Re: compatibility
On Wed, Feb 10, 2010 at 6:55 PM, Vadkan Jozsef jozsi.avad...@gmail.com wrote: did anyone managed to use a: huawei e1752 alcatel x200 under openbsd? Are they attaching to umsm? Can you post output of usbdevs -dv ? cheers david
Re: current on HP EliteBook 8530w
This looks same as problem on my hp 6930p. According to Jordan it's some reference counting bug somewhere. This makes it boot... Index: dsdt.c === RCS file: /cvs/src/sys/dev/acpi/dsdt.c,v retrieving revision 1.157 diff -u -p -r1.157 dsdt.c --- dsdt.c 5 Dec 2009 02:38:11 - 1.157 +++ dsdt.c 23 Jan 2010 23:49:46 - @@ -3629,7 +3629,7 @@ aml_xparse(struct aml_scope *scope, int aml_freevalue(opargs[1]); /* Create Object Reference */ - _aml_setvalue(opargs[1], AML_OBJTYPE_OBJREF, opcode, opargs[0]); + _aml_setvalue(opargs[1], AML_OBJTYPE_INTEGER, 0xDEADBEEF, opargs[0]); aml_xaddref(opargs[1], CondRef); /* Mark that we found it */ Yes it does, on my HP 8630w too. Which gives me acpiec(4) back. Which fixes acpibat(4) and acpiac(4). Which fixes apmd(8), who correctly reports the battery/AC status now and dims/brightens the screen accordingly, too. It worked for a while, but recently, having acpiec enabled (with the patch above) results in acpitz0 at acpi0: critical temperature 115 degC acpitz1 at acpi0: critical temperature 105 degC acpitz2 at acpi0: critical temperature 112 degC acpitz3 at acpi0: critical temperature 112 degC acpitz4 at acpi0: critical temperature 90 degC acpitz5 at acpi0: critical temperature 112 degC at boot and the system shuts down (correctly). So I had to disable even the patched acpiec again. Is anyone experiencing the same? Is there any point upgrading to a 4.7-beta snapshot in this regard? Thanks Jan
Re: OpenBGP filter question
On 10.2.2010 P3. 21:32, Stuart Henderson wrote: On 2010-02-10, Ivo Chutkinopen...@bgone.net wrote: Hello misc, Would the following filter work? match to $my_upstream_1 source-as {some_as} set prepend-self 4 I would like to prepend my as to make as path longer for some_as trough my_upstream_1 and make it to prefer path trough my_upstream_2. It does not produce error with bgpd-n but there is no effect as well. Are you certain it has no effect (and how?) - you can't rely on AS path prepending to change how traffic flows, if someone gives you a higher localpref they'll use that path irrespective of the path length. Hi Stuart, I am certain as I don't see my prepend on some_as looking glass. The actual filter looks like this without the comment: match to $spnet_bg #(AS8717) sourse_as 9070 set prepend-seff 4 and this is what I see on 9070 looking glass: inet.0: 5185 destinations, 8315 routes (5184 active, 0 holddown, 1 hidden) + = Active Route, - = Last Active, * = Both A DestinationP Prf Metric 1 Metric 2 Next hopAS path * 87.120.100.0/24B 170115212.116.129.38 34224 20682 I B 170115 99 212.116.129.66 34224 20682 I B 170115212.116.135.81 8717 20682 I {master:0} where 20682 is my as. Filter like: match to $spnet_gl prefix {$net3 $net4 $net5 $net6} set prepend-self 2 works perfect but it prepends all as paths from this neighbor and it changes the routes to me. I am aware of local preference. Thanks for the help, Ivo
Re: current on HP EliteBook 8530w
On Feb 11 10:34:57, Jan Stary wrote: This looks same as problem on my hp 6930p. According to Jordan it's some reference counting bug somewhere. This makes it boot... Index: dsdt.c === RCS file: /cvs/src/sys/dev/acpi/dsdt.c,v retrieving revision 1.157 diff -u -p -r1.157 dsdt.c --- dsdt.c5 Dec 2009 02:38:11 - 1.157 +++ dsdt.c23 Jan 2010 23:49:46 - @@ -3629,7 +3629,7 @@ aml_xparse(struct aml_scope *scope, int aml_freevalue(opargs[1]); /* Create Object Reference */ - _aml_setvalue(opargs[1], AML_OBJTYPE_OBJREF, opcode, opargs[0]); + _aml_setvalue(opargs[1], AML_OBJTYPE_INTEGER, 0xDEADBEEF, opargs[0]); aml_xaddref(opargs[1], CondRef); /* Mark that we found it */ Yes it does, on my HP 8630w too. Which gives me acpiec(4) back. Which fixes acpibat(4) and acpiac(4). Which fixes apmd(8), who correctly reports the battery/AC status now and dims/brightens the screen accordingly, too. It worked for a while, but recently, having acpiec enabled (with the patch above) results in acpitz0 at acpi0: critical temperature 115 degC acpitz1 at acpi0: critical temperature 105 degC acpitz2 at acpi0: critical temperature 112 degC acpitz3 at acpi0: critical temperature 112 degC acpitz4 at acpi0: critical temperature 90 degC acpitz5 at acpi0: critical temperature 112 degC Sorry. These messages are always there and probably describe the temperatures _considered_ to be critical. Right? What it actually says on boot is acpitz2: Critical temperature, shutting down. I don't think there's anything hotter than 112C while the machine boots. With acpiec disabled, the above does not happen and sysctl hw.sensors reports reasonable temperatures; the systme runs fine (except acpibat and others are confused). at boot and the system shuts down (correctly). So I had to disable even the patched acpiec again. Is anyone experiencing the same? Is there any point upgrading to a 4.7-beta snapshot in this regard? Thanks Jan
Re: current on HP EliteBook 8530w
Hi, On Thu, Feb 11, 2010 at 11:34 AM, Jan Stary h...@stare.cz wrote: [...] It worked for a while, but recently, having acpiec enabled (with the patch above) results in acpitz0 at acpi0: critical temperature 115 degC acpitz1 at acpi0: critical temperature 105 degC acpitz2 at acpi0: critical temperature 112 degC acpitz3 at acpi0: critical temperature 112 degC acpitz4 at acpi0: critical temperature 90 degC acpitz5 at acpi0: critical temperature 112 degC Sorry. These messages are always there and probably describe the temperatures _considered_ to be critical. Right? I had the same doubt some time ago, but you are right, these are the temperature limits. What it actually says on boot is acpitz2: Critical temperature, shutting down. I don't think there's anything hotter than 112C while the machine boots. With acpiec disabled, the above does not happen and sysctl hw.sensors reports reasonable temperatures; the systme runs fine (except acpibat and others are confused). I had, with my previous HP nw9440 notebook, the same problem with those sensors, but not always. And more often when the machine was first started, almost never on reboots. I think there is something inside that needs to settle down before the readings make sense. I saw -2000 degrees on one of them, one time. :-) If I waited a few minutes (sometime even a few seconds), pausing in rc.securelevel at the softraid password prompt I scripted, the problem solved itself. You could try putting some sleep xxx inside rc.securelevel and see if the messages continue to pop up on the console or stop. The system does not shut itself down until the boot is over. HTH. Ciao, D.
