36:13 -0800 > From: "Devarshi Shah" > Subject: RE: [Csci551-talk] socket programs > To: > Message-ID: <200401300436.i0U4aK007687@vapor.isi.edu> > Content-Type: text/plain; charset="us-ascii" > > Hi Nishant, > > If I am correct, you would also need to link the appropriate libraries. > (ex. "-l socket" option). > > Devarshi > > Devarshi P. Shah > -------------------------------------------------------------- > Graduate Student (Computer Science), > University of Southern California, LA. > -------------------------------------------------------------- > Web : http://devarshi.shah.name > email : devarshi@ieee.org > Cell : 323-363-3791, Home :323-373-0318 > -------------------------------------------------------------- > > > -----Original Message----- > From: csci551-talk-bounces@mailman.isi.edu > [csci551-talk-bounces@mailman.isi.edu] On Behalf Of nishant > agarwal > Sent: Thursday, January 29, 2004 8:11 PM > To: csci551-talk@ISI.EDU > Subject: [Csci551-talk] socket programs > > I have done: > #include and so on for sys/types.h , netinet/in.h, > arpa/inet.h and all the files suggested in the local man pages. But yet > I > get symbol referencing errors for socket, bind and inet_addr . what cud > be > the problem. > > Also, what is the way to examine the status of my ports using netstat. > How > do I identify my localhost local address there so that I can grep and > see > the status of my machine ports. > > cheers > Nishant > > > > > ------------------------------ > > Message: 7 > Date: Fri, 30 Jan 2004 08:39:09 -0800 > From: John Heidemann > Subject: Re: [Csci551-talk] socket programs > To: nishant agarwal > Cc: csci551-talk@ISI.EDU > Message-ID: <200401301639.i0UGd9wN004840@dash.isi.edu> > Content-Type: text/plain; charset=US-ASCII > > On Thu, 29 Jan 2004 20:11:14 PST, nishant agarwal wrote: > >I have done: > >#include and so on for sys/types.h , netinet/in.h, > arpa/inet.h and all the files suggested in the local man pages. But yet > I get symbol referencing errors for socket, bind and inet_addr . what > cud be the problem. > > > > Sockets have an abstract layer (socket/bind/etc.) that is protocol > addressing independent, and then specific headers for each protocol > family (IPv4, IPv6, OSI, etc.). > > if you want to get sockaddr_in or inet_addr, those are in the protocol > family-specific header. On my system (a Linux box), see man ip(7) for > details. > > > >Also, what is the way to examine the status of my ports using netstat. > How do I identify my localhost local address there so that I can grep > and see the status of my machine ports. > > > >cheers > >Nishant > > ifconfig will tell you about your local address, among other ways. > > -John Heidemann > > > ------------------------------ > > _______________________________________________ > Csci551-talk mailing list > Csci551-talk@mailman.isi.edu > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > > End of Csci551-talk Digest, Vol 1, Issue 154 > ******************************************** > > > > ------------------------------ > > _______________________________________________ > Csci551-talk mailing list > Csci551-talk@mailman.isi.edu > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > > End of Csci551-talk Digest, Vol 2, Issue 1 > ****************************************** > From naa at usc.edu Sun Feb 1 12:12:46 2004 From: naa at usc.edu (nishant agarwal) Date: Sun Feb 1 12:12:51 2004 Subject: [Csci551-talk] Re: Csci551-talk Digest, Vol 2, Issue 1 Message-ID: <4fcf350d79.50d794fcf3@usc.edu> maybe professor wants us to code our own discussion board for the class for project 2. sounds like a good idea :) ----- Original Message ----- From: rahul pilani Date: Sunday, February 1, 2004 12:10 pm Subject: [Csci551-talk] Re: Csci551-talk Digest, Vol 2, Issue 1 > Wouldnt it be better to use learn.usc.edu's discussion board .. rather > than a mailing list like this one?.. i know this is more direct.. but that > is more sophisticated and much more convenient.. > just a suggestion.. > > Rahul > > ----- Original Message ----- > From: csci551-talk-request@mailman.isi.edu > Date: Sunday, February 1, 2004 12:00 pm > Subject: Csci551-talk Digest, Vol 2, Issue 1 > > > Send Csci551-talk mailing list submissions to > > csci551-talk@mailman.isi.edu > > > > To subscribe or unsubscribe via the World Wide Web, visit > > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > or, via email, send a message with subject or body 'help' to > > csci551-talk-request@mailman.isi.edu > > > > You can reach the person managing the list at > > csci551-talk-owner@mailman.isi.edu > > > > When replying, please edit your Subject line so it is more specific > > than "Re: Contents of Csci551-talk digest..." > > > > > > Today's Topics: > > > > 1. RE: Socket Program (Aaron Tu) > > > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Sat, 31 Jan 2004 17:37:44 -0800 > > From: "Aaron Tu" > > Subject: [Csci551-talk] RE: Socket Program > > To: > > Message-ID: <001001c3e864$02f489e0$6701a8c0@LATUALP> > > Content-Type: text/plain; charset="us-ascii" > > > > Is there a homework assignment (socket program) that I'm not aware of? > > > > Aaron > > > > -----Original Message----- > > From: csci551-talk-bounces@mailman.isi.edu > > [csci551-talk-bounces@mailman.isi.edu] On Behalf Of > > csci551-talk-request@mailman.isi.edu > > Sent: Friday, January 30, 2004 12:00 PM > > To: csci551-talk@mailman.isi.edu > > Subject: Csci551-talk Digest, Vol 1, Issue 154 > > > > Send Csci551-talk mailing list submissions to > > csci551-talk@mailman.isi.edu > > > > To subscribe or unsubscribe via the World Wide Web, visit > > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > or, via email, send a message with subject or body 'help' to > > csci551-talk-request@mailman.isi.edu > > > > You can reach the person managing the list at > > csci551-talk-owner@mailman.isi.edu > > > > When replying, please edit your Subject line so it is more specific > > than "Re: Contents of Csci551-talk digest..." > > > > > > Today's Topics: > > > > 1. LandMark Hierarchy Example (Varun Goel) > > 2. Re: LandMark Hierarchy Example (Yuan Li) > > 3. Re: LandMark Hierarchy Example (Varun Goel) > > 4. socket programs (nishant agarwal) > > 5. Re: LandMark Hierarchy Example (Yuan Li) > > 6. RE: socket programs (Devarshi Shah) > > 7. Re: socket programs (John Heidemann) > > > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Thu, 29 Jan 2004 15:45:12 -0500 > > From: "Varun Goel" > > Subject: [Csci551-talk] LandMark Hierarchy Example > > To: csci551-talk@mailman.isi.edu > > Message-ID: <20040129204512.32399.qmail@usa.com> > > Content-Type: text/plain; charset="iso-8859-1" > > > > Hi, > > > > In the Landmark Hierarchy Example given in the research paper they have > > derived resulting path as 5 hops 1 greater than shortest path but it > > happened due to the fact that router having no entry for destination > > forwarded packet to level 2 router which was three hops away but if it > > had routed the packet to level 1 router which was two hops away and had > > an entry for it too, it could have resulted in shortest path. > > > > I am unable to understand why the packet wasnt routed to level 1 router > > instead? > > > > Please Clarify if anybody know..... > > > > Varun > > -- > > __________________________________________________________ > > Sign-up for your own personalized E-mail at Mail.com > > http://www.mail.com/?sr=signup > > > > Search Smarter - get the new eXact Search Bar for free! > > http://www.exactsearchbar.com/ > > > > > > > > ------------------------------ > > > > Message: 2 > > Date: Thu, 29 Jan 2004 13:40:57 -0800 (PST) > > From: Yuan Li > > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > > To: Varun Goel > > Cc: csci551-talk@mailman.isi.edu > > Message-ID: > > Content-Type: TEXT/PLAIN; charset=US-ASCII > > > > > > In the example(Fig 4), d.i.g's routing table does not have an entry for > > level 1 router n, so it needs to forward the packet to Level 2 router d > > instead. And in its routing table, f(not k) is the next hop to reach d. > > > > 2.2.5 could help