Re: python3 module location (py3-junos-eznc broken)
Stuart Henderson [s...@spacehopper.org] wrote: > > Updating this will need new ports for two modules (transitions and > yamlordereddictloader; tgz attached), and the update of ncclient > that I've just committed. > This works, thank you Stuart.
Re: python3 module location (py3-junos-eznc broken)
Stuart Henderson [s...@spacehopper.org] wrote: > On 2024/01/22 02:00, Daniel Dickman wrote: > > > > > > AttributeError: module 'collections' has no attribute 'MutableMapping' > > > > > > I suspect the module may require python3.9 or earlier? > > > > Release 2.7.0 claims support for python 3.10 so I suggest updating the port > > to the latest release. > > That's probably the best way to go. > There is no distfile source in the Makefile. I assume this means it was captured from github at some point to ftp.openbsd.org and now the ports Makefile should be updated to the current github ports conventions?
python3 module location (py3-junos-eznc broken)
With python 3.10, py3-junos-eznc is broken: class _FactCache(collections.MutableMapping): AttributeError: module 'collections' has no attribute 'MutableMapping' I suspect the module may require python3.9 or earlier? How the heck do you make a port python3 module install for 3.9 instead of 3.10?
Re: librdkafka, libfds, ipfixcol2
Stuart Henderson [s...@spacehopper.org] wrote: > > These ports are to support IPFIXcol2 which is the first set of ports I've > > done in a long time. CMake and all those extra layers of abstraction are a > > bit of a pain in the ass. Can somoene please take a look and let me know how > > close I am getting to the mark here? Is this stuff close to an acceptable > > condition? > > The cmake bits you patched in libfds are because upstream is doing > nonstandard things, cmake ports with libraries aren't usually that > annoying. > Yeah I assumed that, after looking through other ports that use cmake and generate proper shared libs (with the ports version numbers, and without digging through all of these CMakeThings!) > Tidied up a bit here. All the static_assert -> _Static_assert don't > seem needed (and break things on GCC archs) so I've removed them, > and used -D_BSD_SOURCE instead of patching in many files. Couple > of other tweaks (manpage/etc locations, few others). Build output > looks alright here, I haven't tested runtime, don't have netflow/ipfix > setup, what little flow monitoring I do is usually with sflow. > Do they still work for you? > Weird, I couldn't get clang to compile it with static_assert, only converting to _Static_assert worked for me. Thank you Stuart. Checking it out now... Chris
librdkafka, libfds, ipfixcol2
These ports are to support IPFIXcol2 which is the first set of ports I've done in a long time. CMake and all those extra layers of abstraction are a bit of a pain in the ass. Can somoene please take a look and let me know how close I am getting to the mark here? Is this stuff close to an acceptable condition?
NEW: ipfixcol2 2.1.0
DESCR: IPFIXcol2 is a flexible, high-performance NetFlow v5/v9 and IPFIX flow data collector designed to be extensible by plugins. ipfixcol2.tgz Description: application/tar-gz
NEW: libfds 0.2.0
DESCR: The library provides components for parsing and processing IPFIX Messages. IPFIX Data structure parsers, IPFIX Data Record iterators (with Biflow support), IPFIX Template manager, IPFIX Data type coverters (getters, setters, to string), Manager of Information Elements (i.e. id, name, type, etc. of IPFIX fields), Simple XML parser with data type check (implemented as a wrapper over libxml2), Flow file storage (beta) libfds.tgz Description: application/tar-gz
NEW: librdkafka 1.7.0
OK to import? DESCR: librdkafka is a C library implementation of the Apache Kafka protocol, providing Producer, Consumer and Admin clients. It was designed with message delivery reliability and high performance in mind, current figures exceed 1 million msgs/second for the producer and 3 million msgs/second for the consumer. librdkafka.tgz Description: application/tar-gz
Re: Fwd: ports/shells/nsh-10.0.20201006p0-RC
I've got all of your changes (or variations thereof) in the github repository, except for the version number change since it isn't quite time to bump yet. Time to work on the known issues and expand compatibilty with the many features added in the last 3 or 4 years :) Tom Smyth [tom.sm...@wirelessconnect.eu] wrote: > Hello, > > Please find attached patch for nsh > > feedback welcome as my C is rusty, > > I hope this helps, > > details of proposed patch outlined below > > > Architectures Compiled on: > AMD64 > > Compile errors / warnings > 1) warning about an unused variable do not want to delete code just yet > 2) errors about '\0' Null pointers cleared > 3) fixed compatibility with OpenBSD Current and tried to maintain backward > compatibility with release (using #ifdef (March 11 commit in base > for IFXF_INET6_NOPRIVACY depreciated and replaced with IFXF_AUTOCONF6TEMP > 4) fixed warning about self assigned variable with (void)ch; > > Tested: > Manually tested on 6.9 beta current April 11 amd64 > > BUGS Resolved: > show version fixed as there was a sysctl incompatibility that > was throwing an error when "show version" command was run > > Known Issues: > sysctl commands / functions need to be updated in line with changes in base > new interfaces to be added > arptimeout command is not working > > Enhancements: > increased the verbosity of the version displayed when running NSH > so it is easier for the user to see the version and patch level. > > > My sincere thanks to Chris Cappuccio for authoring and opensourcing NSH.
Re: shells/nsh
Marc Espie [es...@nerim.net] wrote: > I've looked a bit at this code. > > I'm on the fence between marking it as BROKEN and give Chris a chance to > fix the code, or downright removing it. > > Specifically, it's currently broken thanks to recent changes to if.h, > but the code warns all over the place. There seems to be a misconception > that '\0' would be a NULL pointer (it's not, it's a character constant). > Well it was to mark the end of arguments in constructed argument lists, but maybe just NULL would be fine. > Looking a bit deeper, clang finds at least one actual bug and some > seriously confused code (strlcpy with size from the src instead of the dest, > which is kind-of okay in this case because it is the same size, but still). > Ugh > I don't know who actually uses nsh, but I wouldn't recommend it in its > present state... After all, you're running code as root, it should at least > keep the warnings down... and it's not like we moved to clang two days ago. > I'll fix this up
Re: Update devel/scons
Stuart Henderson [s...@spacehopper.org] wrote: > > Wouldn't they just do the same they've done with Java when their license > changed (i.e. make people install it themselves)? Or keep on providing the > old binaries themselves but not force the no-longer-supported command line > flags so it works for distribution packagers.. > > (Of course I would very much support them switching to something that > isn't mongodb...) They package the full UniFi controller software in their own products (Cloud Key, UDM) so it needs to be replaced
Re: Update devel/scons
Stuart Henderson [s...@spacehopper.org] wrote: > On 2020/02/06 21:01, Denis Fondras wrote: > > Update to 3.1.2. Will be useful to upgrade MongoDB. > > btw, replacing mongodb with a newer version is blocked until ubiquiti > get round to fixing unifi, though a newer version could be added separately.. I'm not sure that ubiquiti will ever switch to a newer mongo, more likey to something else considering mongodb's new license.
Re: firefox pledge "", syscall 289
Stuart Henderson [s...@spacehopper.org] wrote:
>
> Two simpler ways likely to bypass it, one is the diff below which is
> expected to slow things down for non-pledged programs, the other is to
> disable pledge in firefox which you can do with lines like these in
> prefs.js (replace if already present, add if not).
>
> user_pref("security.sandbox.pledge.content", "moo");
> user_pref("security.sandbox.pledge.main", "moo");
>
> (this is the method for current Firefox, but future updates are
> expected to move this config into a different file).
>
So on the box that has these issues, funny enough /etc/fbtab never gets
honored and so /dev/drm0 stayes owned by root. When I chown to me, it's
apparently not trying to use shm anymore because I stop seeing these
crashes. I think they come back, but it could just be that I reboot at
times and I'm not paying attention to these details.
I haven't yet understood exactly why login_fbtab() isn't happening.
It's called from xenodm and seems to work on some machines. So there's that.
There are two places where login_fbtab is called in xenodm and one of them
has this note: /* XXX not working because of pledge() */
Chris
firefox pledge "", syscall 289
I keep getting random crashes on firefox tabs related to DRI use of shm. The funny thing is, 'ipcs' never shows any allocated shm right before the crash, or at any time. The shmget is entirely superfluous, but it crashes my browser. If I restart firefox, this can go away for hours or a few days, and then it starts trying to do shmget again for a bit. It seems to be partiallyu related to the web page, and its use of the browser. This is both on 6.6 and current. Is anyone else seeing this with an inteldrm based console, or any console and firefox ?? #0 shmget () at -:3 #1 0x13e24027df7c in dri_sw_displaytarget_create () from /usr/X11R6/lib/modules/dri/swrast_dri.so #2 0x13e240434d51 in llvmpipe_resource_create () from /usr/X11R6/lib/modules/dri/swrast_dri.so #3 0x13e2400eaaad in drisw_allocate_textures () from /usr/X11R6/lib/modules/dri/swrast_dri.so #4 0x13e2400e7d9d in dri_st_framebuffer_validate () from /usr/X11R6/lib/modules/dri/swrast_dri.so #5 0x13e23fef93ad in st_framebuffer_validate () from /usr/X11R6/lib/modules/dri/swrast_dri.so #6 0x13e23fefa241 in st_api_make_current () from /usr/X11R6/lib/modules/dri/swrast_dri.so #7 0x13e2400e77f9 in dri_make_current () from /usr/X11R6/lib/modules/dri/swrast_dri.so #8 0x13e2400e627a in driBindContext () from /usr/X11R6/lib/modules/dri/swrast_dri.so #9 0x13e1ec0497ba in drisw_bind_context () from /usr/X11R6/lib/libGL.so.17.1 #10 0x13e1ec04fb8c in glXMakeCurrentReadSGI () from /usr/X11R6/lib/libGL.so.17.1 #11 0x13e1f780fcda in std::__1::vector >::__append () from /usr/local/lib/firefox/libxul.so.84.0 ... libxul trace + broken trace
Re: shells/nsh: cope with mobileip removal
Christian Weisgerber [na...@mips.inka.de] wrote: > The removal of mobileip(4) has broken shells/nsh. > > The patches below simply remove mobileip support from nsh and update > the MANUAL, including a paragraph that became obsolete when mobileip(4) > was added. I don't use nsh at all, but with this it compiles again. > > ok? Perfect...It's also about time for me to update the damn thing Chris
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ch...@cvs.openbsd.org 2018/04/02 09:33:32 Modified files: security/suricata: Makefile Log message: The configure script tries to use -march=native automatically. This is not appropriate for package building. ok sthen@
suricata portable binary
suricata documents this option as "Do not optimize the binary for the hardware
it is built on."
I think it is appropriate for the amd64 build since not every machine is the
ports build machine.
Index: Makefile
===
RCS file: /cvs/ports/security/suricata/Makefile,v
retrieving revision 1.4
diff -u -p -u -r1.4 Makefile
--- Makefile17 Mar 2018 13:03:14 - 1.4
+++ Makefile1 Apr 2018 05:42:47 -
@@ -38,6 +38,8 @@ CONFIGURE_STYLE = gnu
CONFIGURE_ENV = ac_cv_path_HAVE_PDFLATEX= \
ac_cv_path_HAVE_GIT_CMD=
+CONFIGURE_ARGS = --disable-gccmarch-native
+
post-install:
${INSTALL_DATA_DIR} ${PREFIX}/share/examples/suricata/rules
${INSTALL_DATA} ${WRKSRC}/*.config ${PREFIX}/share/examples/suricata
Re: Chromium and Iridium no longer working (current 2016-07-10T00:21:52Z)
Martijn Rijkeboer [mart...@bunix.org] wrote: > > I can't tell from your message whether this is the following problem: > > There is a diff being discussed that makes pledge "recvfd" stricter. It > > is plausible that Theo put this diff into the most recent snapshot. One > > result is that this diff will make chromium and iridium crash because > > they do not make "recvfd" promises. Do you see messages like > > > > chrome(99124): syscall 27 "recvfd" > > > > in your dmesg? > > > > If that's the case, chrome should still work with the old kernel or with > > a kernel compiled from source. > > Thanks for your response, this is indeed the problem. My dmesg is full of > "recvfd" en "stdio" pledge messages. > This diff appears to be in -current now. You'll need to re build the chromium/iridium port.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ch...@cvs.openbsd.org 2016/05/31 12:30:23 Modified files: net/flowd : Makefile distinfo net/flowd/pkg : PLIST Log message: Move up to flowd github version. IPFIX support was hiding out there in the cloud! ok sthen@ djm@
Re: UPDATE: misc/redshift
Ryan Freeman [r...@slipgate.org] wrote: > > Context may help. By 'this', I mean the geoclue.conf hint: > > [redshift] > allowed=true > system=false > users= > absolutely
Re: mail/sympa, sysutils/backuppc: remove forced mod_fastcgi dep
Stuart Henderson [s...@spacehopper.org] wrote:
> fastcgi support is built-in to pretty much every modern
> web-server, no need for an apache1 dep to make it "easier".
>
It may be worth removing spawn-fcgi as well, since slowcgi is in base.
