similar to: Patch: More of kqueue() support.

Hi, I would like to submit the attached patch. It implements IO loop using FreeBSD's kqueue/kevent syscalls. It is based on snapshot of CVS HEAD as of 2005-12-12. I could only give it limited testing on FreeBSD 5.4 but it works fine so far. Vaclav Haisman -------------- next part -------------- diff -rN -u old-dovecot-cvs/autogen.sh new-dovecot-cvs/autogen.sh ---

The attached patch should fix the problem with dying imap on "Unrecognized event". The problem is that when we register a handle for IO_ERROR only, we still can get readable/writable event without EV_EOF being set. This case was not handled. -- Vaclav Haisman -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: io-kq.diff URL:

Hello, attached patch adds experimental ioloop handler which uses BSD kqueue API. It compiles and survives my small test program, but not thoroughly tested because I have no suitable BSD machine to torture. Brave owners of (Free|Net|Open)BSD please test and report results :) Best regards. -- Andrey Panin | Linux and UNIX system administrator pazke at donpac.ru | PGP key: wwwkeys.pgp.net

Chris Cappuccio wrote: > So I just got one like this before a login: > > May 16 21:24:58 m1 dovecot: imap-login: Unrecognized event: kevent {.ident = 0, .filter = 0xffffffff, .flags = 0x0001, .fflags = 0x00000000, .data = 0x0000002b} > May 16 21:24:58 m1 dovecot: imap-login: mode: 0x004x > May 16 21:24:58 m1 dovecot: child 10716 (login) killed with signal 6 > > The OpenBSD

Hi, I have noticed from logs that (only handful of occurrences during whole week) sometimes there is some kind of unrecognized event on handle and the branch with i_panic() in ioloop-kqueue.c:io_loop_handler_run() catches it. This patch allows the i_panic() to print some more info about the nature of the unrecognized event so that I can find out what it is about. Since it is really rare event I

Hi! Can you try these two patches? Aki On 24.10.2016 08:48, Aki Tuomi wrote: > Ok so that timeval makes no sense. We'll look into it. > > Aki > >> On October 24, 2016 at 12:22 AM Larry Rosenman <larryrtx at gmail.com> wrote: >> >> >> doveadm(mrm): Debug: http-client: conn 127.0.0.1:9998 [1]: Got 200 response >> for request [Req38: PUT

Hi, dovecot tries to use OpenSSL's PRNG to generate random numbers if there is no /dev/urandom found. Unfortunately, it is flawed in its present form, since the PRNG is not seeded before RAND_bytes() is called in src/lib/randgen.c (on systems which have /dev/urandom, OpenSSL automatically seeds its PRNG from the urandom device). Here's a patch to address this issue: it tries to seed

doveadm(mrm): Debug: http-client: conn 127.0.0.1:9998 [1]: Got 200 response for request [Req38: PUT http://localhost:9998/tika/] (took 296 ms + 8 ms in queue) doveadm(mrm): Panic: kevent(): Invalid argument Program received signal SIGABRT, Aborted. 0x00000008014e6f7a in thr_kill () from /lib/libc.so.7 (gdb) fr 6 #6 0x00000008011a3e49 in io_loop_handler_run_internal (ioloop=0x801c214e0) at

A few weeks back I upgraded my mail server to -STABLE from a 4.2-STABLE incarnation that had been running for years. Part of recompiling everything on the box involved disassociating my use of the sendmail port I was using and to use the base sendmail in -STABLE. -STABLE builds with libmilter, so I simply recompiled one of my milter clients with the milter headers, objs, etc that were produced

Hey, I noticed that there was an ioloop "module" (if we can call it that) for select and poll and decided to add one for kqueue (aka kevent) BSDs high performance descriptor multiplexing API. I haven't done any of the configure glue stuff but the code is complete and works well. kqueue is available on all recent versions of FreeBSD, NetBSD, OpenBSD and Darwin (and therefore MacOS

Hi, We have a issue with a server pretty much just running dovecot, there's about 250gb of maildirs which has about 200 simultaneous connections to it running on freebsd 7. The machine has stayed up for about 8 days before it panics but sometimes panics after about 24 hours. We are running a PAE kernel (with ULE scheduler) as the machine has 2 (4 core processors) and 8gb of ram, dovecot is

Hi, I've seen some chatter on NFS boards about kqueue being more reliable than inotify when used in NFSv3 and NFSv2. The chatter is a bit old so I don't know if it is true anymore. Anyone have pro/con experience with dovecot on the inotify/kqueue question when using NFS storage? I realize that kqueue is probably a bit slower and causes some delay with IDLE. Also, it may not really

Hello, I'm a new user of Dovecot (since 1.0b8) and I'ld report about a possible bug. Situation: Dovecot 1.0beta9 with kqueue support FreeBSD 6.1-Release x86 machine (IBM x206, P4 3.2 Ghz) local filesystem UFS2 This is an internal, experimental, mail server running Postfix and Dovecot configured with Maildirs e IMAP-only access. Clients are quite few, now, just because it's an

I?ve hit this compile error on macOS 10.15 with the file src/lib/ioloop-notify-kqueue.c: > error: implicit declaration of function 'i_gettimeofday' is invalid in C99 The fix is to #include "time-util.h" > --- ./src/lib/ioloop-notify-kqueue.c 2020-09-18 19:56:38.000000000 -0400 > +++ ./src/lib/ioloop-notify-kqueue.c 2020-09-18 20:54:36.000000000 -0400 > @@ -11,6

Apparently Dovecot with kqueue ioloop enabled causes random crashes in FreeBSD 9.1. Anyone else noticed something like that? I'm wondering if Dovecot's kqueue code has a bug somewhere (I can't find it) or could it be that FreeBSD itself has a bug? Anyway it seems to be returning events for filters that have already been removed.

Hello Dovecot, Version: dovecot-1.0-test61 OS: FreeBSD 5.3-RELEASE i386 Client: Mozilla Thunderbird 1.0 (on WinXP, FreeBSD & Mac OS X) The maillog entries concerning me: - Jan 30 16:16:37 venus dovecot: imap-login: Login: user [10.6.8.3] Jan 30 16:16:37 venus dovecot: imap-login: couldn't remove filter with kqueue: Bad file descriptor Jan 30 16:16:37 venus dovecot: IMAP(user):

I rewrote much of the kqueue code since the old code didn't handle properly the case when both input and output I/O handlers were added to the same file descriptor with different callbacks. This is done commonly in Dovecot, so I'm wondering why more people didn't complain about problems with it.. :) I don't anyway have access to any BSDs so the code is completely untested. Please

Hi, The sample program given below freezes my 5.1R installation. This program is from a bug report - kern/54331. That bug report (http://lists.freebsd.org/pipermail/freebsd-bugs/2003-July/001608.html) is for the shutdown function call, whereas my machine basically just freezes up and reboots on kevent itself. I am also noticing this problem with another program i wrote (but too long to include

Hello, currently I'm working on new ioloop handler which uses epoll(4) API introduced in Linux kernel 2.6. In this API each fd added to fd set by epoll_ctl system call can be accompanied with user supplied data (integer or void pointer). epoll_wait syscall reports arrived events as an array of structures containing event mask and user data. Attached patch replaces fd and condition parameters

So, apparently the original kqueue code was working fine after all, although it was much more complex than it really needed to be. So, this time there's a really simple and pretty kqueue implementation: http://dovecot.org/nightly/dovecot-latest.tar.gz Tested that it seems to work with FreeBSD. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc