From renaud at merit.edu Wed Mar 13 19:32:13 1996 From: renaud at merit.edu (Brian Renaud) Date: Wed, 13 Mar 1996 13:32:13 -0500 Subject: fyi: minor updates for database software Message-ID: <199603131832.NAA09214@merit.edu> I've installed a version of the database software (still on the 1.0.3 base) with the following changes: aspathre: support AS-MACROs in AS extended path regular expressions make grammar conform to AS path RE grammar currently in the RA Toolset (version 3.2) ripe181/src: support line continuations with a '\' at the end of the line (strip out the '\' and do the line joins on input) support AS Macro syntax of /^AS\-[A-Z]([_-]*[A-Z0-9]+)*$/ support RIPE 181 routing policy expressions with the "OR" keyword support community list syntax of /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$ implement quad2int and int2quad versions that work with perl4 & perl5 change PGP scanning to handle possible security hole add RIPE 1.1beta support for passwords on authorise and override send data base responses back to anyone on the cc list for the original message to auto-dbm clean up some messages support deletes of aut-nums which contain optional ("sugar") words (from, to, accept) This distribution is available at: ftp://ftp.ra.net/routing.arbiter/radb/tools/dbase_960312_full.tar.gz for full images of all the files, or ftp://ftp.ra.net/routing.arbiter/radb/tools/dbase_960312_RIPE-1.0.3_diff.tar.g z for a version with a patchable diff file (against RIPE version 1.0.3). I should have another update fairly soon which puts all of the extended path expression handling into perl, so that I can remove the stand-alone syntax checker (aspathre) hack. -Brian -------- Logged at Wed Mar 13 20:11:14 MET 1996 --------- From cengiz at ISI.EDU Wed Mar 13 20:12:08 1996 From: cengiz at ISI.EDU (Cengiz Alaettinoglu) Date: Wed, 13 Mar 1996 11:12:08 -0800 Subject: polling for ripe dbase software changes Message-ID: <199603131912.AA16151@cat.isi.edu> Hi folks, RA is considering: 1) for db-update to bounce back objects which uses ripe-181 style line continuations as error. 2) making the sugar words mandatory on as-in/out, i.e. from to announce accept. 3) not deleting these sugar words from the objects while storing them into the database file (and not reinserting them on whois). We would like to know, if there is a strong opinion against this. Particularly, we would like to know if things would break in your site if we did this. Thanks. Cengiz -- Cengiz Alaettinoglu Information Sciences Institute (310) 822-1511 University of Southern California http://www.isi.edu/~cengiz -------- Logged at Wed Mar 13 22:34:49 MET 1996 --------- From Tony.Bates at mci.net Wed Mar 13 22:33:44 1996 From: Tony.Bates at mci.net (Tony Bates) Date: Wed, 13 Mar 1996 16:33:44 -0500 Subject: fyi: minor updates for database software In-Reply-To: Your message of Wed, 13 Mar 1996 13:32:13 EST. <199603131832.NAA09214@merit.edu> Message-ID: <199603132133.QAA19357@postoffice.reston.mci.net> Brian, a few comments. Why are you making these changes before we have an RPSL spec. These are basically breaking RFC1786 in some cases without having anything else actually defined. Brian Renaud writes: * * I've installed a version of the database software (still on the 1.0.3 * base) with the following changes: * ... * support line continuations with a '\' at the end of the line * (strip out the '\' and do the line joins on input) * * support AS Macro syntax of /^AS\-[A-Z]([_-]*[A-Z0-9]+)*$/ * To quote from RFC1786. ... Each attribute has the following syntax: as-macro: The name of a macro containing at least two Autonomous Systems grouped together for ease of administration. Format: AS- The should be in upper case and not contain any spe- cial characters. Example: as-macro: AS-EBONE Status: mandatory, only one line allowed ... So this breaks this. * support RIPE 181 routing policy expressions with the "OR" keyword * Uhh.. yeah good catch. * support community list syntax of /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$ * This means I can have a community of AS-MACRO now - something we wanted to avoid. --Tony. -------- Logged at Wed Mar 13 22:53:45 MET 1996 --------- From renaud at merit.edu Wed Mar 13 22:52:21 1996 From: renaud at merit.edu (Brian Renaud) Date: Wed, 13 Mar 1996 16:52:21 -0500 Subject: fyi: minor updates for database software In-Reply-To: Your message of Wed, 13 Mar 1996 16:33:44 -0500. <199603132133.QAA19357@postoffice.reston.mci.net> Message-ID: <199603132152.QAA20293@merit.edu> > Why are you making these changes before we have an RPSL spec. These > are basically breaking RFC1786 in some cases without having anything > else actually defined. The syntax changes are in response to requests/common errors of users. Perhaps I was rather premature on this though. I'll be glad to back out the as-macro and