> ok?
>
> Index: mail/sympa/Makefile
> ===
> RCS file: /cvs/ports/mail/sympa/Makefile,v
> retrieving revision 1.33
> diff -u -p -r1.33 Makefile
> --- mail/sympa/Makefile 24 Nov 2015 09:11:05 - 1.33
> +++ mail/sympa/Makefile 28 Apr 2016 15:48:41 -
> @@ -4,7 +4,7 @@ COMMENT = mailing list manager with web
>
> VERSION =6.1.24
> DISTNAME = sympa-${VERSION}
> -REVISION = 0
> +REVISION = 1
>
> CATEGORIES = mail www
>
> @@ -61,7 +61,6 @@ RUN_DEPENDS = archivers/p5-Archive-Zip \
> textproc/p5-Template \
> textproc/p5-XML-LibXML \
> www/spawn-fcgi \
> - www/mod_fastcgi \
> www/mhonarc \
> www/p5-libwww \
> www/p5-HTML-Format \
> Index: sysutils/backuppc/Makefile
> ===
> RCS file: /cvs/ports/sysutils/backuppc/Makefile,v
> retrieving revision 1.12
> diff -u -p -r1.12 Makefile
> --- sysutils/backuppc/Makefile20 Sep 2015 07:23:36 - 1.12
> +++ sysutils/backuppc/Makefile28 Apr 2016 15:48:41 -
> @@ -3,6 +3,7 @@
> COMMENT =remote backup software system
>
> DISTNAME = BackupPC-3.3.1
> +REVISION = 0
> PKGNAME =${DISTNAME:L}
>
> CATEGORIES = sysutils
> @@ -26,8 +27,7 @@ RUN_DEPENDS = textproc/p5-XML-RSS \
> devel/p5-File-RsyncP \
> archivers/p5-Archive-Zip \
> www/p5-libwww \
> - www/spawn-fcgi \
> - www/mod_fastcgi
> + www/spawn-fcgi
>
> do-install:
> cd ${WRKSRC} && perl ./configure.pl \
Re: rsyslog [Re: CVS: cvs.openbsd.org: ports]
Stuart Henderson [s...@spacehopper.org] wrote: > On 2016/03/06 05:18, Antoine Jacoutot wrote: > > CVSROOT:/cvs > > Module name:ports > > Changes by: ajacou...@cvs.openbsd.org 2016/03/06 05:18:31 > > > > Modified files: > > sysutils/rsyslog: Makefile > > sysutils/rsyslog/patches: patch-runtime_nsd_gtls_c > > > > Log message: > > Fix build with GnuTLS >= 3.4 > > On a side note, this port could use an update... > > > > chris@ asked about this the other day too. here's a possible diff, > it builds but I haven't tried running it yet. > > also since I don't use this I have no idea if we actually want > liblogging-stdlog or if it's ok to disable for now. > > sample config is from chris. > Just an anecdotal report, this port has been working for me for logging remote syslog to PostgreSQL. Chris
Re: shell/nsh is broken
Yes that looks right. I think there will be other changes to adapt to current too.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ch...@cvs.openbsd.org 2015/05/05 11:19:57 Modified files: telephony/spandsp: Makefile distinfo telephony/spandsp/patches: patch-configure telephony/spandsp/pkg: PLIST Log message: Update to spandsp 0.0.6, enable sse2 on amd64 ok brad@ sthen@
Re: php-version for roundcubemail
Stuart Henderson [st...@openbsd.org] wrote: Problem is with PHP extension modules outside of the main php release, as these are compiled for the default PHP version in port (5.4 at present). So e.g. you won't be able to use things like pecl-memcache, xcache, etc from packages with php-5.5 until the default version is flipped over. (If you want to try it yourself building things from ports you can change MODPHP_VERSION in php.port.mk and rebuild everything depending on it). Why not just build them for each in-tree PHP version?
gnome on 5.6-current
I setup a box w/gnome for someone using the latest amd64 snapshot + gnome snapshot package. The box has radeondrm with the fw_update installed. I followed the directions in the pkg-readmes to start dbus-daemon, avahi, gdm. The stuff all starts, and gdm pops up the gnome screen with the clock at the top and some buttons for a calendar (and some slider options that don't do much of anything). But, that's all that pops up. There's no login or options to do anything else that I can figure out! Am I missing something obvious? From reading other people's experiences, it sounds like gnome should just work at this point. fvwm works fine, of course.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ch...@cvs.openbsd.org 2014/06/19 16:45:56 Modified files: net/iodine : Makefile distinfo net/iodine/patches: patch-src_Makefile patch-src_iodine_c patch-src_iodined_c Added files: net/iodine/patches: patch-src_client_c Log message: Update iodine DNS tunnel to 0.7.0 ok sthen@
Update for net/iodine
Like this? Index: Makefile === RCS file: /cvs/ports/net/iodine/Makefile,v retrieving revision 1.13 diff -u -p -u -r1.13 Makefile --- Makefile11 Oct 2013 23:50:18 - 1.13 +++ Makefile19 Jun 2014 19:43:02 - @@ -2,8 +2,7 @@ COMMENT= tunnel IPv4 data through DNS -DISTNAME= iodine-0.5.2 -REVISION= 0 +DISTNAME= iodine-0.7.0 CATEGORIES=net HOMEPAGE= http://code.kryo.se/iodine/ Index: distinfo === RCS file: /cvs/ports/net/iodine/distinfo,v retrieving revision 1.3 diff -u -p -u -r1.3 distinfo --- distinfo2 Jun 2009 06:27:05 - 1.3 +++ distinfo19 Jun 2014 19:43:02 - @@ -1,5 +1,2 @@ -MD5 (iodine-0.5.2.tar.gz) = aVI0PMRhSFf4PbuBJHhx5w== -RMD160 (iodine-0.5.2.tar.gz) = LA1VIzwqYVeRGluGi8U40nwesIU= -SHA1 (iodine-0.5.2.tar.gz) = KNDQDlE+IKp/38xhhy3uDsAy8RI= -SHA256 (iodine-0.5.2.tar.gz) = y8FGZdMlY0MY69b3kvjcIikQG9WlUjJCBXBThhL2peQ= -SIZE (iodine-0.5.2.tar.gz) = 50788 +SHA256 (iodine-0.7.0.tar.gz) = rStArPFCExbsFYANzeD1h6sx19b4kfqLmWfE3tk8AT4= +SIZE (iodine-0.7.0.tar.gz) = 96181 Index: patches/patch-src_Makefile === RCS file: /cvs/ports/net/iodine/patches/patch-src_Makefile,v retrieving revision 1.2 diff -u -p -u -r1.2 patch-src_Makefile --- patches/patch-src_Makefile 30 Mar 2009 09:17:45 - 1.2 +++ patches/patch-src_Makefile 19 Jun 2014 19:43:02 - @@ -1,23 +1,16 @@ -$OpenBSD: patch-src_Makefile,v 1.2 2009/03/30 09:17:45 sthen Exp $ src/Makefile.orig Sat Mar 21 13:07:49 2009 -+++ src/Makefile Mon Mar 30 10:09:57 2009 -@@ -1,4 +1,4 @@ --CC = gcc -+CC ?