Commit | Line | Data |
---|---|---|
09169146 GW |
1 | # |
2 | # TODO,v 3.3 1993/11/09 23:20:16 kardel Exp | |
3 | # | |
4 | This file contains problems known to the authors that still need to be done. | |
5 | We would appreciate if you could spare some of your time to look through | |
6 | these topics and help us with some open questions. Most of the topics | |
7 | pertain to specific architectures where we have no direct access or not | |
8 | the time or expertise to currently track down the problem further. | |
9 | If you don't know what we are talking about in the topics don't bother | |
10 | with finding out - somebody else will probably solve that problem. | |
11 | ||
12 | Before you try to send a solution to mills@udel.edu please check whether | |
13 | this problem still exists in the distribution on louie.udel.edu. | |
14 | ||
15 | Thank you for your help ! | |
16 | Dave Mills | |
17 | Frank Kardel | |
18 | Piete Brooks | |
19 | ||
20 | Open issues: | |
21 | ||
22 | HPUX: | |
23 | - Time warp | |
24 | During the last few month disturbing reports about xntp setting | |
25 | preposterous times during periods of high load have been reported | |
26 | on HPUX 8 and 9. The theory is that the adjtimed message queue | |
27 | gets deleted. Symptoms are that xntp() complains about interrupted | |
28 | system calls in adjtime()-emulation and the time is set to some | |
29 | invalid date. Also the adjtimed seems to have problems. We could | |
30 | need some help here by an experienced HPUX guru. | |
31 | Files affected: adjtime/* | |
32 | ||
33 | Apollo: | |
34 | - terminal affiliation | |
35 | Check whether thing are still correct in respect to breaking | |
36 | terminal affiliation - horrible stories are told in the code. | |
37 | File affected: xntpd/ntpd.c |