| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2013-04-23 | support src/libexec/identd's -e option in src/usr.sbin/identd, ok dlg@ | Stuart Henderson | |
| 2013-04-23 | handle large ino_t | Theo de Raadt | |
| 2013-04-23 | handle large ino_t | Theo de Raadt | |
| 2013-04-23 | sizeof(&foo) isnt as good as sizeof(foo). | David Gwynne | |
| found by sthen@ | |||
| 2013-04-23 | use stat to see if ~/.noident is there to avoid reimplementing a security | David Gwynne | |
| issue from 1988 when using open. pointed out by deraadt@ | |||
| 2013-04-23 | i copied too much from the tftpd manpage. identd doesnt take a directory | David Gwynne | |
| argument. | |||
| 2013-04-23 | fix usage to reflect the currently accepted getopt things. | David Gwynne | |
| 2013-04-23 | i forgot to add -n and -N to the synopsis | David Gwynne | |
| 2013-04-23 | remove support for specifying which port you want to run on. there's no | David Gwynne | |
| point. well argued by deraadt@ | |||
| 2013-04-23 | add support for returning uids instead of usernames via -n, like libexec | David Gwynne | |
| identd. | |||
| 2013-04-23 | add support for -N as per libexec/identd. this lets users put | David Gwynne | |
| .noident in their homedir to have this identd return HIDDEN-USER instead of their username. | |||
| 2013-04-22 | prepare for more time bits; ok millert@ deraadt@ | Otto Moerbeek | |
| 2013-04-22 | When sync'ing, look for lease by hwaddr and then ipaddr. Not hwaddr | Kenneth R Westerback | |
| twice. Simplify logic. Fixes loops while sync'ing for sthen@. Spotted by & fixed tested by sthen@. | |||
| 2013-04-22 | handle ENOBUFS on the SEQPACKET socketpair between the child and the parent | David Gwynne | |
| procs. if this happens it would mean the parent has a backlog of work cos of slow username resolutions. in that situation the child should tell the client theres an error, but not die. this factors out a bit of code to handle generating errors for the client and closing the socket. reviewed by jmatthew@ | |||
| 2013-04-21 | use arc4random_uniform(); ok djm | Theo de Raadt | |
| plus some other markings about signal races... | |||
| 2013-04-21 | Tweak #include order to work during the big time_t transition | Philip Guenther | |
| noted by matthieu@ | |||
| 2013-04-21 | handle exceedingly long uptimes | Theo de Raadt | |
| 2013-04-21 | IPv6 might still be around in 2038, replace magic timeval | Florian Obser | |
| {0x7fffffff, 0x7fffffff} with an explicit flag that timers should be stopped. input deraadt@, kettenis@ "Looks like a nice solution." deraadt@ OK benno@ | |||
| 2013-04-21 | delete (uncompiled) hacked usleep function. | Ted Unangst | |
| young minds should not be exposed to such code. | |||
| 2013-04-21 | Unbreak edquota by fixing the temporary file name template. Whilst here, | Joel Sing | |
| fold pathnames.h into edquota.c - we do not really need a header file for a single define. ok deraadt@ guenther@ millert@ | |||
| 2013-04-21 | deregister. no binary change. | Ted Unangst | |
| 2013-04-21 | as in dvmrpd: | Theo de Raadt | |
| the DVMRP protocol generation number can work beyond 2038 because it is unsigned. Cast it specifically, with a comment, to make this more clear. | |||
| 2013-04-21 | the DVMRP protocol generation number can work beyond 2038 because it | Theo de Raadt | |
| is unsigned. Cast it specifically, with a comment, to make this more clear. | |||
| 2013-04-21 | revert 1.45. it depended on a kernel change we will not be making, and | Ted Unangst | |
| the old code was cleaner | |||
| 2013-04-21 | convert select to poll, since the select code had descriptor limits | Theo de Raadt | |
| ok tedu | |||