Strange problem | routing issue
Hello, We are facing a strange problem while trying to use openbgpd. We would like to use openbgpd as our core BGP router on a Dell Poweredge R210 server with quad core xeon/ 8 GB RAM/ 6 x1GB ports (2 on board Broadcom and quad port Intel adapter). 4.6 stable installation failed because it could not detect hard disk (I guess its related to the controller). Then we installed current 4.7 amd64(10 Feb build) which detected the HDD and installed properly. After configuring BGPD everything camp up nicely. --The Setup- Customers-Cisco SW 2950 (8xVLAN)-(1GB Vlan trunk)--|||BGPD Server||-(100Mbps)--Upstream carrier - -- Bgpd.conf # $OpenBSD: bgpd.conf,v 1.9 2009/11/20 19:51:05 claudio Exp $ # global configuration AS xxx13 router-id xxx.xxx.53.1 network xxx.xxx.52.0/22 # neighbors and peers neighbor xxx.xxx.58.21 { remote-as xx55 descr upstream announceall tcp md5sig password xxx } neighbor xxx.xxx.52.202 { remote-as xxx23 descr customer announcedefault-route tcp md5sig password x multihop3 local-address xxx.xxx.53.9 } # filter out prefixes longer than 24 or shorter than 8 bits deny from any allow from any inet prefixlen 8 - 24 # accept a default route (since the previous rule blocks this) allow from xxx.xxx.58.21 prefix 0.0.0.0/0 allow from xxx.xxx.52.202 prefix xxx.xxx.168.0/24 # filter bogus networks deny from any prefix 10.0.0.0/8 prefixlen = 8 deny from any prefix 172.16.0.0/12 prefixlen = 12 deny from any prefix 192.168.0.0/16 prefixlen = 16 deny from any prefix 169.254.0.0/16 prefixlen = 16 deny from any prefix 192.0.2.0/24 prefixlen = 24 deny from any prefix 224.0.0.0/4 prefixlen = 4 deny from any prefix 240.0.0.0/4 prefixlen = 4 Network interfaces (Hostname.em0) upstream carrier inet xxx.xxx.58.22 255.255.255.252 xxx.xxx.58.23 media 100baseTX mediaopt full-duplex description Upstream carrier (Hostname.vlan101) cisco router facing customers and bgpd (for TDM interfaces) inet xxx.xxx.53.1 255.255.255.252 xxx.xxx.53.3 vlandev em3 description Cisco 2821 ! route add xxx.xxx.53.16/30 xxx.xxx.53.2 ! route add xxx.xxx.53.200/29 xxx.xxx.53.2 ! route add xxx.xxx.53.20/30 xxx.xxx.53.2 ! route add xxx.xxx.53.92/30 xxx.xxx.53.2 ! route add xxx.xxx.53.40/30 xxx.xxx.53.2 ! route add xxx.xxx.53.112/28 xxx.xxx.53.2 ! route add xxx.xxx.53.52/30 xxx.xxx.53.2 ! route add xxx.xxx.53.248/29 xxx.xxx.53.2 ! route add xxx.xxx.53.80/30 xxx.xxx.53.2 ! route add xxx.xxx.53.84/30 xxx.xxx.53.2 ! route add xxx.xxx.53.88/30 xxx.xxx.53.2 (Hostname.vlan102) Customer1 inet xxx.xxx.53.9 255.255.255.252 xxx.xxx.53.11 vlandev em3 description customer ! route add xxx.xxx.52.0/24 xxx.xxx.53.10 ! route add xxx.xxx.54.0/25 xxx.xxx.53.10 (hostname.vlan103) Customer2 inet xxx.xxx.53.13 255.255.255.252 xxx.xxx.53.15 vlandev em3 description customer2 ! route add xxx.xxx.53.192/29 xxx.xxx.53.14 And like that we have around 8 customers all going thru vlan trunk configured between bgpd server and cisco switch. ---The Problem- As soon as we start traffic bgp server starts behaving strangely. for example if we ping any IP, customer side or towards upstream from the bgpd server, first few seconds we get no route to host and after few seconds it starts getting the response. When we try to ping the same IP again, behavior remain unchanged. which means it can't get the route for few seconds. We have checked the media connections and under normal conditions (without bgpd and vlans) everything works fine. I am not sure if its related to so many vlans, bgpd configuration or bug in routing daemon. Another strange thing I have noticed is that it shows 10M memory in TOP