you to better understand routing in Landmark Hierarchy. > > > > Thanks > > > > Yuan > > > > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > > > Hi, > > > > > > In the Landmark Hierarchy Example given in the research paper they > > have derived resulting path as 5 hops 1 greater than shortest path but > > it happened due to the fact that router having no entry for destination > > forwarded packet to level 2 router which was three hops away but if it > > had routed the packet to level 1 router which was two hops away and had > > an entry for it too, it could have resulted in shortest path. > > > > > > I am unable to understand why the packet wasnt routed to level 1 > > router instead? > > > > > > Please Clarify if anybody know..... > > > > > > Varun > > > -- > > > __________________________________________________________ > > > Sign-up for your own personalized E-mail at Mail.com > > > http://www.mail.com/?sr=signup > > > > > > Search Smarter - get the new eXact Search Bar for free! > > > http://www.exactsearchbar.com/ > > > > > > > > > > > ------------------------------ > > > > Message: 3 > > Date: Thu, 29 Jan 2004 17:15:45 -0500 > > From: "Varun Goel" > > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > > To: "Yuan Li" > > Cc: csci551-talk@mailman.isi.edu > > Message-ID: <20040129221545.7728.qmail@usa.com> > > Content-Type: text/plain; charset="iso-8859-1" > > > > But Yuan if u will carefully see u will notice that router in question > > do have an entry for level 1 router d.i.i through k so it could have > > routed to d.i.i instead of d.d.d which seems more appropriate as it is > > just two hops away i.e in the range of d.i.g (r0=2 given) > > > > > > Please clarify... > > > > Thanks > > > > Varun > > > > ----- Original Message ----- > > > > > > From: Yuan Li > > Date: Thu, 29 Jan 2004 13:40:57 -0800 (PST) > > To: Varun Goel > > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > > > > > > > > In the example(Fig 4), d.i.g's routing table does not have an entry > > for > > > level 1 router n, so it needs to forward the packet to Level 2 router > > d > > > instead. And in its routing table, f(not k) is the next hop to reach > > d. > > > > > > 2.2.5 could help you to better understand routing in Landmark > > Hierarchy. > > > > > > Thanks > > > > > > Yuan > > > > > > > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > > > > > Hi, > > > > > > > > In the Landmark Hierarchy Example given in the research paper they > > have derived resulting path as 5 hops 1 greater than shortest path but > > it happened due to the fact that router having no entry for destination > > forwarded packet to level 2 router which was three hops away but if it > > had routed the packet to level 1 router which was two hops away and had > > an entry for it too, it could have resulted in shortest path. > > > > > > > > I am unable to understand why the packet wasnt routed to level 1 > > router instead? > > > > > > > > Please Clarify if anybody know..... > > > > > > > > Varun > > > > -- > > > > __________________________________________________________ > > > > Sign-up for your own personalized E-mail at Mail.com > > > > http://www.mail.com/?sr=signup > > > > > > > > Search Smarter - get the new eXact Search Bar for free! > > > > http://www.exactsearchbar.com/ > > > > > > > > > > > -- > > __________________________________________________________ > > Sign-up for your own personalized E-mail at Mail.com > > http://www.mail.com/?sr=signup > > > > Search Smarter - get the new eXact Search Bar for free! > > http://www.exactsearchbar.com/ > > > > > > > > ------------------------------ > > > > Message: 4 > > Date: Thu, 29 Jan 2004 20:11:14 -0800 > > From: nishant agarwal > > Subject: [Csci551-talk] socket programs > > To: csci551-talk@ISI.EDU > > Message-ID: > > Content-Type: text/plain; charset=us-ascii > > > > I have done: > > #include and so on for sys/types.h , netinet/in.h, > > arpa/inet.h and all the files suggested in the local man pages. But yet > > I get symbol referencing errors for socket, bind and inet_addr . what > > cud be the problem. > > > > Also, what is the way to examine the status of my ports using netstat. > > How do I identify my localhost local address there so that I can grep > > and see the status of my machine ports. > > > > cheers > > Nishant > > > > > > > > ------------------------------ > > > > Message: 5 > > Date: Thu, 29 Jan 2004 20:36:54 -0800 (PST) > > From: Yuan Li > > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > > To: Varun Goel > > Cc: csci551-talk@mailman.isi.edu > > Message-ID: > > Content-Type: TEXT/PLAIN; charset=US-ASCII > > > > > > > > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > > > But Yuan if u will carefully see u will notice that router in question > > > do have an entry for level 1 router d.i.i through k so it could have > > > routed to d.i.i instead of d.d.d which seems more appropriate as it > > is > > > just two hops away i.e in the range of d.i.g (r0=2 given) > > > > But g can not say for sure that d.i.i will find a way to t, though in > > this > > case it happens that i has an entry for t. > > > > So in general when g does not have an entry for n, it needs to forward > > the > > packets to upper level router, d in this case. d for sure has an entry > > for n, and the packets can thus be forwarded. > > > > > > Thanks > > > > Yuan > > > > > > > > > > > > > Please clarify... > > > > > > Thanks > > > > > > Varun > > > > > > ----- Original Message ----- > > > > > > > > > From: Yuan Li > > > Date: Thu, 29 Jan 2004 13:40:57 -0800 (PST) > > > To: Varun Goel > > > Subject: Re: [Csci551-talk] LandMark Hierarchy Example > > > > > > > > > > > In the example(Fig 4), d.i.g's routing table does not have an entry > > for > > > > level 1 router n, so it needs to forward the packet to Level 2 > > router d > > > > instead. And in its routing table, f(not k) is the next hop to > > reach d. > > > > > > > > 2.2.5 could help you to better understand routing in Landmark > > Hierarchy. > > > > > > > > Thanks > > > > > > > > Yuan > > > > > > > > > > > > On Thu, 29 Jan 2004, Varun Goel wrote: > > > > > > > > > Hi, > > > > > > > > > > In the Landmark Hierarchy Example given in the research paper they > > have derived resulting path as 5 hops 1 greater than shortest path but > > it happened due to the fact that router having no entry for destination > > forwarded packet to level 2 router which was three hops away but if it > > had routed the packet to level 1 router which was two hops away and had > > an entry for it too, it could have resulted in shortest path. > > > > > > > > > > I am unable to understand why the packet wasnt routed to level 1 > > router instead? > > > > > > > > > > Please Clarify if anybody know..... > > > > > > > > > > Varun > > > > > -- > > > > > __________________________________________________________ > > > > > Sign-up for your own personalized E-mail at Mail.com > > > > > http://www.mail.com/?sr=signup > > > > > > > > > > Search Smarter - get the new eXact Search Bar for free! > > > > > http://www.exactsearchbar.com/ > > > > > > > > > > > > > > > -- > > > __________________________________________________________ > > > Sign-up for your own personalized E-mail at Mail.com > > > http://www.mail.com/?sr=signup > > > > > > Search Smarter - get the new eXact Search Bar for free! > > > http://www.exactsearchbar.com/ > > > > > > > > > > > ------------------------------ > > > > Message: 6 > > Date: Thu, 29 Jan 2004 20:36:13 -0800 > > From: "Devarshi Shah" > > Subject: RE: [Csci551-talk] socket programs > > To: > > Message-ID: <200401300436.i0U4aK007687@vapor.isi.edu> > > Content-Type: text/plain; charset="us-ascii" > > > > Hi Nishant, > > > > If I am correct, you would also need to link the appropriate libraries. > > (ex. "-l socket" option). > > > > Devarshi > > > > Devarshi P. Shah > > -------------------------------------------------------------- > > Graduate Student (Computer Science), > > University of Southern California, LA. > > -------------------------------------------------------------- > > Web : http://devarshi.shah.name > > email : devarshi@ieee.org > > Cell : 323-363-3791, Home :323-373-0318 > > -------------------------------------------------------------- > > > > > > -----Original Message----- > > From: csci551-talk-bounces@mailman.isi.edu > > [csci551-talk-bounces@mailman.isi.edu] On Behalf Of nishant > > agarwal > > Sent: Thursday, January 29, 2004 8:11 PM > > To: csci551-talk@ISI.EDU > > Subject: [Csci551-talk] socket programs > > > > I have done: > > #include and so on for sys/types.h , netinet/in.h, > > arpa/inet.h and all the files suggested in the local man pages. But yet > > I > > get symbol referencing errors for socket, bind and inet_addr . what cud > > be > > the problem. > > > > Also, what is the way to examine the status of my ports using netstat. > > How > > do I identify my localhost local address there so that I can grep and > > see > > the status of my machine ports. > > > > cheers > > Nishant > > > > > > > > > > ------------------------------ > > > > Message: 7 > > Date: Fri, 30 Jan 2004 08:39:09 -0800 > > From: John Heidemann > > Subject: Re: [Csci551-talk] socket programs > > To: nishant agarwal > > Cc: csci551-talk@ISI.EDU > > Message-ID: <200401301639.i0UGd9wN004840@dash.isi.edu> > > Content-Type: text/plain; charset=US-ASCII > > > > On Thu, 29 Jan 2004 20:11:14 PST, nishant agarwal wrote: > > >I have done: > > >#include and so on for sys/types.h , netinet/in.h, > > arpa/inet.h and all the files suggested in the local man pages. But yet > > I get symbol referencing errors for socket, bind and inet_addr . what > > cud be the problem. > > > > > > > Sockets have an abstract layer (socket/bind/etc.) that is protocol > > addressing independent, and then specific headers for each protocol > > family (IPv4, IPv6, OSI, etc.). > > > > if you want to get sockaddr_in or inet_addr, those are in the protocol > > family-specific header. On my system (a Linux box), see man ip(7) for > > details. > > > > > > >Also, what is the way to examine the status of my ports using netstat. > > How do I identify my localhost local address there so that I can grep > > and see the status of my machine ports. > > > > > >cheers > > >Nishant > > > > ifconfig will tell you about your local address, among other ways. > > > > -John Heidemann > > > > > > ------------------------------ > > > > _______________________________________________ > > Csci551-talk mailing list > > Csci551-talk@mailman.isi.edu > > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > > > > > End of Csci551-talk Digest, Vol 1, Issue 154 > > ******************************************** > > > > > > > > ------------------------------ > > > > _______________________________________________ > > Csci551-talk mailing list > > Csci551-talk@mailman.isi.edu > > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > > > > > End of Csci551-talk Digest, Vol 2, Issue 1 > > ****************************************** > > > > From liyuan at pollux.usc.edu Mon Feb 2 09:54:57 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Mon Feb 2 09:55:04 2004 Subject: [Csci551-talk] Need to change my office hour today In-Reply-To: <4b2fc4d500.4d5004b2fc@usc.edu> Message-ID: Dear Class, I will have to change my office hour today. I will hold my office hour today (Feb 02) from 6:00pm to 7:00pm in SAL229. And if you can not make it, please email me for appointments. The change is for this week only! Thanks Yuan From meghajos at usc.edu Mon Feb 2 12:11:30 2004 From: meghajos at usc.edu (megha joshi) Date: Mon Feb 2 12:11:39 2004 Subject: [Csci551-talk] (no subject) Message-ID: <2b12d92ac669.2ac6692b12d9@usc.edu> hi, I am not clear about what exactly does route oscillations mean So if anyone can please clarify, thanks, megha From johnh at ISI.EDU Mon Feb 2 14:31:14 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon Feb 2 14:31:18 2004 Subject: [Csci551-talk] homework 1 has been posted Message-ID: <200402022231.i12MVEkg008866@dash.isi.edu> CSci551 homework 1 has been posted to the class web site. -John Heidemann From johnh at ISI.EDU Mon Feb 2 15:03:38 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon Feb 2 15:03:41 2004 Subject: [Csci551-talk] socket programs In-Reply-To: Message-ID: <200402022303.i12N3c5V009658@dash.isi.edu> On Fri, 30 Jan 2004 12:26:42 PST, parag shah wrote: > >add "-lnsl -lsocket -lresolv" to the end of the compile command These are system dependent. What you give works for Solaris, but is unnecessary on FreeBSD or Linux. -John Heidemann From johnh at ISI.EDU Mon Feb 2 15:10:42 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon Feb 2 15:10:45 2004 Subject: [Csci551-talk] Re: Csci551-talk Digest, Vol 2, Issue 1 In-Reply-To: <4b2fc4d500.4d5004b2fc@usc.edu> Message-ID: <200402022310.i12NAgkI009705@dash.isi.edu> On Sun, 01 Feb 2004 12:10:58 PST, rahul pilani wrote: >Wouldnt it be better to use learn.usc.edu's discussion board .. rather than a mailing list like this one?.. i know this is more direct.. but that is more sophisticated and much more convenient.. >just a suggestion.. > I don't really want to have this discussion on the mailing list as it's not related to class content. However, any students that want to send me mail privately about their preferences, I'd be happy to consider them for future semesters (we're not changing this semester!). In particular, I'd like to know WHY other systems might be better (or more sophisticated and/or convenient). (Personally, I've used both web moderated bulletin boards and mailing lists, and found that web boards can be nice for some things, but you get ONLY what's on the web. While a mailing list means everyone can use their favorite mailers, and local tools to search the list, etc. Thus it's not so clear (at least to me) which is more sophisticated and convenient.) -John Heidemann From johnh at ISI.EDU Mon Feb 2 15:11:20 2004 From: johnh at ISI.EDU (John Heidemann) Date: Mon Feb 2 15:11:25 2004 Subject: [Csci551-talk] RE: Socket Program In-Reply-To: <001001c3e864$02f489e0$6701a8c0@LATUALP> Message-ID: <200402022311.i12NBKiZ009735@dash.isi.edu> On Sat, 31 Jan 2004 17:37:44 PST, "Aaron Tu" wrote: >Is there a homework assignment (socket program) that I'm not aware of? No. Not... yet. -John Heidemann From johnh at ISI.EDU Wed Feb 4 10:26:41 2004 From: johnh at ISI.EDU (John Heidemann) Date: Wed Feb 4 10:26:52 2004 Subject: [Csci551-talk] route oscillation [was: (no subject)] In-Reply-To: <2b12d92ac669.2ac6692b12d9@usc.edu> Message-ID: <200402041826.i14IQfVL021082@dash.isi.edu> On Mon, 02 Feb 2004 12:11:30 PST, megha joshi wrote: >hi, > >I am not clear about what exactly does route oscillations mean >So if anyone can please clarify, >thanks, >megha > Route oscillations are when a path between two hosts changes frequently over a short period of time, such as between two different paths. For example, assume a multi-homed host was connected to two ISPs, a primary and a backup. When the primary failed, the backup would take over. If the link with the primary was flakey, the route might osciallate (switch) between primary and backup frequently. Follow on questions to think about (no need to answer to the list): why would a route switch frequently? Why would an oscillation be a problem (or not be a problem)? What mechanisms exist to reduce the effects of oscillating routes? -John Heidemann From bshukla at usc.edu Wed Feb 4 13:30:20 2004 From: bshukla at usc.edu (bhavin shukla) Date: Wed Feb 4 13:30:32 2004 Subject: [Csci551-talk] questions on the shaikh paper Message-ID: <58139a584235.58423558139a@usc.edu> hi, I have the following questions on the Shaikh paper titled " Routing stability in congested networks" 1) What is the significance of the LOT ? Why has it been calculated for all the cases that they have experimented on? Why is it so imp? 2) How did they arrive on the values of parameters A1 and A2 in D2U cycle of BGP. Also what is a1 and a2 in the equations given? thanks bhavin From sharifza at aludra.usc.edu Thu Feb 5 14:00:31 2004 From: sharifza at aludra.usc.edu (Mehdi Sharifzadeh) Date: Thu Feb 5 14:00:40 2004 Subject: [Csci551-talk] Re: questions on the shaikh paper In-Reply-To: <200402052000.i15K0Fk19824@gamma.isi.edu> Message-ID: > I have the following questions on the Shaikh paper titled " Routing stability in congested networks" > > 1) What is the significance of the LOT ? Why has it been calculated for all the cases that they have experimented on? Why is it so imp? It is used to assign the cost to the transition from the last state to the init state. This is because it takes LOT time for the receiver timer to expire and find that the message has been lost. That's when the state of the whole system should be initiated to the init state. > > 2) How did they arrive on the values of parameters A1 and A2 in D2U cycle of BGP. Also what is a1 and a2 in the equations given? > Shaikh's master thesis [20] has a nice explanation of these parameters and Ra and Rb states (http://www.research.att.com/~ashaikh/papers/ms-thesis00.pdf). The initial 3*RTT is the value for RTO which is doubled in the case of each retransmission. --Mehdi > thanks > bhavin > > > > ------------------------------ > > _______________________________________________ > Csci551-talk mailing list > Csci551-talk@mailman.isi.edu > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > > End of Csci551-talk Digest, Vol 2, Issue 6 > ****************************************** > From khurania at usc.edu Fri Feb 6 01:12:14 2004 From: khurania at usc.edu (puneet khurania) Date: Fri Feb 6 01:12:16 2004 Subject: [Csci551-talk] Wanna make a Study Group Message-ID: <133c018581.18581133c0@usc.edu> Hi all, I want to make a study group to discuss papers, assignments etc. Anybody who is interested, please let me know. Regards, Puneet Khurania M.S. (Computer Science) University of Southern California Mailing Address: 3006, Royal Street Apartment No # 33, Los Angeles CA 90007 Ph: (213)-745-8766 From johnh at ISI.EDU Thu Feb 12 21:50:58 2004 From: johnh at ISI.EDU (John Heidemann) Date: Thu Feb 12 21:51:07 2004 Subject: [Csci551-talk] Re: questions on the shaikh paper In-Reply-To: Message-ID: <200402130550.i1D5owaH005985@dash.isi.edu> On Thu, 05 Feb 2004 14:00:31 PST, Mehdi Sharifzadeh wrote: > >> I have the following questions on the Shaikh paper titled " Routing stability in congested networks" >> >> 1) What is the significance of the LOT ? Why has it been calculated for all the cases that they have experimented on? Why is it so imp? > >It is used to assign the cost to the transition from the last state to the >init state. This is because it takes LOT time for the receiver timer to >expire and find that the message has been lost. That's when the state of >the whole system should be initiated to the init state. Exactly. It's just there to represent the remaining time until hold-time expires. >> 2) How did they arrive on the values of parameters A1 and A2 in D2U cycle of BGP. Also what is a1 and a2 in the equations given? >> > >Shaikh's master thesis [20] has a nice explanation of these parameters >and Ra and Rb states >(http://www.research.att.com/~ashaikh/papers/ms-thesis00.pdf). >The initial 3*RTT is the value for RTO which is >doubled in the case of each retransmission. And ultimately these values derive from TCP timeout handling. That's the subject of this week's class :-) -John Heidemann From dchopra at usc.edu Sun Feb 15 14:01:46 2004 From: dchopra at usc.edu (divya chopra) Date: Sun Feb 15 14:01:48 2004 Subject: [Csci551-talk] hw 1 question Message-ID: <7099bc70bf8a.70bf8a7099bc@usc.edu> Is RTP an internet transport protocol or an application layer protocol built over UDP/TCP? --Divya From naa at usc.edu Sun Feb 15 14:43:27 2004 From: naa at usc.edu (nishant agarwal) Date: Sun Feb 15 14:43:30 2004 Subject: [Csci551-talk] hw 1 question Message-ID: <72b65c72ef82.72ef8272b65c@usc.edu> See RFC 3550 for complete info on RTP. I am attaching it for ur reference. www.rfc-editor.org is the place to search for and get RFCs NishanT > > > ----- Original Message ----- > From: divya chopra > Date: Sunday, February 15, 2004 2:01 pm > Subject: [Csci551-talk] hw 1 question > > > Is RTP an internet transport protocol or an application layer protocol > > built > > over UDP/TCP? > > > > --Divya > > > > > From paragsha at usc.edu Sun Feb 15 14:53:40 2004 From: paragsha at usc.edu (parag shah) Date: Sun Feb 15 14:53:42 2004 Subject: [Csci551-talk] hw 1 question Message-ID: <71dc1a71c828.71c82871dc1a@usc.edu> this is kinda tricky q as RPT has all the imp properties of an transport layer protocol and can be used as one. however as of now it used as an application layer protocol and I have it specifically seen it classified as one at some instances. hope this helps parag ----- Original Message ----- From: divya chopra Date: Sunday, February 15, 2004 2:01 pm Subject: [Csci551-talk] hw 1 question > Is RTP an internet transport protocol or an application layer protocol > built > over UDP/TCP? > > --Divya > > From johnh at ISI.EDU Sun Feb 15 17:49:25 2004 From: johnh at ISI.EDU (John Heidemann) Date: Sun Feb 15 17:49:26 2004 Subject: [Csci551-talk] hw 1 question In-Reply-To: <71dc1a71c828.71c82871dc1a@usc.edu> Message-ID: <200402160149.i1G1nPcZ019997@dash.isi.edu> On Sun, 15 Feb 2004 14:53:40 PST, parag shah wrote: >this is kinda tricky q as RPT has all the imp properties of an transport layer >protocol and can be used as one. > >however as of now it used as an application layer protocol and I have it >specifically seen it classified as one at some instances. > >hope this helps As Parag says, RTP is an odd case. For the purposes of homework q1, you can think of RTP with UDP as a transport protocol. -John Heidemann From johnh at ISI.EDU Wed Feb 18 17:41:24 2004 From: johnh at ISI.EDU (John Heidemann) Date: Wed Feb 18 17:41:24 2004 Subject: [Csci551-talk] project A posted Message-ID: <200402190141.i1J1fO0U010906@dash.isi.edu> Project A has been posted to the class web site. -John Heidemann From vbahl at usc.edu Wed Feb 18 20:12:05 2004 From: vbahl at usc.edu (varun bahl) Date: Wed Feb 18 20:12:09 2004 Subject: [Csci551-talk] doubt in Question 1(g) Message-ID: <11a60fc11a6bf1.11a6bf111a60fc@usc.edu> HW1 , Q1 (g) asks which Internet Protocol is best suited for the application selected in part (e). However it does not state which layer of the protocol stack we need to write about.Please clarify the doubt. Cheers, Varun. From khurania at usc.edu Wed Feb 18 22:45:31 2004 From: khurania at usc.edu (puneet khurania) Date: Wed Feb 18 22:45:33 2004 Subject: [Csci551-talk] doubt on Ubiquitous Computing ques Message-ID: <1210aa912120f1.12120f11210aa9@usc.edu> Smart classroom in itself is a big application cosisting of diff sub applications and hence differnt Qos requirements. So, can we assume one sub application of it or not. Regards, Puneet Khurania M.S. (Computer Science) University of Southern California Mailing Address: 3006, Royal Street Apartment No # 33, Los Angeles CA 90007 Ph: (213)-745-8766 From khurania at usc.edu Wed Feb 18 23:35:33 2004 From: khurania at usc.edu (puneet khurania) Date: Wed Feb 18 23:35:36 2004 Subject: [Csci551-talk] One more doubt Message-ID: <1ccce1c3ed.1c3ed1ccce@usc.edu> What is the difference between 2(c) and 2(e) and what is diff b/w 2(h) 2(j) Regards, Puneet Khurania M.S. (Computer Science) University of Southern California Mailing Address: 3006, Royal Street Apartment No # 33, Los Angeles CA 90007 Ph: (213)-745-8766 From johnh at ISI.EDU Fri Feb 20 06:47:12 2004 From: johnh at ISI.EDU (John Heidemann) Date: Fri Feb 20 06:47:14 2004 Subject: [Csci551-talk] doubt in Question 1(g) In-Reply-To: <11a60fc11a6bf1.11a6bf111a60fc@usc.edu> Message-ID: <200402201447.i1KElC03001541@dash.isi.edu> On Wed, 18 Feb 2004 20:12:05 PST, varun bahl wrote: > >HW1 , Q1 (g) asks which Internet Protocol is best suited for the application >selected in part (e). > >However it does not state which layer of the protocol stack we need to write >about.Please clarify the doubt. If it doesn't specificy a particular layer, then you may choose. -John Heidemann From johnh at ISI.EDU Fri Feb 20 06:48:07 2004 From: johnh at ISI.EDU (John