community syntax changes if people feel that is the right thing to do. > * support community list syntax of /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$ > * > This means I can have a community of AS-MACRO now - something we > wanted to avoid. Actually no, the change comment wasn't as comprehensive as it could have been. The follow up check to prevent communities from being keywords or of the form AS-xxx is still in the code. -Brian -------- Logged at Wed Mar 13 22:55:05 MET 1996 --------- From Tony.Bates at mci.net Wed Mar 13 22:54:30 1996 From: Tony.Bates at mci.net (Tony Bates) Date: Wed, 13 Mar 1996 16:54:30 -0500 Subject: fyi: minor updates for database software In-Reply-To: Your message of Wed, 13 Mar 1996 16:52:21 EST. <199603132152.QAA20293@merit.edu> Message-ID: <199603132154.QAA19902@postoffice.reston.mci.net> Brian Renaud writes: * * > Why are you making these changes before we have an RPSL spec. These * > are basically breaking RFC1786 in some cases without having anything * > else actually defined. * * The syntax changes are in response to requests/common errors of users. * * Perhaps I was rather premature on this though. I'll be glad to back * out the as-macro and community syntax changes if people feel that * is the right thing to do. * I think so but am happy to hear other peoples opinion. * > * support community list syntax of /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$ * > * * > This means I can have a community of AS-MACRO now - something we * > wanted to avoid. * * Actually no, the change comment wasn't as comprehensive as it could have * been. The follow up check to prevent communities from being keywords or * of the form AS-xxx is still in the code. * Okay, yes - I just checked the code. Sorry about that one. --Tony -------- Logged at Wed Mar 13 23:01:10 MET 1996 --------- From cengiz at ISI.EDU Wed Mar 13 23:02:12 1996 From: cengiz at ISI.EDU (Cengiz Alaettinoglu) Date: Wed, 13 Mar 1996 14:02:12 -0800 Subject: fyi: minor updates for database software In-Reply-To: <199603132154.QAA19902@postoffice.reston.mci.net> References: <199603132152.QAA20293@merit.edu> <199603132154.QAA19902@postoffice.reston.mci.net> Message-ID: <199603132202.AA16443@cat.isi.edu> I think it is OK to make these changes, because, 1) they are backward compatible with RFC1786 2) we already have other extensions to RFC1786 implemented Tony Bates (Tony.Bates at mci.net) on March 13: > > Brian Renaud writes: > * > * > Why are you making these changes before we have an RPSL spec. These > * > are basically breaking RFC1786 in some cases without having anything > * > else actually defined. > * > * The syntax changes are in response to requests/common errors of users. > * > * Perhaps I was rather premature on this though. I'll be glad to back > * out the as-macro and community syntax changes if people feel that > * is the right thing to do. > * > I think so but am happy to hear other peoples opinion. > > * > * support community list syntax of /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$ > * > * > * > This means I can have a community of AS-MACRO now - something we > * > wanted to avoid. > * > * Actually no, the change comment wasn't as comprehensive as it could have > * been. The follow up check to prevent communities from being keywords or > * of the form AS-xxx is still in the code. > * > Okay, > yes - I just checked the code. Sorry about that one. > > --Tony Cengiz -- Cengiz Alaettinoglu Information Sciences Institute (310) 822-1511 University of Southern California http://www.isi.edu/~cengiz -------- Logged at Thu Mar 14 18:59:33 MET 1996 --------- From renaud at merit.edu Thu Mar 14 18:59:30 1996 From: renaud at merit.edu (Brian Renaud) Date: Thu, 14 Mar 1996 12:59:30 -0500 Subject: fyi: minor updates for database software In-Reply-To: Your message of Wed, 13 Mar 1996 13:32:13 -0500. Message-ID: <199603141759.MAA25687@merit.edu> Well, it seems that we did a really bad job of testing one of the changes in the final version of this code. If you grabbed a copy of the code with the intent of using it, please get a new copy: ftp://ftp.ra.net/routing.arbiter/radb/tools/dbase_960314_full.tar.gz or ftp://ftp.ra.net/routing.arbiter/radb/tools/dbase_960314_RIPE-1.0.3_diff.tar.gz which contains a revised version of dbadd.pl. My apologies. -Brian at Wed, 13 Mar 1996 13:32:13 EST, you wrote: > > I've installed a version of the database software (still on the 1.0.3 > base) with the following changes: > > aspathre: > support AS-MACROs in AS extended path regular expressions > > make grammar conform to AS path RE grammar currently in the > RA Toolset (version 3.2) > > ripe181/src: > support line continuations with a '\' at the end of the line > (strip out the '\' and do the line joins on input) > > support AS Macro syntax of /^AS\-[A-Z]([_-]*[A-Z0-9]+)*$/ > > support