| 2013-04-21 | cleaner and easier to use signed long long. ok deraadt | Ted Unangst | |
| 2013-04-20 | Don't use random() because it's not safe. Use arc4random() instead of | YASUOKA Masahiko | |
| it. Suggested by deraadt. | |||
| 2013-04-20 | Remove extra info from usage message; that's what the manual is for. | Todd C. Miller | |
| OK jmc@ | |||
| 2013-04-20 | casts of (time_t *) lead to bugs | Theo de Raadt | |
| 2013-04-20 | Crank C_MAXFILE static limit of files to serve; note that this is actually | Miod Vallat | |
| a limit of the number of files in /usr/mdec/rbootd/, not a limit of files listed in /etc/rbootd.conf. The existing limit of 10 is too small for some networks. | |||
| 2013-04-20 | Use offsetof() instead of rolling our own; gets rid of truncation warnings | Miod Vallat | |
| when building on 64-bit hosts. "looks good" deraadt@ | |||
| 2013-04-20 | replace select with poll | Theo de Raadt | |
| ok miod | |||
| 2013-04-20 | remove one of the last stupid (unsigned) casts in the tree | Theo de Raadt | |
| 2013-04-20 | spacing | Theo de Raadt | |
| 2013-04-20 | remove irrelevant comment | Theo de Raadt | |
| 2013-04-20 | SSL_CTX_set_timeout only handles long for the delta timeval, so constrain | Theo de Raadt | |
| it with MIN and LONG_MAX. It is only an interval, so it is fine. suggestion by djm, ok benno | |||
| 2013-04-20 | print large time_t nicely; ok benno | Theo de Raadt | |
| 2013-04-20 | PPPoE server was broken since last configuration rework. Calling | YASUOKA Masahiko | |
| pppoed_init() was accidentally deleted. Revert it in npppd_init() and rearrange l2tpd_init() and pptpd_init() to the same place. ok todd | |||
| 2013-04-20 | Remove extraneous "\n" erroneously introduced by move to strftime(). | Kenneth R Westerback | |
| Spotted by sthen@. | |||
| 2013-04-20 | Add -o option to getopt string and usage. | Todd C. Miller | |
| 2013-04-19 | Log (at LOG_INFO) which interfaces dhcpd listens to and (as suggested by | Stuart Henderson | |
| krw) their address. As dhcpd uses BPF, it's not possible to find this information out from fstat/netstat. ok krw@ | |||
| 2013-04-19 | Make -n the default and add -o flag for the old RFC 868 time protocol | Todd C. Miller | |
| (which uses a 32-bit value for its wire protocol). OK deraadt@ ian@ | |||
| 2013-04-19 | for datagrams, assume other protocols should fail | Theo de Raadt | |
| ok millert | |||
| 2013-04-19 | Constrain the 2038 check to only when sizeof(time_t) == sizeof(int). | Theo de Raadt | |
| Adding more powerful checking is too difficult. We'll fix this problem by moving to 64-bit time_t.. and once we do, we don't want this code to falsely trigger. ok millert | |||
| 2013-04-19 | comment incorrect; rdate has -n support now too | Theo de Raadt | |
| 2013-04-19 | sw_vp is now in struct nfs_diskless. | Florian Obser | |
| input/ok deraadt@, ok miod@ | |||
| 2013-04-19 | print tv_sec properly | Theo de Raadt | |
| 2013-04-18 | struct timezone is so 1980 | Theo de Raadt | |
| 2013-04-17 | Replace hand-rolled date printing/parsing code with strftime()/strptime(). | Kenneth R Westerback | |
| Use timegm() because all the dates are in UTC and the 'standard' routines can't handle the truth. Remove some 'time_t is 32bits' assumptions. Print 'UTC' at the end of dates in the leases file, rather than mentioning that all dates are UTC in a comment at the top of the file. Feedback and suggestions from guenther@, kettenis@, otto@, tedu@, deraadt@. | |||
| 2013-04-17 | Adapt cron and at for future large time_t and tv_sec types. These were some | Theo de Raadt | |
| of the harder programs to adapt, so let me know if problems happen. tested by guenther, gilles, chl, others | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |