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