Age | Commit message (Collapse) | Author |
|
to 1024 session per process (esp. with keep-alive). Now the fd limit is
the new maximum and relayd will make sure to not accept too many sessions.
The tcp backlog config maximum is now 512, adjust manpage accordingly.
OK benno@ deraadt@
|
|
Ensures that the olatch decrements uniformly, independent of system
wall clock jumps.
While here, roll the olatch computation in i8253_do_readback() into
a loop, and leverage the timespec macros in sys/time.h. Both make the
code a lot more readable.
ok mlarkin@
|
|
This implements RFC 3430, with the exception of processing multiple
incoming requests in parallel (Section 2.1). This required too much
code and is optional anyway.
Initial review by reyk@, very thorough reviews by jca@. Thanks!
OK jca@, gerhard@
|
|
set to HOST_DOWN.
Noticed and fixed by Rivo Nurges <Rivo DOT Nurges AT smit DOT ee>
ok and reminder florian@
|
|
Ok ccardenas@
|
|
"listen on * port 80".
While here accept up to 16 addresses from DNS or interface groups.
requested by & "lovely" deraadt@
OK kn@
|
|
So that events fire punctually, even if the system clock is changed.
ok jca@
|
|
addresses could be listed with prefix.
Initial diff from Ryan Kavanagh, tweaked by me.
Ok millert@ gilles@
|
|
Ryan Kavanagh reported on github that certain domains have misconfigured
SPF records. https://github.com/OpenSMTPD/OpenSMTPD/issues/844
Ok millert@ gilles@
|
|
OK jca, stsp
|
|
removal of the agreement url by now.
OK benno
|
|
AF_UNIX is both the historical _and_ standard name, so prefer and recommend
it in the headers, manpages, and kernel.
ok miller@ deraadt@ schwarze@
|
|
ok deraadt@
|
|
connection and eventually stop answering queries because of file
descriptor starvation.
Problem reported by, minimal testcase provided and testing by trondd
_AT_ kagu-tsuchi.com, thanks!
Testing Nick Holland and millert
OK deraadt
|
|
Problem reported by jj@ on bugs@, fix based on
https://gitlab.isc.org/isc-projects/bind9/commit/084ba95b083dc55fd10631ad43fa8fff48707648
(under ISC license) by Caspar Schutijser.
|
|
the vlan specs have the priority of 0 and 1 swapped on the wire,
which is how the kernel handles them. eg, if you use pf to set prio
1, it will end up being 0 on the wire. this makes 0 on the wire
come out as 1 in tcpdump so it is consistent with the rest of the
tooling.
ok henning@
|
|
on 4-byte AS path. Since this function is also called with 2-byte ASPATH
attributes. Instead inline the check using the as_size.
Problem found by Tom Beard (tom at exilien.uk), thanks for the report
OK job@
|
|
|
|
errors.
Pointed out by benno
|
|
testing tb, sthen
OK benno, sthen
|
|
loaded. This makes sure that the local prefix specied in the config is
always used.
OK ccardenas@
|
|
These are the direct sources from NLnet Labs upstream, minus these:
compat contrib libunbound/python pythonmod testcode testdata winrc
ok deraadt@ jakob@
|
|
ok sthen
|
|
From Raf Czlonka, ok sthen@
|
|
both cases. Avoids race in the exit path. ok sthen@ florian@
|
|
ok gilles@
|
|
been the default for a long time now) but users still assume this is the
way to update aliases within the MTA.
when smtpctl is invoked as newaliases, have it notify the daemon that it
has had the aliases table updated, no matter what backend it uses, as it
will at worst be a noop and at best do what user expected.
ok eric@
|
|
port is stored in the listener structure using the network order.
ok eric@
|
|
prevent possible loops on badly configured spf records.
ok gilles@
|
|
|
|
certificate for some time.
While they stopped doing so it seems best to follow redirects anyway.
More status codes pointed out by & OK sthen
OK benno
|
|
Also changes an error message in vmctl to reflect same.
|
|
|
|
ok sthen
knock out a dud Tn while here
|
|
in packet loss on busy and big networks and therefor to instability.
From camield@
|
|
in packet loss on busy and big networks and therefor to instability.
Also adjust the SO_SNDBUF to real 64k just for consistency.
From camiel@
|
|
from Andrew Marks
ok deraadt@
|
|
a recent cleanup we accidentally removed one of the ways breaking some .net
lib apparently, sorry.
this commit reintroduces the "pass password in the same line as username"
method
ok millert@
|
|
ok deraadt@ mestre@ gilles@
|
|
- Fix memory leak in zone file read of unknown rr formatted RRs.
- Fix memory leak when rehashing nsec3 after axfr or zonefile read,
in the selectively allocated precompiled nsec3 hashes.
|
|
|
|
|
|
ok jmc@
|
|
|
|
OK espie@
|
|
|
|
note that this is only part of raf's original diff: the rest is for
code. although i'm sure the diff itself is correct, i don;t judge
myself qualified to decide whether those changes are warranted.
still, i'd appreciate at least a reply to raf's diff from another
developer (yay! nay!)
|
|
ok krw@, tweaks by jmc@
|
|
Use rtc_reschedule_per in mc146818_start instead of re arming the
periodic interrupt without checking if it's enabled in REGB.
reported by Dave Voutila
ok phessler@
|
|
- have confirm go thru f() every time (adding
confirm_defaults_to_no/confirm_defaults_to_yes apis... verbose but effective)
- check DESCR in pkg_create, as suggested by sthen@
- don't do the whole safe dance when not needed (optimization)
- misnamed allowed -> forbidden (schwarze@)
This effectively prevents DESCR from using diacritics. Sorry!
Ingo makes a compelling argument that using UTF8 while installing a new
box is currently fairly dangerous, considering tty setups.
This affects about 10 ports at the moment, all of which were fixed before
this commit.
|