= gcc - COMMONOBJS = tun.o dns.o read.o encoding.o login.o base32.o base64.o md5.o common.o - CLIENTOBJS = iodine.o - CLIENT = ../bin/iodine -@@ -10,7 +10,8 @@ ARCH = `uname -m` +--- src/Makefile.orig Thu Jun 19 12:05:05 2014 src/Makefile Thu Jun 19 12:05:38 2014 +@@ -9,7 +9,8 @@ LIBPATH = -L. - LDFLAGS = -lz `sh osflags $(TARGETOS) link` $(LIBPATH) --CFLAGS = -c -g -Wall -D$(OS) -pedantic `sh osflags $(TARGETOS) cflags` + LDFLAGS += -lz `sh osflags $(TARGETOS) link` $(LIBPATH) +-CFLAGS += -std=c99 -c -g -Wall -D$(OS) -pedantic `sh osflags $(TARGETOS) cflags` +CFLAGS ?= -g -+CFLAGS += -c -Wall -D$(OS) -pedantic `sh osflags $(TARGETOS) cflags` ++CFLAGS += -std=c99 -c -Wall -D$(OS) -pedantic `sh osflags $(TARGETOS) cflags` all: stateos $(CLIENT) $(SERVER) -@@ -18,18 +19,15 @@ stateos: +@@ -17,18 +18,15 @@ @echo OS is $(OS), arch is $(ARCH) $(CLIENT): $(COMMONOBJS) $(CLIENTOBJS) @@ -32,10 +25,10 @@ $OpenBSD: patch-src_Makefile,v 1.2 2009/ - @$(CC) $(COMMONOBJS) $(SERVEROBJS) -o $(SERVER) $(LDFLAGS) + $(CC) $(COMMONOBJS) $(SERVEROBJS) -o $(SERVER) $(LDFLAGS) - .c.o: + .c.o: - @echo CC $ - @$(CC) $(CFLAGS) $ -o $@ + $(CC) $(CFLAGS) $ -o $@ - clean: - @echo Cleaning src/ + base64u.o client.o iodined.o: base64u.h + base64u.c: base64.c Index: patches/patch-src_client_c === RCS file: patches/patch-src_client_c diff -N patches/patch-src_client_c --- /dev/null 1 Jan 1970 00:00:00 - +++ patches/patch-src_client_c 19 Jun 2014 19:43:02 - @@ -0,0 +1,16 @@ +--- src/client.c.orig Thu Jun 19 12:23:49 2014 src/client.c Thu Jun 19 12:23:21 2014 +@@ -121,11 +121,11 @@ + b64u = get_base64u_encoder(); + b128 = get_base128_encoder(); + dataenc = get_base32_encoder(); +- rand_seed = ((unsigned int) rand()) 0x; ++ rand_seed = ((unsigned int) arc4random()) 0x; + send_ping_soon = 1; /* send ping immediately after startup */ + conn = CONN_DNS_NULL; + +- chunkid = ((unsigned int) rand()) 0x; ++ chunkid = ((unsigned int) arc4random()) 0x; + chunkid_prev = 0; + chunkid_prev2 = 0; + Index: patches/patch-src_iodine_c === RCS file: /cvs/ports/net/iodine/patches/patch-src_iodine_c,v retrieving revision 1.2 diff -u -p -u -r1.2 patch-src_iodine_c --- patches/patch-src_iodine_c 30 Mar 2009 09:17:45 - 1.2 +++ patches/patch-src_iodine_c 19 Jun 2014 19:43:02 - @@ -1,18 +1,24 @@ -$OpenBSD: patch-src_iodine_c,v 1.2 2009/03/30 09:17:45 sthen Exp $ - -Drop privileges and chroot by default. - src/iodine.c.orig Sat Mar 21 14:07:49 2009 -+++ src/iodine.c Mon Mar 30 10:25:08 2009 -@@ -1009,9 +1009,9 @@ main(int argc, char **argv) - int autodetect_frag_size; - - memset(password, 0, 33); +--- src/iodine.c.orig Mon Jun 16 13:28:43 2014 src/iodine.c Thu Jun 19 12:22:17 2014 +@@ -155,11 +155,10 @@ + #ifndef WINDOWS32 + pw = NULL; + #endif - username = NULL; + username = _iodine; +
Re: Fix: shells/nsh
Sorry I didn't get back to you on this earlier. The time_t stuff is already updated in github. But it all looks fine to me. Chris
Freeswitch
Does Freeswitch compile and run on OpenBSD? I thought it was already done, but just noticed it isn't in the official ports tree yet.
Re: Freeswitch
James Turner [ja...@calminferno.net] wrote: On Fri, Nov 22, 2013 at 03:50:25PM -0800, Chris Cappuccio wrote: Does Freeswitch compile and run on OpenBSD? I thought it was already done, but just noticed it isn't in the official ports tree yet. There seems to be a port in openbsd-wip [0]. [0] https://github.com/jasperla/openbsd-wip/tree/master/telephony/freeswitch I noticed that and also this: https://github.com/tbullock/freeswitch-openbsd http://tbullock.comlore.com/2013/03/freeswitch-on-openbsd.html Seems to me that the SQLite problem is resolved by simply using the freeswitch version of it. Chris
Re: NEW: sysutils/binwalk
Vadim Zhukov [persg...@gmail.com] wrote: It tries to use pymatplotlib, which we don't have, for entropy analisis. Probably, this should be mentioned in DESCR, to clear the doubt if this dependency was missed by accident. We don't have py-matplotlib? Except, we do. /usr/ports/graphics/py-matplotlib/
Re: Chromium core dumps
Stuart Henderson [st...@openbsd.org] wrote: On 2013/10/25 22:46, Fred wrote: Hi ports@ I have had an issue with Chromium for the last couple of week where it will segmentation fault if I click on the tools button and select settings. Apart from this Chromium seems to run fine... It's a known problem (tools/about and tools/settings broken on amd64) but I haven't heard any ideas of why it's happening yet. Likewise no ideas about the breakage with pretty much any complex web pages on i386 (which has been like this since around April, from looking at ports@ archives). I've been running it on i386 and other than history immediately crashing it, complex web pages work fine. I was using firefox but chrome is so much faster at this point, it's not even funny. Firefox is also constantly crashing on my relatively low memory i386 laptop.