command where server has 8GB RAM. I am pasting the dmesg and other logs for your info. - System info CPU0 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU1 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU2 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU3 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU4 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU5 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU6 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle CPU7 states: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle Memory: Real: 11M/59M act/tot Free: 2915M Swap: 0K/3318M used/tot PID USERNAME PRI NICE SIZE RES STATE WAIT TIMECPU COMMAND 9436 root 20 3428K 3132K sleep/1 select0:00 0.00% sshd 16649 root 20 3368K 3096K idle select0:00 0.00% sshd 18987 _bgpd 20 776K 1220K idle poll
Валютные (суммовые и курсовые) разницы в современных условиях, с учетом займовezz
17 QP5P2QP0P;Q 2010P3. PPQP:P2P0. PP0P;QQP=QP5 (QQ2QP5 P8 P:QQQPP2QP5) QP0P7P=P8QQ P2 QPP2QP5PP5P=P=QQ QQP;PP2P8QQ, Q QQP5QPP P7P0P9PPP2. = P!P5PP8P=P0Q P?QP5P4P=P0P7P=P0QP5P= P4P;Q: QP8P=P0P=QPP2QQ QP0P1PQP=P8P:PP2 P:PPP?P0P=P8P9. P PP PPP PPPP PQQQPP2QP5 QP0P7P=P8QQ. * P$PQPP8QPP2P0P=P8P5 P4PP3PP2PQP=QQ PQP=PQP5P=P8P9 P?QP8 QP0QQP5QP0Q P2 P2P0P;QQP5. * PP?QP5P4P5P;P5P=P8P5 P:QQQP0 P?QP8 QP0QQP5QP0Q P2 P2P0P;QQP5. * PP5QP2P8QP=QP5 P4PP:QPP5P=QQ P?QP8 QP0QQP5QP0Q P2 P2P0P;QQP5. * PQQP3P0P;QP5QQP:P8P9 QQP5Q P?P P:QQQPP2QP QP0P7P=P8QP0P. * PQQQPP2QP5 QP0P7P=P8QQ P2 P1QQP3P0P;QP5QQP:PP9 PQQP5QP=PQQP8. * PP0P;PP3PP2QP9 QQP5Q P?P P:QQQPP2QP QP0P7P=P8QP0P. * PPP! P?P P:QQQPP2QP QP0P7P=P8QP0P. * PQPQPP;P5P=P8P5 QQP5QP0-QP0P:QQQQ, P2 QP2QP7P8 Q P=P0P;P8QP8P5P QP0QQP5QPP2 P2 P2P0P;QQP5. * PPP;PP6P8QP5P;QP=QP5 P:QQQPP2QP5 QP0P7P=P8QQ Q P?QPP4P0P2QP0 (P2 QPP QP8QP;P5 Q P:QP?PQQ QPP2P0QPP2). * PQQP8QP0QP5P;QP=QP5 P:QQQPP2QP5 QP0P7P=P8QQ Q P?QPP4P0P2QP0 (P2 QPP QP8QP;P5 Q P:QP?PQQ QPP2P0QPP2). * PPP;PP6P8QP5P;QP=QP5 P:QQQPP2QP5 QP0P7P=P8QQ Q P?PP:QP?P0QP5P;Q (P2 QPP QP8QP;P5 P8PP?PQQ QPP2P0QPP2). * PQQP8QP0QP5P;QP=QP5 P:QQQPP2QP5 QP0P7P=P8QQ Q P?PP:QP?P0QP5P;Q (P2 QPP QP8QP;P5 P8PP?PQQ QPP2P0QPP2). * PQQQPP2QP5 QP0P7P=P8QQ P?QP8 P?QP8PP1QP5QP5P=P8P8 P?PQQP5P4P=P8QP5QP:P8Q QQP;QP3. * PQQQPP2QP5 QP0P7P=P8QQ P?P P4PP3PP2PQP0P P0QP5P=P4Q P8 P8P=QQ QQP;QP3 b 5P=QQ PP?QP5P4P5P;P5P=P8Q P:QQQP0. * PQQQPP2QP5 QP0P7P=P8QQ P?QP8 PP:P0P7P0P=P8P8 P?PQQP5P4P=P8QP5QP:P8Q QQP;QP3. * PQQQPP2QP5 QP0P7P=P8QQ, P2PP7P=P8P:P0QQP8P5 P?QP8 P?PP:QP?P:P5 P8P;P8 P?QPP4P0P6P5 P8P=PQQQP0P=P=PP9 P2P0P;QQQ. * PQQQPP2QP5 QP0P7P=P8QQ P?P P2P0P;QQP=PPQ QQP5QQ. * PQQQPP2QP5 QP0P7P=P8QQ P?P QP0QQP5QP0P P?QP5P4QQP0P2P8QP5P;QQQP2P0 QPQQP8P9QP:PP9 PQP3P0P=P8P7P0QP8P8 P7P0 QQP1P5P6PP. * PQQQPP2QP5 QP0P7P=P8QQ P?P QP0QQP5QP0P P?QP5P4QQP0P2P8QP5P;QQQP2P0 P8P=PQQQP0P=P=PP9 PQP3P0P=P8P7P0QP8P8 Q P3PP;PP2P=PP9 PQP3P0P=P8P7P0QP8P5P9. * PQQQPP2QP5 QP0P7P=P8QQ P?P P?PP;QQP0P5PQP QP5P;P5P2PPQ QP8P=P0P=QP8QPP2P0P=P8Q P8 P?PQQQP?P;P5P=P8QP. * PP7P=PQQ P2 QQQP0P2P=QP9 P:P0P?P8QP0P;, P2P0P;QQP=QP5 QP5P=P=QP5 P1QPP0P3P8 P8 P8P=QP5 P2P=P5PP1PQPQP=QP5 P0P:QP8P2Q. * PP0P;QQP=QP9 QQQP0P2P=QP9 P:P0P?P8QP0P;, QP5P=P=QP5 P1QPP0P3P8 P8 P8P=QP5 P2P=P5PP1PQPQP=QP5 P?P0QQP8P2Q. * PQQQPP2QP5 QP0P7P=P8QQ P?P P4PP;P3PP2QP PP1QP7P0QP5P;QQQP2P0P P8 P?QPQP5P=QP0P Q P:QP5P4P8QPQP0. * PQQQPP2QP5 QP0P7P=P8QQ P?P P4PP;P3PP2QP PP1QP7P0QP5P;QQQP2P0P P8 P?QPQP5P=QP0P Q P4PP;P6P=P8P:P0. P!Q2QP5 QP0P7P=P8QQ. * PP0QP0QQQPQP8QP5QP:P8P5 P?PQP;P5P4QQP2P8Q PP8QQPP0 P$PP! P P$ PQ 24.08.2009 N 3-1-07/674 P P?QP8PP5P=P5P=P8P8 PPP! P?P P4PP3PP2PQP0P PP1 PP?P;P0QP5 P2 QQP;PP2P=QQ P5P4P8P=P8QP0Q P8 P8Q P?QP5PP4PP;P5P=P8P5. * P$PQPP8QPP2P0P=P8P5 P4PP3PP2PQP=QQ PQP=PQP5P=P8P9 P?QP8 QP0QQP5QP0Q P2 QQP;PP2P=QQ P5P4P8P=P8QP0Q. PP?QP5P4P5P;P5P=P8P5 P:QQQP0 P?QP8 QP0QQP5QP0Q P2 QQP;PP2P=QQ P5P4P8P=P8QP0Q. * PP5QP2P8QP=QP5 P4PP:QPP5P=QQ P?QP8 QP0QQP5QP0Q P2 QQP;PP2P=QQ P5P4P8P=P8QP0Q. PQQP3P0P;QP5QQP:P8P9 QQP5Q P?P QQ2QP QP0P7P=P8QP0P. P!Q2QP5 QP0P7P=P8QQ P2 P1QQP3P0P;QP5QQP:PP9 PQQP5QP=PQQP8. PP0P;PP3PP2QP9 QQP5Q P?P QQ2QP QP0P7P=P8QP0P. PPP! P?P QQ2QP QP0P7P=P8QP0P. PQPQPP;P5P=P8P5 QQP5QP0-QP0P:QQQQ, P2 QP2QP7P8 Q P=P0P;P8QP8P5P QP0QQP5QPP2 P2 QQP;PP2P=QQ P5P4P8P=P8QP0Q. * PPP;PP6P8QP5P;QP=QP5 QQ2QP5 QP0P7P=P8QQ Q P?QPP4P0P2QP0. PQQP8QP0QP5P;QP=QP5 QQ2QP5 QP0P7P=P8QQ Q P?QPP4P0P2QP0. PPP;PP6P8QP5P;QP=QP5 QQ2QP5 QP0P7P=P8QQ Q P?PP:QP?P0QP5P;Q. PQQP8QP0QP5P;QP=QP5 QQ2QP5 QP0P7P=P8QQ Q P?PP:QP?P0QP5P;Q. * P!Q2QP5 QP0P7P=P8QQ P?QP8 P?QP8PP1QP5QP5P=P8P8 P?PQQP5P4P=P8QP5QP:P8Q QQP;QP3. P!Q2QP5 QP0P7P=P8QQ P?P P4PP3PP2PQP0P P0QP5P=P4Q P8 P8P=QQ QQP;QP3 b 5P=QQ PP?QP5P4P5P;P5P=P8Q P:QQQP0. P!Q2QP5 QP0P7P=P8QQ P?QP8 PP:P0P7P0P=P8P8 P?PQQP5P4P=P8QP5QP:P8Q QQP;QP3. * P P0QQP5QQ P2 QQP;PP2P=QQ
Re: routing and pf at 10Gbps
On Wed, 10 Feb 2010, Mike Williams wrote: Really, nobody firewalls at multi-Gbps? yes, people run firewalls on 10G circuits I am not aware of anyone filtering at 10G who is using off the shelf hardware, with open source O/S. Large enterprises use either commercial firewalls, for example Juniper Netscreens, or build systems using FPGA cards with locally produced code. Either way the filtering is done in hardware. In my experience the Netscreen 5x00 firewalls sold with 10G cards and MGT3 card can not do line rate 10G, though it was marketed as capable of 10G filtering. The newer, ie more expensive Juniper SRX firewalls supposedly can do it. They are based on Juniper heavy iron routers. diana
SEMINARIO INTERNACIONAL: ¿CÓMO REALIZAR ENTREVISTAS DE SELECCIÓN DE PERSONAL POR COMPETENCIAS?-Técnicas de evaluación y registro de evidencias conductuales
Emarketing de Costa Rica S.A. Ofrecemos: !!! Envios a nuestras bases de datos mas de 300.000 emails!!! !!! Video emails... Lo ultimo en publicidad por Correo Electronico !!! !!! Envios a sus bases de datos !!! ( seguridad y confiabilidad ) !!! Plataforma personal para que haga sus envios !!! ( Patrocinadores de aldeas S.O.S y fundacion Samuel ) Telefonos (506 ) 22 97 14 04 / (506) 89 24 62 64 !!! Nueve aC1os de experiencia !!! www.emarketingcr.net / ven...@emarketingcr.net Si desea desinscribirse haga click Aqui [IMAGE]
Re: Azalia problem: no sound
On Sun, Jan 31, 2010 at 03:00:24AM +, Jacob Meuser wrote: On Sun, Jan 31, 2010 at 01:53:50AM +, Jacob Meuser wrote: On Sat, Jan 30, 2010 at 07:22:35PM +0100, Holger Mikolon wrote: Hi, a similar fix works for my Dell Studio 1555 (details below). I only had to adjust the subid. I guess this should probably apply to all Dell with IDT 92HD73* codecs. that seems to be what linux is doing. please test this if you have a Dell machine with an azalia with an IDT codec. -- jake...@sdf.lonestar.org SDF Public Access UNIX System - http://sdf.lonestar.org This works well on the Dell E4300 ok drahn@ Index: azalia_codec.c === RCS file: /cvs/src/sys/dev/pci/azalia_codec.c,v retrieving revision 1.139 diff -u -p azalia_codec.c --- azalia_codec.c3 Nov 2009 17:31:30 - 1.139 +++ azalia_codec.c31 Jan 2010 02:57:39 - @@ -166,6 +166,9 @@ azalia_codec_init_vtbl(codec_t *this) break; case 0x111d7675: this-name = IDT 92HD73C1;/* aka 92HDW74C1 */ + if ((this-subid 0x) == 0x1028) { /* DELL */ + this-qrks |= AZ_QRK_GPIO_UNMUTE_0; + } break; case 0x111d7676: this-name = IDT 92HD73E1;/* aka 92HDW74E1 */ @@ -175,10 +178,7 @@ azalia_codec_init_vtbl(codec_t *this) break; case 0x111d76b2: this-name = IDT 92HD71B7; - if (this-subid == 0x02631028 ||/* DELL_E5500 */ - this-subid == 0x02501028 || /* DELL_M4400 */ - this-subid == 0x02331028 ||/* DELL_E6400 */ - this-subid == 0x024f1028) {/* DELL_E6500 */ + if ((this-subid 0x) == 0x1028) { /* DELL */ this-qrks |= AZ_QRK_GPIO_UNMUTE_0; } break; Dale Rahn dr...@dalerahn.com