Heidemann) Date: Fri Feb 20 06:48:07 2004 Subject: [Csci551-talk] doubt on Ubiquitous Computing ques In-Reply-To: <1210aa912120f1.12120f11210aa9@usc.edu> Message-ID: <200402201448.i1KEm7vK001561@dash.isi.edu> On Wed, 18 Feb 2004 22:45:31 PST, puneet khurania wrote: >Smart classroom in itself is a big application cosisting of diff sub >applications and hence differnt Qos requirements. So, can we assume one sub >application of it or not. If you feel you need to do something more specific to make a clear point, then do so (and say why you did so). -John Heidemann From johnh at ISI.EDU Fri Feb 20 06:49:43 2004 From: johnh at ISI.EDU (John Heidemann) Date: Fri Feb 20 06:49:43 2004 Subject: [Csci551-talk] One more doubt In-Reply-To: <1ccce1c3ed.1c3ed1ccce@usc.edu> Message-ID: <200402201449.i1KEnhbd001604@dash.isi.edu> On Wed, 18 Feb 2004 23:35:33 PST, puneet khurania wrote: >What is the difference between 2(c) and 2(e) >and what is diff b/w 2(h) 2(j) One is about where the functionality should be placed. The other is about where the end point is. These may or may not be in the same place, and may or may not be for different reasons, as is discussed in the paper and was discussed in class. -John Heidemann From devarshi at ieee.org Fri Feb 20 21:09:34 2004 From: devarshi at ieee.org (Devarshi Shah) Date: Fri Feb 20 21:09:44 2004 Subject: [Csci551-talk] Project A : UDP Port Assignments Message-ID: <1077340174.18935.11.camel@maverick> Hi, Was reading up on the Project definition.. had a small question.. The description in Part A asks us to include the UDP Ports of the neighbour routers in the configuration downloaded from the Manager to a Router via TCP. Does this mean that the UDP Ports are preassigned ? (I dont think so - based on what was said earlier in the same section) If not, this implies that now there needs to be a 2 way communication between the Manager and the Router - in the first one the Router sends over its UDP port number, gets the config and then disconnects from the Manager.. Subsequently, it reconnects from to the Manager (after all the Routers have checked in with their UDP Ports) to get the UDP Ports of its neighbours.. Would this be correct ? Please correct me if I am going wrong somewhere in my reasoning !!! --- Devarshi From myopenideas at yahoo.com Sat Feb 21 11:04:11 2004 From: myopenideas at yahoo.com (Nirav Jasapara) Date: Sat Feb 21 11:04:23 2004 Subject: [Csci551-talk] Project A : UDP Port Assignments In-Reply-To: <1077340174.18935.11.camel@maverick> Message-ID: <20040221190411.13822.qmail@web41709.mail.yahoo.com> --- Devarshi Shah wrote: > Hi, > > Was reading up on the Project definition.. had a > small question.. > The description in Part A asks us to include the UDP > Ports of the > neighbour routers in the configuration downloaded > from the Manager to a > Router via TCP. Does this mean that the UDP Ports > are preassigned ? i think your child process can open up a UDP port [ at random .. htons(0) ] and can send this info to the parent ( Manager ) or you could preselect the udp ports and send these to the routers, to bind to, but this kind of hardcoding should be avioded. > > If not, this implies that now there needs to be a 2 > way communication > between the Manager and the Router - in the first > one the Router sends > over its UDP port number,gets the config and then > disconnects from the > Manager.. Subsequently, it reconnects from to the > Manager (after all the or you could just keep the socket open with the Manager process till the program runs [ which does provide you with two way communication ]... why disconnect and reconnect frequently > Routers have checked in with their UDP Ports) to get > the UDP Ports of > its neighbours.. Would this be correct ? Please > correct me if I am going > wrong somewhere in my reasoning !!! > > > --- Devarshi > 'stay free - Nirav Jasapara. " God is real .... unless defined int " __________________________________ Do you Yahoo!? Yahoo! Mail SpamGuard - Read only the mail you want. http://antispam.yahoo.com/tools From devarshi at ieee.org Sat Feb 21 12:49:07 2004 From: devarshi at ieee.org (Devarshi Shah) Date: Sat Feb 21 12:49:10 2004 Subject: [Csci551-talk] Project A : UDP Port Assignments In-Reply-To: <20040221190411.13822.qmail@web41709.mail.yahoo.com> References: <20040221190411.13822.qmail@web41709.mail.yahoo.com> Message-ID: <1077396547.5877.2.camel@maverick> On Sat, 2004-02-21 at 11:04, Nirav Jasapara wrote: > --- Devarshi Shah wrote: > > Hi, > > > > Was reading up on the Project definition.. had a > > small question.. > > The description in Part A asks us to include the UDP > > Ports of the > > neighbour routers in the configuration downloaded > > from the Manager to a > > Router via TCP. Does this mean that the UDP Ports > > are preassigned ? > > i think your child process can open up a UDP port [ at > random .. htons(0) ] and can send this info to the > parent ( Manager ) or you could preselect the udp > ports and send these to the routers, to bind to, but > this kind of hardcoding should be avioded. > I agree completely.... > > > > If not, this implies that now there needs to be a 2 > > way communication > > between the Manager and the Router - in the first > > one the Router sends > > over its UDP port number,gets the config and then > > disconnects from the > > Manager.. Subsequently, it reconnects from to the > > Manager (after all the > > or you could just keep the socket open with the > Manager process till the program runs [ which does > provide you with two way communication ]... why > disconnect and reconnect frequently > You have a point here... But my only concern was that there was no point in keeping the connection unless it was needed... However, it does make sense to keep it initially connected, and later, after the config info has been totally exchanged, to disconnect it ! > > Routers have checked in with their UDP Ports) to get > > the UDP Ports of > > its neighbours.. Would this be correct ? Please > > correct me if I am going > > wrong somewhere in my reasoning !!! > > > > > > --- Devarshi > > > > 'stay free > - Nirav Jasapara. > > " God is real .... unless defined int " > > __________________________________ > Do you Yahoo!? > Yahoo! Mail SpamGuard - Read only the mail you want. > http://antispam.yahoo.com/tools From myopenideas at yahoo.com Sat Feb 21 13:24:48 2004 From: myopenideas at yahoo.com (Nirav Jasapara) Date: Sat Feb 21 13:24:55 2004 Subject: [Csci551-talk] Project A : UDP Port Assignments In-Reply-To: <1077396547.5877.2.camel@maverick> Message-ID: <20040221212448.50949.qmail@web41704.mail.yahoo.com> --- Devarshi Shah wrote: > On Sat, 2004-02-21 at 11:04, Nirav Jasapara wrote: > > --- Devarshi Shah wrote: > > > Hi, > > > > > > Was reading up on the Project definition.. had a > > > small question.. > > > The description in Part A asks us to include the > UDP > > > Ports of the > > > neighbour routers in the configuration > downloaded > > > from the Manager to a > > > Router via TCP. Does this mean that the UDP > Ports > > > are preassigned ? > > > > i think your child process can open up a UDP port > [ at > > random .. htons(0) ] and can send this info to the > > parent ( Manager ) or you could preselect the udp > > ports and send these to the routers, to bind to, > but > > this kind of hardcoding should be avioded. > > > > I agree completely.... > > > > > > > If not, this implies that now there needs to be > a 2 > > > way communication > > > between the Manager and the Router - in the > first > > > one the Router sends > > > over its UDP port number,gets the config and > then > > > disconnects from the > > > Manager.. Subsequently, it reconnects from to > the > > > Manager (after all the > > > > or you could just keep the socket open with the > > Manager process till the program runs [ which does > > provide you with two way communication ]... why > > disconnect and reconnect frequently > > > > You have a point here... But my only concern was > that there > was no point in keeping the connection unless it was > needed... we might