Re: Preview of 5.4?
snapshots ? Jones, Bill [bill.jo...@lpsvcs.com] wrote: Is there going to be a preview of 5.4? http://ftp.openbsd.org/pub/OpenBSD/5.4/ ? Thx/Bill The information contained in this message is proprietary and/or confidential. If you are not the intended recipient, please: (i) delete the message and all copies; (ii) do not disclose, distribute or use the message in any manner; and (iii) notify the sender immediately. In addition, please be aware that any message addressed to our domain is subject to archiving and review by persons other than the intended recipient. Thank you. -- Man cannot live without a permanent trust in something indestructible within himself, though both that indestructible something and his own trust in it may remain permanently concealed from him - Franz Kafka
Re: segfaults firefox20 i386
Martijn van Duren [martijn...@gmail.com] wrote: Try to move i386 from MODCLANG_ARCHS to MODGCC4_ARCHS, run make clean and then rebuild. Thus helped me in a similar issue. After an hour of online gaming, I think I can safely say that building it with gcc4.6 fixes this issue for me. Unfortunately I don't seem to be able to build firefox-debug with clang, because of Memory Exhausted issues, so I can't trace the problem. If more people experience this problem on i386 it might be wise to switch www/mozilla-firefox back to gcc for this architecture. While we're talking about web browsers... Ever since chrome was switched to clang from gcc, it has been significantly slower on my t61 laptop. Not any noticeable difference on my core i3 desktop. Both are loaded with 8GB RAM, amd64.
Re: ruby on openbsd
Chris Cappuccio [ch...@nmedia.net] wrote: Ted Unangst [t...@tedunangst.com] wrote: On Fri, May 03, 2013 at 15:46, Jeremy Evans wrote: Basically, the native code in the libv8 gem doesn't compile correctly on OpenBSD. It probably requires patches (lang/libv8 requires patches to compile). There are two ways to go about fixing it: 1) Get libv8 to work. You could try installing an updated gcc port and see if its g++ will work. If not, you could try building an OpenBSD port for the ruby libv8 gem, and either getting it to use the system libv8, or porting over the lang/libv8 patches so that it builds. From the libv8 gem readme, it appears you can install it with the system v8 as so: gem install libv8 -- --with-system-v8 And then start all over and hope you can pass go. That is totally fucking sweet. So now that Ruby is up and running... I ocassionally get ruby 2.0 to crash on thin20 start. But not always. And there appears to be some looping behavior. Is this showing anything obvious to anyone? # gdb /usr/local/bin/ruby20 ruby20.core GNU gdb 6.3 Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type show copying to see the conditions. There is absolutely no warranty for GDB. Type show warranty for details. This GDB was configured as amd64-unknown-openbsd5.3...(no debugging symbols found) Core was generated by `ruby20'. Program terminated with signal 6, Aborted. Reading symbols from /usr/lib/libpthread.so.17.1...done. Loaded symbols for /usr/lib/libpthread.so.17.1 Loaded symbols for /usr/local/bin/ruby20 Reading symbols from /usr/local/lib/libruby20.so.0.0...done. Loaded symbols for /usr/local/lib/libruby20.so.0.0 Reading symbols from /usr/lib/libm.so.8.0...done. Loaded symbols for /usr/lib/libm.so.8.0 Symbols already loaded for /usr/lib/libpthread.so.17.1 Reading symbols from /usr/lib/libc.so.68.1...done. Loaded symbols for /usr/lib/libc.so.68.1 Reading symbols from /usr/libexec/ld.so...done. Loaded symbols for /usr/libexec/ld.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/enc/encdb.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/enc/encdb.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/enc/trans/transdb.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/enc/trans/transdb.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/pathname.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/pathname.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/etc.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/etc.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/digest/sha1.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/digest/sha1.so Reading symbols from /usr/lib/libssl.so.19.0...done. Loaded symbols for /usr/lib/libssl.so.19.0 Reading symbols from /usr/lib/libcrypto.so.22.0...done. Loaded symbols for /usr/lib/libcrypto.so.22.0 Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/digest.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/digest.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/socket.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/socket.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/fcntl.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/fcntl.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/openssl.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/openssl.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/stringio.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/stringio.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/strscan.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/strscan.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/date_core.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/date_core.so Reading symbols from /usr/local/lib/ruby/gems/2.0/gems/eventmachine-1.0.3/lib/rubyeventmachine.so...done. Loaded symbols for /usr/local/lib/ruby/gems/2.0/gems/eventmachine-1.0.3/lib/rubyeventmachine.so Reading symbols from /usr/lib/libstdc++.so.55.0...done. Loaded symbols for /usr/lib/libstdc++.so.55.0 Reading symbols from /usr/local/lib/ruby/gems/2.0/gems/thin-1.5.1/lib/thin_parser.so...done. Loaded symbols for /usr/local/lib/ruby/gems/2.0/gems/thin-1.5.1/lib/thin_parser.so Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/psych.so...done. Loaded symbols for /usr/local/lib/ruby/2.0/x86_64-openbsd/psych.so Reading symbols from /usr/local/lib/libyaml.so.0.0...done. Loaded symbols for /usr/local/lib/libyaml.so.0.0 Reading symbols from /usr/local/lib/ruby/2.0/x86_64-openbsd/bigdecimal.so...done
ruby on openbsd
hi trying to get some rails stuff working but the ruby-gem port keeps trying to install libv8 when it is already installed via ports. clue stick? # thin20 start Using rack adapter Could not find libv8-3.11.8.13 in any of the sources Run `bundle install` to install missing gems.
