1*2aef6930SMark Murray@(#) README.IRIX 1.2 94/12/28 18:45:58 2*2aef6930SMark Murray 3*2aef6930SMark MurrayIn the past few months I received several messages with questions from 4*2aef6930SMark Murraypeople that tried to use my tcp wrapper on IRIX 5.x. Some mysteries 5*2aef6930SMark Murraycould be solved via email, and then some remained. 6*2aef6930SMark Murray 7*2aef6930SMark MurrayToday I finally had a chance to do some tests on someones IRIX 5.2 8*2aef6930SMark Murraysystem. Here is my first-hand experience with wrapper release 6.3. 9*2aef6930SMark Murray 10*2aef6930SMark Murray(1) Inetd is broken. Normally one edits inetd.conf, sends a HUP signal 11*2aef6930SMark Murray to inetd and that's it. With IRIX evil things happen: inetd is too 12*2aef6930SMark Murray stupid to remember that it is already listening on a port. 13*2aef6930SMark Murray 14*2aef6930SMark Murray In order to modify an entry in inetd.conf, first comment it out 15*2aef6930SMark Murray with a # at the beginning of the line, kill -HUP the inetd, then 16*2aef6930SMark Murray uncomment the inetd.conf entry and kill -HUP again. 17*2aef6930SMark Murray 18*2aef6930SMark Murray Even with this amount of care I have seen inetd messing up, like 19*2aef6930SMark Murray calling rusersd when I make a talk connection. Even killing and 20*2aef6930SMark Murray restarting inetd does not solve all problems. 21*2aef6930SMark Murray 22*2aef6930SMark Murray I find it hard to believe, it but the best thing to do with IRIX is 23*2aef6930SMark Murray to reboot after changing inetd.conf. 24*2aef6930SMark Murray 25*2aef6930SMark Murray(2) When tcpd is built according to the irix4 Makefile rules, it 26*2aef6930SMark Murray appears to work as expected with TCP-based services such as 27*2aef6930SMark Murray fingerd, and with UDP-based services such as ntalk and tftp. 28*2aef6930SMark Murray 29*2aef6930SMark Murray(3) It does NOT work with RPC over UDP services such as rusersd and 30*2aef6930SMark Murray rstatd: the wrapper hangs in the recvfrom() system call, and I 31*2aef6930SMark Murray have spent several hours looking for ways to work around it. No 32*2aef6930SMark Murray way. After finding that none of the applicable socket primitives 33*2aef6930SMark Murray can be made to work (recvfrom recvmsg) I give up. So, the IRIX RPC 34*2aef6930SMark Murray services cannot be wrapped until SGI fixes their system so that it 35*2aef6930SMark Murray works like everyone elses code (HP Sun Dec AIX and so on). 36*2aef6930SMark Murray 37*2aef6930SMark Murray(4) I didn't even bother to try the RPC over TCP services. 38*2aef6930SMark Murray 39*2aef6930SMark Murray(5) When an IRIX 5.2 system is a NIS client, it can have problems with 40*2aef6930SMark Murray hosts that have more than one address: the wrapper will see only 41*2aef6930SMark Murray one address, and may complain when PARANOID mode is on. The fix is 42*2aef6930SMark Murray to change the name service lookup order in /etc/resolv.conf so that 43*2aef6930SMark Murray your system tries DNS before NIS (hostresorder bind nis local). 44*2aef6930SMark Murray 45*2aef6930SMark Murray(6) IRIX 5.2 is not System V.4, and it shows. Do not link with the 46*2aef6930SMark Murray -lsocket and -lnsl libraries. They are completely broken, and the 47*2aef6930SMark Murray wrapper will be unable to figure out the client internet address. 48*2aef6930SMark Murray So, TLI services cannot be wrapped until SGI fixes their system so 49*2aef6930SMark Murray that it works the way it is supposed to. 50*2aef6930SMark Murray 51*2aef6930SMark MurrayI am not impressed by the quality of the IRIX system software. There 52*2aef6930SMark Murrayare many things that work on almost every other system except with IRIX. 53*2aef6930SMark Murray 54*2aef6930SMark Murray Wietse 55