stevo@elroy.UUCP (10/14/87)
Hello there. I am the co-SA for machine "elroy" at JPL in Pasadena, CA. This posting concerns a bad UUCP map entry for a machine here at JPL that has caused many people/mailers to attempt to route mail over a non-existent UUCP link. I am not sure that the news groups I am posting to are the most appropriate ones, so if you aren't interested in this type of problem, accept my apologies and hit 'n' now. Anyway, here is the scoop. Here at JPL there is a machine called "devvax." The problem stems from the fact that there is a machine at Cornell that goes by the same name. When another JPL machine, "jplpro", published its UUCP map entry, it advertised a LOCAL connection to devvax. Since JPL's devvax has not published a map entry but there is already one for Cornell's devvax, pathalias etc. think that jplpro has a very cheap link to Cornell. So, if you are closer to JPL than you are to Cornell, you may find that the pathalias-generated route to Cornell attempts to use the non-existent link. A fix for the problem is in the works, and consists mostly of fixing jplpro's map entry (probably by renaming devvax to jpl-devvax or something). In the meantime, this posting is intended to inform SA's that may have experienced this problem or who know users that have. I should say here that I have no affiliation with any machine in this discussion other than elroy. The problem came to light because the path to jplpro usually goes through elroy. After I started seeing a lot of bounced mail coming back from JPL's devvax, I investigated. The SA's of jplpro and (JPL's) devvax have been informed, and have told me that they are "working on it." -steve /* Steve Groom, Image Analysis Systems Group, M/S 168-522 * Jet Propulsion Laboratory, Pasadena, CA 91109 * ARPA: stevo@elroy.jpl.nasa.gov UUCP: ...!cit-vax!elroy!stevo * "Whaddaya want for nothin'?!?!" */