Re: ruby on openbsd
Jeremy Evans [jeremyeva...@gmail.com] wrote: On Fri, May 3, 2013 at 10:13 AM, Chris Cappuccio ch...@nmedia.net wrote: hi trying to get some rails stuff working but the ruby-gem port keeps trying to install libv8 when it is already installed via ports. clue stick? # thin20 start Using rack adapter Could not find libv8-3.11.8.13 in any of the sources Run `bundle install` to install missing gems. It appears to be trying to install the ruby libv8 gem, not the libv8 shared library (which is probably what you mean by installed via ports). What happens when you run `bundle install`? What are the contents of Gemfile and Gemfile.lock? Well I installed libv8 from ports and gem from ports. But then it tries to install libv8 again. Not knowing anything about ruby/rails/whatever this stuff is, I was hoping someone could tell me how to make it work on OpenBSD ports ruby/gem :) Gemfile: # cat Gemfile source 'https://rubygems.org' gem 'rails', '3.2.12' # Bundle edge Rails instead: # gem 'rails', :git = 'git://github.com/rails/rails.git' #gem 'sqlite3' gem 'pg' gem 'mysql2', :git = 'git://github.com/brianmario/mysql2.git' gem 'devise' gem 'formtastic' gem 'geocoder' gem 'therubyracer' gem 'will_paginate', '~ 3.0' gem 'ri_cal' gem 'nested_form' gem 'libv8' gem 'carrierwave' gem 'google-api-client' gem 'omniauth-google-oauth2' gem 'ancestry' gem 'whenever', :require = false gem 'activerecord-postgres-hstore' gem 'thin' gem haml-rails gem 'acts_as_list' gem 'formtastic-bootstrap' gem 'vestal_versions', :git = 'git://github.com/laserlemon/vestal_versions' gem 'public_activity' gem 'twilio-ruby' gem 'googlestaticmap' gem 'whenever' # # Gems used only for assets and not required # in production environments by default. group :assets do gem 'sass-rails', '~ 3.2.3' gem 'coffee-rails', '~ 3.2.1' gem 'bootstrap-sass' # See https://github.com/sstephenson/execjs#readme for more supported runtimes # gem 'therubyracer', :platforms = :ruby gem 'uglifier', '= 1.0.3' gem 'jquery-ui-rails' gem highcharts-rails, ~ 3.0.0 end gem 'jquery-timepicker-rails' gem 'jquery-rails' # To use ActiveModel has_secure_password # gem 'bcrypt-ruby', '~ 3.0.0' # To use Jbuilder templates for JSON # gem 'jbuilder' # Use unicorn as the app server # gem 'unicorn' # Deploy with Capistrano # gem 'capistrano' # To use debugger # gem 'debugger' Gemfile.lock: GIT remote: git://github.com/brianmario/mysql2.git revision: 72d0d22eea7b230253afc5e1f18bd743ccdc713d specs: mysql2 (0.3.12b6) GIT remote: git://github.com/laserlemon/vestal_versions revision: c7aa512666b22e1a47b4fb36bb1679cb847a532e specs: vestal_versions (1.2.3) activerecord (~ 3.0) activesupport (~ 3.0) GEM remote: https://rubygems.org/ specs: actionmailer (3.2.12) actionpack (= 3.2.12) mail (~ 2.4.4) actionpack (3.2.12) activemodel (= 3.2.12) activesupport (= 3.2.12) builder (~ 3.0.0) erubis (~ 2.7.0) journey (~ 1.0.4) rack (~ 1.4.5) rack-cache (~ 1.2) rack-test (~ 0.6.1) sprockets (~ 2.2.1) activemodel (3.2.12) activesupport (= 3.2.12) builder (~ 3.0.0) activerecord (3.2.12) activemodel (= 3.2.12) activesupport (= 3.2.12) arel (~ 3.0.2) tzinfo (~ 0.3.29) activerecord-postgres-hstore (0.7.5) activerecord (= 3.1) pg-hstore (= 1.1.5) rake activeresource (3.2.12) activemodel (= 3.2.12) activesupport (= 3.2.12) activesupport (3.2.12) i18n (~ 0.6) multi_json (~ 1.0) acts_as_list (0.2.0) activerecord (= 3.0) addressable (2.3.3) ancestry (1.3.0) activerecord (= 2.3.14) arel (3.0.2) autoparse (0.3.2) addressable (= 2.3.1) extlib (= 0.9.15) multi_json (= 1.0.0) bcrypt-ruby (3.0.1) bootstrap-sass (2.3.0.1) sass (~ 3.2) builder (3.0.4) carrierwave (0.8.0) activemodel (= 3.2.0) activesupport (= 3.2.0) chronic (0.9.1) coffee-rails (3.2.2) coffee-script (= 2.2.0) railties (~ 3.2.0) coffee-script (2.2.0) coffee-script-source execjs coffee-script-source (1.6.1) daemons (1.1.9) devise (2.2.3) bcrypt-ruby (~ 3.0) orm_adapter (~ 0.1) railties (~ 3.1) warden (~ 1.2.1) erubis (2.7.0) eventmachine (1.0.3) execjs (1.4.0) multi_json (~ 1.0) extlib (0.9.16) faraday (0.8.6) multipart-post (~ 1.1) formtastic (2.2.1) actionpack (= 3.0) formtastic-bootstrap (2.0.0) formtastic (~ 2.2) formtastic-bootstrap geocoder (1.1.6) google-api-client (0.6.2) addressable (= 2.3.2) autoparse (= 0.3.2) extlib (= 0.9.15) faraday (~ 0.8.4) jwt (= 0.1.5) launchy (= 2.1.1) multi_json (= 1.0.0) signet (= 0.4.4) uuidtools (= 2.1.0) googlestaticmap (1.1.3) haml (4.0.0) tilt
Re: ruby on openbsd
Jeremy Evans [jeremyeva...@gmail.com] wrote: Like I mentioned earlier, there's libv8 the shared library, and libv8 the ruby gem. The port you installed is for the shared library. The rails app is complaining because it wants you to install the libv8 ruby gem. The Gemfile includes gem 'libv8', so it appears to be a direct dependency of the rails app you are working on. I think it uses node.js. Or somehow other uses libv8. I mentioned that I installed libv8 from ports so that you could provide me with the secret handshake for make it all work together! What I think you're saying is that bundle should just work ? You didn't mention what happens when you run `bundle install`. As the rails app you are working on uses bundler, running `bundle install` is basically the first thing you should do after cloning the repository. bundle tries to compile libv8 and fails because g++44, g++45, g++46 don't exist (eg++ of course) which appear to be the debian/ubuntu names for explicit versions of g++. When I run gem by hand, it tries to compile libv8 with g++ and runs into some error (that may be related to the fact that it isn't g++ 4.4 or higher.) Building native extensions. This could take a while... ERROR: Error installing libv8: ERROR: Failed to build gem native extension. /usr/local/bin/ruby20 extconf.rb creating Makefile Compiling v8 for x64 Using python 2.7.4 Unable to find a compiler officially supported by v8. It is recommended to use GCC v4.4 or higher Using compiler: g++ Unable to find a compiler officially supported by v8. It is recommended to use GCC v4.4 or higher cc1plus: warnings being treated as errors In file included from ../src/elements.h:33, from ../src/objects-inl.h:38, from ../src/v8.h:60, from ../src/accessors.cc:28: ../src/heap.h: In member function 'v8::internal::byte** v8::internal::Heap::store_buffer_top_address()': ../src/heap.h:1317: warning: dereferencing type-punned pointer will break strict-aliasing rules gmake[1]: *** [/usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/v8_base/src/accessors.o] Error 1 gmake: *** [x64.release] Error 2 /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/ext/libv8/location.rb:36:in `block in verify_installation!': libv8 did not install properly, expected binary v8 archive '/usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/tools/gyp/libv8_base.a'to exist, but it was not found (Libv8::Location::Vendor::ArchiveNotFound) from /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/ext/libv8/location.rb:35:in `each' from /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/ext/libv8/location.rb:35:in `verify_installation!' from /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/ext/libv8/location.rb:26:in `install!' from extconf.rb:7:in `main' GYP_GENERATORS=make \ build/gyp/gyp --generator-output=out build/all.gyp \ -Ibuild/standalone.gypi --depth=. \ -Dv8_target_arch=x64 \ -S.x64 -Dv8_enable_backtrace=1 -Dv8_can_use_vfp2_instructions=true -Darm_fpu=vfpv2 -Dv8_can_use_vfp3_instructions=true -Darm_fpu=vfpv3 gmake[1]: Entering directory `/usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out' CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/allocation.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/atomicops_internals_x86_gcc.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/bignum.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/bignum-dtoa.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/cached-powers.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/conversions.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/diy-fp.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/dtoa.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/fast-dtoa.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/fixed-dtoa.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/once.o CXX(target) /usr/local/lib/ruby/gems/2.0/gems/libv8-3.16.14.1/vendor/v8/out/x64.release/obj.target/preparser_lib/src/preparse-data.o CXX(target)
Re: ruby on openbsd
Jeremy Evans [jeremyeva...@gmail.com] wrote: You didn't mention what happens when you run `bundle install`. As the rails app you are working on uses bundler, running `bundle install` is basically the first thing you should do after cloning the repository. I got the bright idea to ln -s /usr/local/bin/eg++ /usr/local/bin/g++46 and then bundle got further... but obviously this version of libv8 doesn't even support openbsd. Installing libv8 (3.11.8.13) with native extensions Gem::Installer::ExtensionBuildError: ERROR: Failed to build gem native extension. /usr/local/bin/ruby18 extconf.rb creating Makefile which: g++44: Command not found. ../src/platform-linux.cc:34:23: fatal error: sys/prctl.h: No such file or directory compilation terminated. gmake[1]: *** [/usr/local/lib/ruby/gems/1.8/gems/libv8-3.11.8.13/vendor/v8/out/x64.release/obj.target/v8_base/src/platform-linux.o] Error 1 gmake: *** [x64.release] Error 2 Using compiler: /usr/local/bin/g++46 GYP_GENERATORS=make \ build/gyp/gyp --generator-output=out build/all.gyp \ -Ibuild/standalone.gypi --depth=. \ -Dv8_target_arch=x64 \ -S.x64 -Dv8_can_use_vfp_instructions=true gmake[1]: Entering directory `/usr/local/lib/ruby/gems/1.8/gems/libv8-3.11.8.13/vendor/v8/out'
Re: ruby on openbsd
Ted Unangst [t...@tedunangst.com] wrote: On Fri, May 03, 2013 at 15:46, Jeremy Evans wrote: Basically, the native code in the libv8 gem doesn't compile correctly on OpenBSD. It probably requires patches (lang/libv8 requires patches to compile). There are two ways to go about fixing it: 1) Get libv8 to work. You could try installing an updated gcc port and see if its g++ will work. If not, you could try building an OpenBSD port for the ruby libv8 gem, and either getting it to use the system libv8, or porting over the lang/libv8 patches so that it builds. From the libv8 gem readme, it appears you can install it with the system v8 as so: gem install libv8 -- --with-system-v8 And then start all over and hope you can pass go. That is totally fucking sweet.
Re: mplayer: Audio device got stuck
Jan Stary [h...@stare.cz] wrote: While playing audio, mplayer accasionaly gets stuck with A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% Audio device got stuck! A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% Audio device got stuck! A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% Audio device got stuck! A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% Audio device got stuck! A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% Audio device got stuck! A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% Audio device got stuck! A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% Audio device got stuck! A:1036.4 (17:16.3) of 1913.0 (31:53.0) 1.2% It seems some check was introduced into mplayer about a year ago http://mplayerhq.hu/pipermail/mplayer-cvslog/2012-February.txt whether the audio device times out too much. I am seeing this more if I simultaneously run firefox, or xombrero with many open windows. Is anybody seeing the same? My audio device works fine beside this. Appears to be something that will be improved if/when ratchov gets his audio MP lock latency diff in the kernel. You can get the same effect by using GENERIC instead of GENERIC.MP.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ch...@cvs.openbsd.org 2013/02/12 09:23:33 Modified files: shells/nsh : Makefile distinfo shells/nsh/pkg : DESCR Removed files: shells/nsh/patches: patch-conf_c patch-if_c Log message: update to 1.0 ok sthen, naddy
Re: scribus displays strangely and emits errors
Tim Howe [th...@bendtel.net] wrote: Chris Cappuccio pointed out to me that it is probably a resource issue. It then occurred that it only works fine when PostgreSQL isn't running. I made the sysctl changes as specified in the PostgreSQL pkg-readme and that appears to have resolved that issue. ipcs can be helpful in this regard, when anything crashes and doesn't clean up its resources, or needs resources in excess of what is default (which can be the case with postgresql and large databases)
Re: NEW: editors/vigor
It can only exist in the ports tree with proper balance. Imagine a world without yin AND yang, Coke and Pepsi, Ford and Chevy, tomato and tomato, democrat and republican, etc. There need to be proper balancing, countermading forces to keep vigor from spiraling into a zombie drug causing our nation's youths to eat faces. Alexander Bluhm [alexander.bl...@gmx.net] wrote: Hi, Do we want to allow the evil vigor editor to reside in our ports tree? Based on the User Friendly comic strip storyline from 4 to 14 January 2000, Vigor brings all the features of traditional Unix vi, plus the friendly and helpful Vigor Assistant. http://vigor.sourceforge.net/ bluhm -- Keep them laughing half the time, scared of you the other half. And always keep them guessing. -- Clair George
Re: Rollback to 0.8.9/0.12.9 (Was: Unresponsive -current + rtorrent)
David Coppa [dco...@gmail.com] wrote: This new release is an epic fail, I don't want to ship it with 5.2. Apart from the huge ipi storms that lead to system freezes, it also segfaults with strict malloc flags :( Latest fix to src/sys/kern/kern_synch.c from haesbaert@ does not help here... Has anyone tried running the newer rtorrent version with kernels from sthen's archive and see if the IPI problem exists throughout, or is new? I know this gets tricky around the __tfork and earlier rthreads changes. On the contrary 0.8.9/0.12.9 is much more stable, with ipi rarely over 100 and has no problems with malloc J flag or S. So, here's my rollback diff. Updating from 0.13.2/0.9.2 with 'pkg_add -ui' works fine (I've added EPOCH). I'd like to commit it, if nobody has objections. Your honor, I object. This man trashed my apartment. AND HE INTERRUPTED ME WHILE I WAS WATCHING OW, MY BALLS!
chrome embedded pdf viewer
Newer versions of chromium on OpenBSD have a working embedded pdf viewer. It only works for certain types of PDFs embedded, when you click on a pdf document, instead of opening it in the viewer, it always wants to download it. Has anyone found a way to fix this?
alpine
has anyone else noticed that alpine is very incompatible with rthreads? i sent diagnostics to guenther that were so bad, he didn't even reply.