Re: writing to usb very slow
On Sat, Jan 16, 2010 at 01:44:49AM +0100, T. Tofus von Blisstein wrote: Hello, I have noticed that writing to a usb drive is slow. What does slow mean? It means that compared to other OS's. Which OS's? Yes, it hurts: the penguin. Hello there, I came across a cheap USB flash drive, particularly a SanDisk Cruzer U3: umass0 at uhub1 port 1 configuration 1 interface 0 SanDisk SanDisk Cruzer rev 2.00/2.00 addr 2 umass0: using SCSI over Bulk-Only scsibus3 at umass0: 2 targets, initiator 0 sd1 at scsibus3 targ 1 lun 0: SanDisk, SanDisk Cruzer, 8.02 SCSI0 0/direct removable sd1: 3863MB, 512 bytes/sec, 7913471 sec total Reading and writing from/to a raw disk /dev/rsd1c gives same exact rates as on other OSes: $ sudo dd if=/dev/rsd1c of=/dev/null bs=128k count=1600 1600+0 records in 1600+0 records out 209715200 bytes transferred in 8.822 secs (23771541 bytes/sec) $ sudo dd if=/dev/zero of=/dev/rsd1c bs=128k count=1600 1600+0 records in 1600+0 records out 209715200 bytes transferred in 19.253 secs (10892315 bytes/sec) Which are reasonable numbers. Now let's put a filesystem on it: $ sudo fdisk -i sd1 Do you wish to write new MBR and partition table? [n] y Writing MBR at offset 0. $ sudo disklabel -E sd1 Label editor (enter '?' for help at any prompt) a partition: [a] offset: [63] size: [7903917] FS type: [4.2BSD] w q No label changes. $ sudo newfs sd1a /dev/rsd1a: 3859.3MB in 7903916 sectors of 512 bytes 20 cylinder groups of 202.47MB, 12958 blocks, 25984 inodes each super-block backups (for fsck -b #) at: 32, 414688, 829344, 1244000, 1658656, 2073312, 2487968, 2902624, 3317280, 3731936, 4146592, 4561248, 4975904, 5390560, 5805216, 6219872, 6634528, 7049184, 7463840, 7878496, $ sudo mount /dev/sd1a /mnt/test; cd /mnt/test $ sudo dd if=/dev/zero of=./x bs=128k count=1600 1600+0 records in 1600+0 records out 209715200 bytes transferred in 254.224 secs (824923 bytes/sec) (confirmed that pengingwrites is 0 right after dd exits with systat iostat) As you can see here, 800KB/s is quite low, compared to raw read rate. What can be the cause? I don't know, but let's try formatting sd1a with frag-size of 8192 and block-size of 65536 ... (same exact sequence of steps as above, just adjust frag-size and block-size) ... $ time (sudo dd if=/dev/zero of=./x bs=128k count=3200; sync) 3200+0 records in 3200+0 records out 419430400 bytes transferred in 21.194 secs (19789417 bytes/sec) 0m42.44s real 0m0.00s user 0m0.33s syste 419430400/42.44=9,882,902.92 That's almost 10MB/s Perhaps people who are having problems with their slow USB flash drives can try the same experiment, see if it helps.
Re: anyone need old PC crap?
On Mon, Feb 8, 2010 at 6:44 AM, J.C. Roberts list-...@designtools.org wrote: On Sun, 07 Feb 2010 22:12:06 -0500 Nick Holland n...@holland-consulting.net wrote: ropers wrote: You (or anyone else, really) wouldn't happen to have any 1st or 2nd generation PC stuff (as in, IBM 5150 PC / IBM 5155 Portable, or IBM 5160 PC XT)? http://en.wikipedia.org/wiki/IBM_5150 http://en.wikipedia.org/wiki/IBM_5155 http://en.wikipedia.org/wiki/IBM_5160 please answer off-list. Do not feed the old computer crap addition I have... :-/ Your subconscious speaks otherwise. lulz...