need the connection again as the Manager will ( in later stages ) need to pass the information about other AS's to all the routers > However, it does make sense to keep it initially > connected, and > later, after the config info has been totally > exchanged, to disconnect > it ! > > > > Routers have checked in with their UDP Ports) to > get > > > the UDP Ports of > > > its neighbours.. Would this be correct ? Please > > > correct me if I am going > > > wrong somewhere in my reasoning !!! > > > > > > > > > --- Devarshi > > > > > > > 'stay free > > - Nirav Jasapara. > > > > " God is real .... unless defined int " > > > > __________________________________ > > Do you Yahoo!? > > Yahoo! Mail SpamGuard - Read only the mail you > want. > > http://antispam.yahoo.com/tools > __________________________________ Do you Yahoo!? Yahoo! Mail SpamGuard - Read only the mail you want. http://antispam.yahoo.com/tools From poojapat at usc.edu Sun Feb 22 13:55:53 2004 From: poojapat at usc.edu (pooja patel) Date: Sun Feb 22 13:55:55 2004 Subject: [Csci551-talk] Re: Csci551-talk Digest, Vol 2, Issue 13 Message-ID: <7ce3bf7cd85d.7cd85d7ce3bf@usc.edu> An HTML attachment was scrubbed... URL: http://gamma.isi.edu/pipermail/csci551-talk/attachments/20040222/915d67d6/attachment.html From devarshi at ieee.org Sun Feb 22 16:42:48 2004 From: devarshi at ieee.org (Devarshi Shah) Date: Sun Feb 22 16:42:49 2004 Subject: [Csci551-talk] Re: Csci551-talk Digest, Vol 2, Issue 13 In-Reply-To: <7ce3bf7cd85d.7cd85d7ce3bf@usc.edu> References: <7ce3bf7cd85d.7cd85d7ce3bf@usc.edu> Message-ID: <1077496967.7701.1.camel@maverick> On Sun, 2004-02-22 at 13:55, pooja patel wrote: > Hi.. > > In Project A while parsing the input configuration file the > requirements state that we r supposed to detect comments anywhere in > the file > > Do we have to take "//" "*/" into consideration? The only comments that we need to take into consideration are the ones which begin with "#". > > Do we have to take a scenario like > > int i=5 ;//Value of i is 5 > > or > > int i=5 ;*/ This is fdognfgodfnhobdnb > > binghiunhuihoiugfnhiguhniu > > ghingfih gfihnfgh/* > > into consideration?? > > Pooja.----- Original Message ----- > > From: csci551-talk-request@mailman.isi.edu > > Date: Sunday, February 22, 2004 12:00 pm > > Subject: Csci551-talk Digest, Vol 2, Issue 13 > > > Send Csci551-talk mailing list submissions to > > csci551-talk@mailman.isi.edu > > > > To subscribe or unsubscribe via the World Wide Web, visit > > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > or, via email, send a message with subject or body 'help' to > > csci551-talk-request@mailman.isi.edu > > > > You can reach the person managing the list at > > csci551-talk-owner@mailman.isi.edu > > > > When replying, please edit your Subject line so it is more specific > > than "Re: Contents of Csci551-talk digest..." > > > > > > Today's Topics: > > > > 1. Re: Project A : UDP Port Assignments (Devarshi Shah) > > 2. Re: Project A : UDP Port Assignments (Nirav Jasapara) > > > > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Sat, 21 Feb 2004 12:49:07 -0800 > > From: Devarshi Shah > > Subject: Re: [Csci551-talk] Project A : UDP Port Assignments > > To: csci551-talk@ISI.EDU > > Message-ID: <1077396547.5877.2.camel@maverick> > > Content-Type: text/plain > > > > On Sat, 2004-02-21 at 11:04, Nirav Jasapara wrote: > > > --- Devarshi Shah wrote: > > > > Hi, > > > > > > > > Was reading up on the Project definition.. had a > > > > small question.. > > > > The description in Part A asks us to include the UDP > > > > Ports of the > > > > neighbour routers in the configuration downloaded > > > > from the Manager to a > > > > Router via TCP. Does this mean that the UDP Ports > > > > are preassigned ? > > > > > > i think your child process can open up a UDP port [ at > > > random .. htons(0) ] and can send this info to the > > > parent ( Manager ) or you could preselect the udp > > > ports and send these to the routers, to bind to, but > > > this kind of hardcoding should be avioded. > > > > > > > I agree completely.... > > > > > > > > > > If not, this implies that now there needs to be a 2 > > > > way communication > > > > between the Manager and the Router - in the first > > > > one the Router sends > > > > over its UDP port number,gets the config and then > > > > disconnects from the > > > > Manager.. Subsequently, it reconnects from to the > > > > Manager (after all the > > > > > > or you could just keep the socket open with the > > > Manager process till the program runs [ which does > > > provide you with two way communication ]... why > > > disconnect and reconnect frequently > > > > > > > You have a point here... But my only concern was that there > > was no point in keeping the connection unless it was needed... > > However, it does make sense to keep it initially connected, and > > later, after the config info has been totally exchanged, to > disconnect > > it ! > > > > > > Routers have checked in with their UDP Ports) to get > > > > the UDP Ports of > > > > its neighbours.. Would this be correct ? Please > > > > correct me if I am going > > > > wrong somewhere in my reasoning !!! > > > > > > > > > > > > --- Devarshi > > > > > > > > > > 'stay free > > > - Nirav Jasapara. > > > > > > " God is real .... unless defined int " > > > > > > __________________________________ > > > Do you Yahoo!? > > > Yahoo! Mail SpamGuard - Read only the mail you want. > > > http://antispam.yahoo.com/tools > > > > > > > > ------------------------------ > > > > Message: 2 > > Date: Sat, 21 Feb 2004 13:24:48 -0800 (PST) > > From: Nirav Jasapara > > Subject: Re: [Csci551-talk] Project A : UDP Port Assignments > > To: csci551-talk@mailman.isi.edu > > Message-ID: <20040221212448.50949.qmail@web41704.mail.yahoo.com> > > Content-Type: text/plain; charset=us-ascii > > > > --- Devarshi Shah wrote: > > > On Sat, 2004-02-21 at 11:04, Nirav Jasapara wrote: > > > > --- Devarshi Shah wrote: > > > > > Hi, > > > > > > > > > > Was reading up on the Project definition.. had a > > > > > small question.. > > > > > The description in Part A asks us to include the > > > UDP > > > > > Ports of the > > > > > neighbour routers in the configuration > > > downloaded > > > > > from the Manager to a > > > > > Router via TCP. Does this mean that the UDP > > > Ports > > > > > are preassigned ? > > > > > > > > i think your child process can open up a UDP port > > > [ at > > > > random .. htons(0) ] and can send this info to the > > > > parent ( Manager ) or you could preselect the udp > > > > ports and send these to the routers, to bind to, > > > but > > > > this kind of hardcoding should be avioded. > > > > > > > > > > I agree completely.... > > > > > > > > > > > > > If not, this implies that now there needs to be > > > a 2 > > > > > way communication > > > > > between the Manager and the Router - in the > > > first > > > > > one the Router sends > > > > > over its UDP port number,gets the config and > > > then > > > > > disconnects from the > > > > > Manager.. Subsequently, it reconnects from to > > > the > > > > > Manager (after all the > > > > > > > > or you could just keep the socket open with the > > > > Manager process till the program runs [ which does > > > > provide you with two way communication ]... why > > > > disconnect and reconnect frequently > > > > > > > > > > You have a point here... But my only concern was > > > that there > > > was no point in keeping the connection unless it was > > > needed... > > > > > > we might need the connection again as the Manager will > > ( in later stages ) need to pass the information about > > other AS's to all the routers > > > > > > > However, it does make sense to keep it initially > > > connected, and > > > later, after the config info has been totally > > > exchanged, to disconnect > > > it ! > > > > > > > > Routers have checked in with their UDP Ports) to > > > get > > > > > the UDP Ports of > > > > > its neighbours.. Would this be correct ? Please > > > > > correct me if I am going > > > > > wrong somewhere in my reasoning !!! > > > > > > > > > > > > > > > --- Devarshi > > > > > > > > > > > > > 'stay free > > > > - Nirav Jasapara. > > > > > > > > " God is real .... unless defined int " > > > > > > > > __________________________________ > > > > Do you Yahoo!? > > > > Yahoo! Mail SpamGuard - Read only the mail you > > > want. > > > > http://antispam.yahoo.com/tools > > > > > > > > > __________________________________ > > Do you Yahoo!? > > Yahoo! Mail SpamGuard - Read only the mail you want. > > http://antispam.yahoo.com/tools > > > > > > ------------------------------ > > > > _______________________________________________ > > Csci551-talk mailing list > > Csci551-talk@mailman.isi.edu > > http://mailman.isi.edu/mailman/listinfo/csci551-talk > > > > > > End of Csci551-talk Digest, Vol 2, Issue 13 > > ******************************************* > > From devarshi at ieee.org Sun Feb 22 17:01:02 2004 From: devarshi at ieee.org (Devarshi Shah) Date: Sun Feb 22 17:01:05 2004 Subject: [Csci551-talk] Need a few clarifications... Message-ID: <1077498062.7701.20.camel@maverick> Hi, I had a few questions about the project definition, especially the part 3. Was wondering if anyone else too had similar concerns... a) The project definition says that "... Note that each router will need to send multiple, increasingly longer routing messages....". How would this happen ? I mean, in Link State routing, I thought that the contents of the LSA message were the routes that the originator is responsible for, and the routers that the originator is connected to. This does not change when you reach steady state (i.e. having discovered your neighbours) unless routers go down... And so, shouldnt the length of the message be fixed (once you have discovered who your immediate neighbours are) ? b) Is a Router supposed to forward a LSA message that it receives from its neighbour to the other neighbours ? (Apart from sending out its own LSA message) --Devarshi From svenkata at usc.edu Sun Feb 22 18:06:18 2004 From: svenkata at usc.edu (suresh venkataramani) Date: Sun Feb 22 18:06:19 2004 Subject: [Csci551-talk] Need a few clarifications... Message-ID: <84e5398505a9.8505a984e539@usc.edu> An HTML attachment was scrubbed... URL: http://gamma.isi.edu/pipermail/csci551-talk/attachments/20040222/1a54f28d/attachment.html From myopenideas at yahoo.com Sun Feb 22 18:08:11 2004 From: myopenideas at yahoo.com (Nirav Jasapara) Date: Sun Feb 22 18:08:19 2004 Subject: [Csci551-talk] Need a few clarifications... In-Reply-To: <1077498062.7701.20.camel@maverick> Message-ID: <20040223020811.5911.qmail@web41704.mail.yahoo.com> --- Devarshi Shah wrote: > a) The project definition says that "... Note that > each router will need > to send multiple, increasingly longer routing > messages....". How would > this happen ? I mean, in Link State routing, I > thought that the > contents of the LSA message were the routes that the > originator is > responsible for, and the routers that the originator > is connected to. > This does not change when you reach steady state > (i.e. having discovered > your neighbours) unless routers go down... And so, > shouldnt the length > of the message be fixed (once you have discovered > who your immediate > neighbours are) ? 'increasing long routing messages ' refers to the startup stage when the routers start building their routing tables and sending out LSA messages. After some time this should converge and then steady stage is reached. > b) Is a Router supposed to forward a LSA message > that it receives from > its neighbour to the other neighbours ? (Apart from > sending out its own > LSA message) a router will send a LSA message only when it observes some change in its routing table. I dont think there is any _forwarding_ done. g'luck. - Nirav Jasapara. " The answer above is purely ment to confuse ppl and does not necessarily reflect the ideas of the author :) " __________________________________ Do you Yahoo!? Yahoo! Mail SpamGuard - Read only the mail you want. http://antispam.yahoo.com/tools From liyuan at pollux.usc.edu Mon Feb 23 01:11:47 2004 From: liyuan at pollux.usc.edu (Yuan Li) Date: Mon Feb 23 01:11:54 2004 Subject: [Csci551-talk] Need a few clarifications... In-Reply-To: <20040223020811.5911.qmail@web41704.mail.yahoo.com> Message-ID: I agree with what Nirav said in this post. Thanks Yuan On Sun, 22 Feb 2004, Nirav Jasapara wrote: > > --- Devarshi Shah wrote: > > > a) The project definition says that "... Note that > > each router will need > > to send multiple, increasingly longer routing > > messages....". How would > > this happen ? I mean, in Link State routing, I > > thought that the > > contents of the LSA message were the routes that the > > originator is > > responsible for, and the routers that the originator > > is connected to. > > This does not change when you reach steady state > > (i.e. having discovered > > your neighbours) unless routers go down... And so, > > shouldnt the length > > of the message be fixed (once you have discovered > > who your immediate > > neighbours are) ? > > 'increasing long routing messages ' refers to the > startup stage when the routers start building their > routing tables and sending out LSA messages. After > some time this should converge and then steady stage > is reached. > > > b) Is a Router supposed to forward a LSA message > > that it receives from > > its neighbour to the other neighbours ? (Apart from > > sending out its own > > LSA message) > > a router will send a LSA message only when it observes > some change in its routing table. I dont think there > is any _forwarding_ done. > > g'luck. > - Nirav Jasapara. > > " The answer above is purely ment to confuse ppl and > does not necessarily reflect the ideas of the author > :) " > > __________________________________ > Do you Yahoo!? > Yahoo! Mail SpamGuard - Read only the mail you want. > http://antispam.yahoo.com/tools > From myopenideas at yahoo.com Mon Feb 23 19:48:19 2004 From: myopenideas at yahoo.com (Nirav Jasapara) Date: Mon Feb 23 19:48:27 2004 Subject: [Csci551-talk] Need a few clarifications... In-Reply-To: <20040223020811.5911.qmail@web41704.mail.yahoo.com> Message-ID: <20040224034819.81518.qmail@web41706.mail.yahoo.com> > > b) Is a Router supposed to forward a LSA message > > that it receives from > > its neighbour to the other neighbours ? (Apart > from > > sending out its own > > LSA message) > > a router will send a LSA message only when it > observes > some change in its routing table. I dont think there > is any _forwarding_ done. After resolving my self_confusion between DV routing and LS routing i think that we are supposed to forward routes to all the routers in the AS ... though i still dont understand the use of many feilds in the given LSA message which seems to be different from the LSA messages used in OSPF ( IS-IS anyone ?? ). i guess things will gradually get more clear as the deadline arrives :) cheers, Nirav. __________________________________ Do you Yahoo!? Yahoo! Mail SpamGuard - Read only the mail you want. http://antispam.yahoo.com/tools From johnh at ISI.EDU Tue Feb 24 21:52:35 2004 From: johnh at ISI.EDU (John Heidemann) Date: Tue Feb 24 21:52:37 2004 Subject: [Csci551-talk] Project A : UDP Port Assignments In-Reply-To: <20040221212448.50949.qmail@web41704.mail.yahoo.com> Message-ID: <200402250552.i1P5qZMk000740@dash.isi.edu> On Sat, 21 Feb 2004 13:24:48 PST, Nirav Jasapara wrote: >--- Devarshi Shah wrote: >> On Sat, 2004-02-21 at 11:04, Nirav Jasapara wrote: >> > --- Devarshi Shah wrote: >> > > Hi, >> > > >> > > Was reading up on the Project definition.. had a >> > > small question.. >> > > The description in Part A asks us to include the >> UDP >> > > Ports of the >> > > neighbour routers in the configuration >> downloaded >> > > from the Manager to a >> > > Router via TCP. Does this mean that the UDP >> Ports >> > > are preassigned ? >> > >> > i think your child process can open up a UDP port >> [ at >> > random .. htons(0) ] and can send this info to the >> > parent ( Manager ) or you could preselect the udp >> > ports and send these