RIPE 181 routing policy expressions with the "OR" keyword > > support community list syntax of /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$ > > implement quad2int and int2quad versions that work with perl4 & perl5 > > change PGP scanning to handle possible security hole > > add RIPE 1.1beta support for passwords on authorise and override > > send data base responses back to anyone on the cc list for the > original message to auto-dbm > > clean up some messages > > support deletes of aut-nums which contain optional ("sugar") words > (from, to, accept) > > This distribution is available at: > > ftp://ftp.ra.net/routing.arbiter/radb/tools/dbase_960312_full.tar.gz > > for full images of all the files, or > > ftp://ftp.ra.net/routing.arbiter/radb/tools/dbase_960312_RIPE-1.0.3_diff.tar .g > z > > for a version with a patchable diff file (against RIPE version 1.0.3). > > I should have another update fairly soon which puts all of the extended > path expression handling into perl, so that I can remove the stand-alone > syntax checker (aspathre) hack. > > -Brian > -------- Logged at Thu Mar 14 19:10:26 MET 1996 --------- From curtis at ans.net Thu Mar 14 19:08:13 1996 From: curtis at ans.net (Curtis Villamizar) Date: Thu, 14 Mar 1996 13:08:13 -0500 Subject: polling for ripe dbase software changes In-Reply-To: Your message of "Wed, 13 Mar 1996 11:12:08 PST." <199603131912.AA16151@cat.isi.edu> Message-ID: <199603141808.NAA17728@brookfield.ans.net> In message <199603131912.AA16151 at cat.isi.edu>, Cengiz Alaettinoglu writes: > > Hi folks, > > RA is considering: > 1) for db-update to bounce back objects which uses ripe-181 style > line continuations as error. > 2) making the sugar words mandatory on as-in/out, i.e. from to > announce accept. > 3) not deleting these sugar words from the objects while storing them > into the database file (and not reinserting them on whois). > > We would like to know, if there is a strong opinion against > this. Particularly, we would like to know if things would break in > your site if we did this. > > Thanks. > > Cengiz > > -- > Cengiz Alaettinoglu Information Sciences Institute > (310) 822-1511 University of Southern California > http://www.isi.edu/~cengiz It would be nice if we could get these changes into all of the databases. If something breaks here, we'll fix it. Please send advanced notice before this goes into effect. (And shoot for a Tuesady night if you can). Curtis -------- Logged at Thu Mar 14 21:33:02 MET 1996 --------- From curtis at ans.net Thu Mar 14 21:31:17 1996 From: curtis at ans.net (Curtis Villamizar) Date: Thu, 14 Mar 1996 15:31:17 -0500 Subject: fyi: minor updates for database software In-Reply-To: Your message of "Wed, 13 Mar 1996 16:33:44 EST." <199603132133.QAA19357@postoffice.reston.mci.net> Message-ID: <199603142031.PAA18362@brookfield.ans.net> In message <199603132133.QAA19357 at postoffice.reston.mci.net>, Tony Bates writes : > * support AS Macro syntax of /^AS\-[A-Z]([_-]*[A-Z0-9]+)*$/ The RFC sez: > Format: > AS- > > The should be in upper case and not contain any spe- > cial characters. > ... > So this breaks this. I would like to have at least digits so I can look for AS macros with other AS numbers embedded in the name. So why not go the whole way. We have a use for this. We will need this for some of the aggregation support. > * support community list syntax of /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$ > * > This means I can have a community of AS-MACRO now - something we > wanted to avoid. Can we just say: /^[A-Za-z]([_-]*[A-Za-z0-9]+)*$/ && !/^AS\-[A-Z]([_-]*[A-Z0-9]+)*$/ (btw- non-technical users would really enjoy this exchange). > --Tony. I'd prefer not to stand still. How about a 2 page RFC describing incremental changes to RFC1786? Curtis -------- Logged at Mon Mar 18 15:24:35 MET 1996 --------- From David.Kessens at ripe.net Mon Mar 18 15:24:33 1996 From: David.Kessens at ripe.net (David.Kessens at ripe.net) Date: Mon, 18 Mar 1996 15:24:33 +0100 (MET) Subject: fyi: minor updates for database software In-Reply-To: <199603131832.NAA09214@merit.edu> from "Brian Renaud" at Mar 13, 96 01:32:13 pm Message-ID: <9603181424.AA01145@belegen.ripe.net> A non-text attachment was scrubbed... Name: not available Type: text Size: 1948 bytes Desc: not available Url : https://www.ripe.net/mailman/private/rr-impl/attachments/19960318/3c7331fb/attachment.pl From David.Kessens at ripe.net Thu Mar 28 15:27:33 1996 From: David.Kessens at ripe.net (David.Kessens at ripe.net) Date: Thu, 28 Mar 1996 15:27:33 +0100 (MET) Subject: new bug fix release of prtraceroute Message-ID: <9603281427.AA15125@belegen.ripe.net> A non-text attachment was scrubbed... Name: not available Type: text Size: 1562 bytes Desc: not available Url : https://www.ripe.net/mailman/private/rr-impl/attachments/19960328/1a9a28a2/attachment.pl