Re: Strange problem | routing issue
Shailesh Tyagi [shail...@novanet.net] wrote: As soon as we start traffic bgp server starts behaving strangely. for example if we ping any IP, customer side or towards upstream from the bgpd server, first few seconds we get no route to host and after few seconds it starts getting the response. When we try to ping the same IP again, behavior remain unchanged. which means it can't get the route for few seconds. We have checked the media connections and under normal conditions (without bgpd and vlans) everything works fine. I am not sure if its related to so many vlans, bgpd configuration or bug in routing daemon. Another strange thing I have noticed is that it shows 10M memory in TOP command where server has 8GB RAM. I am pasting the dmesg and other logs for your info. - It's hard to tell from the jumble of obfuscated output that you posted to the list, it sounds like you have a larger network (/24?) directly configured on a local interface, and you are expecting to reach parts of it (/30, /27, etc...) through BGP. You should better plan your network configuration so that you have no overlapping subnets on any of your router interfaces. Chris
Re: routing and pf at 10Gbps
* Diana Eichert deich...@wrench.com [2010-02-11 17:02]: On Wed, 10 Feb 2010, Mike Williams wrote: Really, nobody firewalls at multi-Gbps? yes, people run firewalls on 10G circuits I am not aware of anyone filtering at 10G who is using off the shelf hardware, with open source O/S. I know of some. I don't remember specifics, dunno wether anybody does linerate and with what kind of packet characteristics. for the OP, I think it has a fair chance to work out, given the hardware is picked right. there might be some experiments needed. basically you want the fastest single core possible (you'll get a multicore CPU and let the others idle), and, most importantly, fastest memory access possible, thus an architecture with fast caches - nehalem should beat core2 there, but i haven't tried yet. you want to run i386 instead of amd64 (this needs a re-check really, it's been long that we did that and amd64 changed a lot). good nics help a lot, but i dunno which 10G ones to use. bus bandwidth/latency etc should not make a difference here on any quite recent not crap hardware you can buy, disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. -- Henning Brauer, h...@bsws.de, henn...@openbsd.org BS Web Services, http://bsws.de Full-Service ISP - Secure Hosting, Mail and DNS Services Dedicated Servers, Rootservers, Application Hosting
Re: routing and pf at 10Gbps
Alcatel-Lucent do a AA-ISA card plugin module for their 7750 range of routers. Which enables you to do filtering at 50GB (and scale it up to 800GB) per 12U router. Having recently investigated this segment for work. Allot, Sonicwall(which is a Linux Variant) and a few others are running FOSS firewalls filtering appliances at 10GB+ and it's not just the router vendors (nortel, cisco, junper,alcatel) that do that sort of speed these days. I can't comment on the hardware blobs that may also be involved in these, as Diana says they will have FPGA's under a freeOS... i.e JunOS is essentially 4.4 BSD with a bunch of Juniper FPGA drivers. -JoelW On 12 February 2010 04:54, Diana Eichert deich...@wrench.com wrote: On Wed, 10 Feb 2010, Mike Williams wrote: Really, nobody firewalls at multi-Gbps? yes, people run firewalls on 10G circuits I am not aware of anyone filtering at 10G who is using off the shelf hardware, with open source O/S. Large enterprises use either commercial firewalls, for example Juniper Netscreens, or build systems using FPGA cards with locally produced code. Either way the filtering is done in hardware. In my experience the Netscreen 5x00 firewalls sold with 10G cards and MGT3 card can not do line rate 10G, though it was marketed as capable of 10G filtering. B The newer, ie more expensive Juniper SRX firewalls supposedly can do it. B They are based on Juniper heavy iron routers. diana
Ganhe um voucher de 500 EUROS na loja FNAC !!!
Recebe um voucher FNAC de 500 NOTA INFORMATIVA: O presente email destina-se znica e exclusivamente a informar potenciais utilizadores e nco pode ser considerado SPAM. De acordo com a legislagco internacional que regulamenta o correio electrsnico, o email nco pode sera ser considerado SPAM quando incluir uma forma do receptor ser removido da lista do emissor. Para deixar de receber estas ofertas no seu e-mail clicar aqui
Re: routing and pf at 10Gbps
On Wed, Feb 10, 2010 at 07:57:44PM +, Mike Williams wrote: Really, nobody firewalls at multi-Gbps? I know some folks at NASA that use OpenBSD firewalls that would make your head spin. And yes, that means multi-Gbps. -- Jason Dixon DixonGroup Consulting http://www.dixongroup.net/
Re: routing and pf at 10Gbps
On Thu, 11 Feb 2010 15:26:18 -0500 Jason Dixon ja...@dixongroup.net wrote: On Wed, Feb 10, 2010 at 07:57:44PM +, Mike Williams wrote: Really, nobody firewalls at multi-Gbps? I know some folks at NASA that use OpenBSD firewalls that would make your head spin. And yes, that means multi-Gbps. Oh my, what a nice article a writeup of their setup would make for undeadly's OpenBSD in the trenches department.