to the routers, to bind to, >> but >> > this kind of hardcoding should be avioded. Hardcoding ports is bad. The assignment suggests how ports should be handled (wihtout hardcoding). >> > > >> > > If not, this implies that now there needs to be >> a 2 >> > > way communication >> > > between the Manager and the Router - in the >> first >> > > one the Router sends >> > > over its UDP port number,gets the config and >> then >> > > disconnects from the >> > > Manager.. Subsequently, it reconnects from to >> the >> > > Manager (after all the >> > >> > or you could just keep the socket open with the >> > Manager process till the program runs [ which does >> > provide you with two way communication ]... why >> > disconnect and reconnect frequently Whether or not you keep a manger to router connection open is up to you, unless it's necessary. I believe for Project A, once the routers are going there is no need for the manager to keep talking to them. (Proj B may change this.) -John Heidemann From johnh at ISI.EDU Tue Feb 24 21:55:51 2004 From: johnh at ISI.EDU (John Heidemann) Date: Tue Feb 24 21:55:51 2004 Subject: [Csci551-talk] Re: Csci551-talk Digest, Vol 2, Issue 13 In-Reply-To: <7ce3bf7cd85d.7cd85d7ce3bf@usc.edu> Message-ID: <200402250555.i1P5tpNO000791@dash.isi.edu> On Sun, 22 Feb 2004 13:55:53 PST, pooja patel wrote: >Hi.. > >In Project A while parsing the input configuration file the requirements state >that we r supposed to detect comments anywhere in the file > >Do we have to take "//" "*/" into consideration? > >Do we have to take a scenario like > >int i=5 ;//Value of i is 5 > >or > >int i=5 ;*/ This is fdognfgodfnhobdnb > >binghiunhuihoiugfnhiguhniu > >ghingfih gfihnfgh/* > >into consideration?? Section 6 that comment lines begin with # (as the first non-whitespace character). For this project, that is the only kind of comment. (The project also doesn't use statements like "int i=5", except maybe in your source code.) -John Heidemann From johnh at ISI.EDU Tue Feb 24 22:01:16 2004 From: johnh at ISI.EDU (John Heidemann) Date: Tue Feb 24 22:01:24 2004 Subject: [Csci551-talk] Need a few clarifications... In-Reply-To: <1077498062.7701.20.camel@maverick> Message-ID: <200402250601.i1P61GoS000856@dash.isi.edu> On Sun, 22 Feb 2004 17:01:02 PST, Devarshi Shah wrote: >Hi, > >I had a few questions about the project definition, especially the part >3. Was wondering if anyone else too had similar concerns... > >a) The project definition says that "... Note that each router will need >to send multiple, increasingly longer routing messages....". How would >this happen ? I mean, in Link State routing, I thought that the >contents of the LSA message were the routes that the originator is >responsible for, and the routers that the originator is connected to. >This does not change when you reach steady state (i.e. having discovered >your neighbours) unless routers go down... And so, shouldnt the length >of the message be fixed (once you have discovered who your immediate >neighbours are) ? You're correct. This was a mistaken holdover from last year's project :-) Also, you will NOT use path information to avoid loops (the next sentence). >b) Is a Router supposed to forward a LSA message that it receives from >its neighbour to the other neighbours ? (Apart from sending out its own >LSA message) You may want to review "reliable flooding" in link state routing. -John Heidemann From johnh at ISI.EDU Tue Feb 24 22:03:15 2004 From: johnh at ISI.EDU (John Heidemann) Date: Tue Feb 24 22:03:17 2004 Subject: [Csci551-talk] Need a few clarifications... In-Reply-To: <20040223020811.5911.qmail@web41704.mail.yahoo.com> Message-ID: <200402250603.i1P63FUE000882@dash.isi.edu> On Sun, 22 Feb 2004 18:08:11 PST, Nirav Jasapara wrote: > >--- Devarshi Shah wrote: > >> a) The project definition says that "... Note that >> each router will need >> to send multiple, increasingly longer routing >> messages....". How would >> this happen ? I mean, in Link State routing, I >> thought that the >> contents of the LSA message were the routes that the >> originator is >> responsible for, and the routers that the originator >> is connected to. >> This does not change when you reach steady state >> (i.e. having discovered >> your neighbours) unless routers go down... And so, >> shouldnt the length >> of the message be fixed (once you have discovered >> who your immediate >> neighbours are) ? > >'increasing long routing messages ' refers to the >startup stage when the routers start building their >routing tables and sending out LSA messages. After >some time this should converge and then steady stage >is reached. Actually, not quite... my statement in the proposal about increasing message sizes was simply incorrect. > >> b) Is a Router supposed to forward a LSA message >> that it receives from >> its neighbour to the other neighbours ? (Apart from >> sending out its own >> LSA message) > >a router will send a LSA message only when it observes >some change in its routing table. I dont think there >is any _forwarding_ done. That's not true for LS routing (although it would be true for DV). In LS routing you send out messages when your LINK status changes, and nodes propgate these via reliable flooding. -John Heidemann From johnh at ISI.EDU Fri Feb 27 14:06:14 2004 From: johnh at ISI.EDU (John Heidemann) Date: Fri Feb 27 14:06:09 2004 Subject: [Csci551-talk] homework 2 has been posted Message-ID: <200402272206.i1RM6Es2019675@dash.isi.edu> Homework 2 has been posted to the class web site. -John Heidemann From pilani at usc.edu Sat Feb 28 02:37:36 2004 From: pilani at usc.edu (rahul pilani) Date: Sat Feb 28 02:37:39 2004 Subject: [Csci551-talk] Re: Query regarding the configuration file Message-ID: <30f83f30e856.30e85630f83f@usc.edu> Was wondering something about the configuration file. It is mentioned in the specs that comments "#" can appear anywhere.. anywhere means start of every line right?.. Also can the order of the various values in the config file change?.. Rahul From johnh at ISI.EDU Sat Feb 28 08:41:54 2004 From: johnh at ISI.EDU (John Heidemann) Date: Sat Feb 28 08:41:55 2004 Subject: [Csci551-talk] Re: Query regarding the configuration file In-Reply-To: <30f83f30e856.30e85630f83f@usc.edu> Message-ID: <200402281641.i1SGftvE030029@dash.isi.edu> On Sat, 28 Feb 2004 02:37:36 PST, rahul pilani wrote: >Was wondering something about the configuration file. It is mentioned in the >specs that comments "#" >can appear anywhere.. anywhere means start of every line right?.. Yes. >Also can the order of the various values in the config file change?.. Which values? The meaning of some values is defined by their order. They can't change. -John Heidemann From devarshi at ieee.org Sun Feb 29 19:35:09 2004 From: devarshi at ieee.org (Devarshi Shah) Date: Sun Feb 29 19:35:11 2004 Subject: [Csci551-talk] Porting from Linux to Solaris... Message-ID: <1078112108.5701.5.camel@maverick> Hi, I just tried to move a portion of the project code which I have running perfectly on Linux to Solaris (aludra), and for some wierd reason, upon running the binary, got a runtime error saying "Broken Pipe". I wonder if anyone has come across such an issue and knows where the incompatibility may be ! -- Devarshi From naa at usc.edu Sun Feb 29 19:45:56 2004 From: naa at usc.edu (nishant agarwal) Date: Sun Feb 29 19:45:58 2004 Subject: [Csci551-talk] Porting from Linux to Solaris... Message-ID: <57641b5722d8.5722d857641b@usc.edu> Ignore SIGPIPE by signal (SIGPIPE, SIG_IGN) and check for the return error when read / write fails due to broken pipe ----- Original Message ----- From: Devarshi Shah Date: Sunday, February 29, 2004 7:35 pm Subject: [Csci551-talk] Porting from Linux to Solaris... > Hi, > > I just tried to move a portion of the project code which I have running > perfectly on Linux to Solaris (aludra), and for some wierd reason, upon > running the binary, got a runtime error saying "Broken Pipe". I wonder > if anyone has come across such an issue and knows where the > incompatibility may be ! > > -- Devarshi > >