Re: alpine
Marc Espie [es...@nerim.net] wrote: On Mon, Mar 05, 2012 at 02:06:05PM -0800, Chris Cappuccio wrote: has anyone else noticed that alpine is very incompatible with rthreads? Nope. Nobody uses alpine. I don't either, but some other folks do. i sent diagnostics to guenther that were so bad, he didn't even reply. More seriously, when ? guenther has been fixing a huge amount of things concerning threads lately, some are still in transit. I know. He is very thorough. And he usually replies very fast but the pathetic nature of my bug reports shut people down. So, hey, first try again with a really current system. Feb 27th kernel and library. I don't think anything significant has gone in since that time. Then, try to figure out what's wrong. Simple test cases do help. Alpine is not a simple testcase. Also, how come guenther gets diagnostics and we don't ? I have no idea what your bug report looked like. Maybe he didn't reply because he had better bug reports to look at ? Probably. Here's what I sent to him. Alpine alternates between crashing with sig11 when trying to send mail via SMTP, or failing to send mail through SMTP with a bogus SMTP error message (but not crashing). Sending e-mail always bails after the MAIL FROM in the SMTP session with a bogus 421 SMTP server error message (that the SMTP server didn't send) 13988 alpine GIO fd 8 wrote 34 bytes MAIL FROM:ch...@ref.nmedia.net\r 13988 alpine RET write 34/0x22 13988 alpine CALL read(0x9,0x204e2f000,0x1000) 13988 alpine RET read 0 13988 alpine CALL close(0x9) 13988 alpine PSIG SIGCHLD caught handler=0x4ba0a0 mask=0x0 13988 alpine RET close 0 13988 alpine CALL sigreturn(0x7f7c4ab0) 13988 alpine RET sigreturn JUSTRETURN 13988 alpine CALL close(0x8) 13988 alpine RET close 0 13988 alpine CALL wait4(0x6401,0x7f7c4e1c,0invalid0,0) 13988 alpine RET wait4 25601/0x6401 13988 alpine CALL sigaction(SIGINT,0x7f7c4e30,0x7f7c4e20) 13988 alpine RET sigaction 0 13988 alpine CALL sigaction(SIGHUP,0x7f7c4e30,0x7f7c4e20) 13988 alpine RET sigaction 0 13988 alpine CALL sigaction(SIGQUIT,0x7f7c4e30,0x7f7c4e20) 13988 alpine RET sigaction 0 13988 alpine CALL gettimeofday(0x7f7c91b0,0x7f7c91c0) 13988 alpine RET gettimeofday 0 13988 alpine CALL gettimeofday(0x7f7c91a0,0x7f7c91b0) 13988 alpine RET gettimeofday 0 13988 alpine CALL gettimeofday(0x7f7c9190,0x7f7c91a0) 13988 alpine RET gettimeofday 0 13988 alpine CALL kill(0xfb132,SIGTHR) 13988 alpine RET kill 0 28402 alpine PSIG SIGTHR caught handler=0x20cbd6de0 mask=0x808 28402 alpine RET nanosleep -1 errno 4 Interrupted system call Here's what happens when you try to send mail and alpine crashes: 13988 alpine GIO fd 1 wrote 81 bytes \^[[23;1H\^[[K\^[[24;1H\^[[K\^[[22;1H\^[[K\^[[7m\^[[22;27H[Sending mail | |]\^[[27m\^[[22;1H 13988 alpine RET write 81/0x51 13988 alpine CALL munmap(0x206518000,0x12000) 13988 alpine RET munmap 0 13988 alpine CALL mmap(0,0x12000,0x3PROT_READ|PROT_WRITE,0x1000MAP_ANON,0x,0) 13988 alpine RET mmap 8710569984/0x20730c000 13988 alpine CALL mprotect(0x20730c000,0x1000,0PROT_NONE) 13988 alpine RET mprotect 0 13988 alpine CALL __tfork(0x7f7c8e80) 13988 alpine RET __tfork 1007295/0xf5ebf 13988 alpine CALL gettimeofday(0x7f7c8d40,0x7f7c8d50) 13988 alpine RET gettimeofday 0 7295 alpine RET rfork 0 13988 alpine CALL gettimeofday(0x7f7c91b0,0x7f7c91c0) 7295 alpine CALL sigprocmask(SIG_BLOCK,0x808) 7295 alpine RET sigprocmask 0 13988 alpine RET gettimeofday 0 7295 alpine CALL nanosleep(0x20731dfb0,0) 13988 alpine CALL access(0x20c079f40,0x1X_OK) 13988 alpine NAMI /usr/sbin/sendmail 13988 alpine RET access 0 13988 alpine CALL pipe(0x7f7c81c0) 13988 alpine RET pipe 0 13988 alpine CALL pipe(0x7f7c81c0) 13988 alpine RET pipe 0 13988 alpine CALL vfork() 26036 alpine PSIG SIGTHR caught handler=0x20cbd6de0 mask=0x0 26036 alpine RET rfork 0 26036 alpine CALL sigreturn(0x7f7c7f50) 26036 alpine RET sigreturn JUSTRETURN 26036 alpine CALL sigprocmask(SIG_BLOCK,0x808) 26036 alpine RET sigprocmask 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036 alpine CALL sched_yield() 26036 alpine RET sched_yield 0 26036
Kamilio/SER
Is anyone working on a Kamilio 3 update by any chance? Also is there any reason why an ancient version of net/ser still exists when Kamilio 1.5 is in the tree?
clamav
PFRAG.shared on the clamav port should list libclamav.so.1.20 and not libclamav.so.0.0 -- Do you even send e-mails? I told you, I'm from the Wild West. I write by hand. -- Chuck Norris