Re: routing and pf at 10Gbps
On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Best, Daniel
Re: routing and pf at 10Gbps
Daniel Ouellet wrote: On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Best, Daniel http://www.youtube.com/watch?v=DF7MroTLDfU
Re: routing and pf at 10Gbps
On Thu, Feb 11, 2010 at 03:07:28PM -0500, Daniel Ouellet wrote: On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Henning, I told you, we should not talk about unfinsihed projects. We planned to announce this in exactly 7 weeks. Anyway, to late, the cat is out of the bag. So Henning and Oga are working at offloading pf into the graphic card cores by using the DRI interface. The shader will evaluate the ruleset and packets in parallel and use the graphic memory for the state table. Additionally if the speed of one card is not enough you can use SLI or crossfire to use multiple cards in parallel. -- :wq Claudio It is just a 3-line diff
Re: routing and pf at 10Gbps
--- On Thu, 2/11/10, Claudio Jeker cje...@diehard.n-r-g.com wrote: From: Claudio Jeker cje...@diehard.n-r-g.com Subject: Re: routing and pf at 10Gbps To: misc@openbsd.org Received: Thursday, February 11, 2010, 5:24 PM On Thu, Feb 11, 2010 at 03:07:28PM -0500, Daniel Ouellet wrote: On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Henning, I told you, we should not talk about unfinsihed projects. We planned to announce this in exactly 7 weeks. Anyway, to late, the cat is out of the bag. So Henning and Oga are working at offloading pf into the graphic card cores by using the DRI interface. The shader will evaluate the ruleset and packets in parallel and use the graphic memory for the state table. Additionally if the speed of one card is not enough you can use SLI or crossfire to use multiple cards in parallel. -- :wq Claudio It is just a 3-line diff You have *got* to be kidding me. - head explodes - --- James A. Peltier james_a_pelt...@yahoo.ca
Re: routing and pf at 10Gbps
On Thu, 2010-02-11 at 23:24 +0100, Claudio Jeker wrote: So Henning and Oga are working at offloading pf into the graphic card cores by using the DRI interface. The shader will evaluate the ruleset and packets in parallel and use the graphic memory for the state table. Additionally if the speed of one card is not enough you can use SLI or crossfire to use multiple cards in parallel. Does this mean I will be able to write match in all scrub (no-df asintropic-filter x4) in my pf.conf and instantly beautify my packets at 10Gb wirespeed? If not a pony would be nice too. Luca
Re: routing and pf at 10Gbps
On Thursday 11 February 2010 19:46:22 Henning Brauer wrote: yes, people run firewalls on 10G circuits I am not aware of anyone filtering at 10G who is using off the shelf hardware, with open source O/S. I know of some. Thanks very much guys, much useful information. I'll be sure to purchase servers with the fastest graphics cards possible! We had a Juniper tech in today, and he suggested the SRX-3600. It's almost certainly faster than any single software platform can do (~18Gbps IMIX), but is it worth 10 times the price, and 5U of rack space? Especially for a parallisable problem. Junos can't come close to the scriptable service monitoring and route manipulation that almost trivially easy on openbsd either. -- Mike Williams
Re: routing and pf at 10Gbps
On Thursday 11 February 2010 22:24:46 Claudio Jeker wrote: Henning, I told you, we should not talk about unfinsihed projects. We planned to announce this in exactly 7 weeks. Anyway, to late, the cat is out of the bag. OpenBSD is going to rock in April! -- Mike Williams
Re: routing and pf at 10Gbps
On 11 Feb 2010 at 23:15, Dirk Mast wrote: Daniel Ouellet wrote: On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Best, Daniel http://www.youtube.com/watch?v=DF7MroTLDfU Actually I was hoping that if it is nothing more than sarcasm Henning would give a hint -- I'm old enough to remember earlier generations of i386 architecture where poorly designed graphics card would affect the entire bus performance to slow down all kinds of I/O (disk, lan, etc.)
Re: routing and pf at 10Gbps
On 22:56, Thu 11 Feb 10, Mike Williams wrote: On Thursday 11 February 2010 19:46:22 Henning Brauer wrote: yes, people run firewalls on 10G circuits I am not aware of anyone filtering at 10G who is using off the shelf hardware, with open source O/S. I know of some. Thanks very much guys, much useful information. I'll be sure to purchase servers with the fastest graphics cards possible! We had a Juniper tech in today, and he suggested the SRX-3600. It's almost certainly faster than any single software platform can do (~18Gbps IMIX), but is it worth 10 times the price, and 5U of rack space? Especially for a parallisable problem. Junos can't come close to the scriptable service monitoring and route manipulation that almost trivially easy on openbsd either. Then break it up in segments and use junos to do the heavy routing and use openbsd on the 1Gbps or slower links. -- Michiel van Baak mich...@vanbaak.eu http://michiel.vanbaak.eu GnuPG key: http://pgp.mit.edu:11371/pks/lookup?op=getsearch=0x71C946BD Why is it drug addicts and computer aficionados are both called users?
Re: routing and pf at 10Gbps
On Fri, Feb 12, 2010 at 9:48 AM, System Administrator ad...@bitwise.net wrote: On 11 Feb 2010 at 23:15, Dirk Mast wrote: Daniel Ouellet wrote: On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Best, Daniel http://www.youtube.com/watch?v=DF7MroTLDfU Actually I was hoping that if it is nothing more than sarcasm Henning would give a hint -- I'm old enough to remember earlier generations of i386 architecture where poorly designed graphics card would affect the entire bus performance to slow down all kinds of I/O (disk, lan, etc.) That's why you see very few servers with video cards. Even well-designed cards can rob the system of precious, precious I/O. Same goes for sound cards (which, from what I've heard, used to create havoc by not lowering its IRQ after each request), floppy drives, anything not needed for the system to function basically. -- Aaron Mason - Programmer, open source addict I've taken my software vows - for beta or for worse
В Э Д: Заключение внешнеторговой сделки. Международный контракт купли-продажиzix
17 - 18 QP5P2QP0P;Q 2010P3. PPQP:P2P0. PP0P:P;QQP5P=P8P5 P2P=P5QP=P5QPQP3PP2PP9 QP4P5P;P:P8. PP5P6P4QP=P0QPP4P=QP9 (P2P=P5QP=P5QPQP3PP2QP9) P:PP=QQP0P:Q P:QP?P;P8-P?QPP4P0P6P8 === P!P5PP8P=P0Q P?QP5P4P=P0P7P=P0QP5P= P4P;Q: P2QP5Q QQP0QQP=P8P:PP2 PPP, P?QP5P4P?QP8P=P8PP0QP5P;P5P9, QQP8QQPP2, QQP:PP2PP4P8QP5P;P5P9 P8 QP?P5QP8P0P;P8QQPP2 P2P=P5QP=P5Q P:PP=PPP8QP5QP:PP3P, QP0PPP6P5P=P=PP3P, Q P:QP?PQQP=P-P8PP?PQQP=PP3P PQP4P5P;PP2, P!PPP P8 P;PP3P8QQPP2 PPP. PP5P;Q QP5PP8P=P0QP0: PP7P=P0P:PPP;P5P=P8P5 Q Q QP0P?P0PP8, QQP;PP2P8QPP8, P7P0P:PP=PP4P0QP5P;QQQP2PP P8 P?PQQP4P:PP P?PP4P3PQPP2P:P8 P7P0P:P;QQP5P=P8Q PP5P6P4QP=P0QPP4P=PP9 QP4P5P;P:P8. PP8P=P8PP8P7P0QP8Q PQP8P1PP: P8 P=P0QQQP5P=P8P9 P?QP8 P7P0P:P;QQP5P=P8P8 PP5P6P4QP=P0QPP4P=QQ QP4P5P;PP:. PQPP3QP0PPP0 QP5PP8P=P0QP0 PQQPP4 P=P0 P2P=P5QP=P8P9 QQP=PP:. PP0P:P;QQP5P=P8P5 PP5P6P4QP=P0QPP4P=PP9 (P2P=P5QP=P5QPQP3PP2PP9) QP4P5P;P:P8 P:QP?P;P8-P?QPP4P0P6P8 QPP2P0QPP2 * PP5QP=P8P:P0 P8 Q QP0P?Q P2QQPP4P0 P=P0 P2P=P5QP=P8P9 QQP=PP: (P8P7QQP5P=P8P5 P:PP=Q QP=P:QQQQ QPP2P0QP=QQ QQP=P:PP2, P2QP1PQ P7P0QQP1P5P6P=PP3P P?P0QQP=P5QP0). * PQP0P?Q P7P0P:P;QQP5P=P8Q P2P=P5QP=P5QPQP3PP2PP9 QP4P5P;P:P8 (P0. P?QP5P4P?P5QP5P3PP2PQP=P0Q P?P5QP5P?P8QP:P0 P?P QPP3P;P0QPP2P0P=P8Q P?QPP5P:QP0 P:PP=QQP0P:QP0; P2P8P4Q PQP5QQ; P1. PQPQPP;P5P=P8P5 P?P8QQPP5P=P=PP3P P4PP3PP2PQP0 b P?PP4P?P8QP0P=P8P5 P:PP=QQP0P:QP0). * PP0P:PP=PP4P0QP5P;QP=PP5 QP5P3QP;P8QPP2P0P=P8P5 P2PP?QPQPP2, QP2QP7P0P=P=QQ Q P7P0P:P;QQP5P=P8P5P P2P=P5QP=P5Q P:PP=PPP8QP5QP:PP9 QP4P5P;P:P8. PP5P6P4QP=P0QPP4P=QP9 (P2P=P5QP=P5QPQP3PP2QP9) P:PP=QQP0P:Q P:QP?P;P8-P?QPP4P0P6P8 QPP2P0QPP2 * PPQQP4PP: P?PP4P3PQPP2P:P8 P8 P7P0P:P;QQP5P=P8Q, QQQQP:QQQP0 P8 QPP4P5QP6P0P=P8P5 P2P=P5QP=P5QPQP3PP2PP3P P:PP=QQP0P:QP0. * PP;P0QQP8QP8P:P0QP8Q P:PP=QQP0P:QP=QQ QQP;PP2P8P9: QQP;PP2P8Q PP1QP7P0QP5P;QP=QP5 P8 P4PP?PP;P=P8QP5P;QP=QP5, QP8P?PP2QP5 P8 P8P=P4P8P2P8P4QP0P;QP=QP5. * PP8P4Q P2P=P5QP=P5QPQP3PP2QQ P:PP=QQP0P:QPP2, PPP=QQP0P:QQ B+P:P;P0QQP8QP5QP:P8P5B; P8 QP8P?PP2QP5. PQPP1P5P=P=PQQP8 P1P0QQP5QP=QQ P:PP=QQP0P:QPP2. * PQP8PP5QP=P0Q QQP5PP0 Q P:QP?PQQP=P-P8PP?PQQP=PP3P P:PP=QQP0P:QP0 P:QP?P;P8-P?QPP4P0P6P8 QPP2P0QPP2. P%P0QP0P:QP5QP8QQP8P:P0 PQP=PP2P=QQ P:PP=QQP0P:QP=QQ QQP;PP2P8P9: P:P0QP5QQP2P, QP5P=P0 P8 P1P0P7P8Q P?PQQP0P2P:P8 * P P0P7P4P5P; B+PP0QP5QQP2PB; P2 P2/Q P:PP=QQP0P:QP5. P!P?PQPP1Q PP?QP5P4P5P;P5P=P8Q P:P0QP5QQP2P0 QPP2P0QP0 P2 PP5P6P4QP=P0QPP4P=PP9 QPQP3PP2P;P5. * PP5P=PPP1QP0P7PP2P0P=P8P5 P2P P2P=P5QP=P5P9 QPQP3PP2P;P5; P2P8P4Q (QP?PQPP1Q QP8P:QP0QP8P8) P2P=P5QP=P5QPQP3PP2QQ QP5P=; P2P8P4Q QP:P8P4PP:, P?QP8PP5P=QP5PQQ P2 PP5P6P4QP=P0QPP4P=PP9 QPQP3PP2P;P5. * PP0P7P8QP=QP5 QQP;PP2P8Q P?PQQP0P2P:P8: QPP;Q P8 QQP=P:QP8P8 P2 PP1P5QP?P5QP5P=P8P8 PP5P6P4QP=P0QPP4P=PP9 QPQP3PP2P;P8. * PPQPP0QP8P2P=P-P?QP0P2PP2P0Q P1P0P7P0 (PQP=PP2P=QP5 P4PP:QPP5P=QQ); QP0P7P1PQ PQP=PP2P=QQ PP5P6P4QP=P0QPP4P=QQ P1P0P7P8QPP2 P?PQQP0P2P:P8. P%P0QP0P:QP5QP8QQP8P:P0 PQP=PP2P=QQ P:PP=QQP0P:QP=QQ QQP;PP2P8P9: P?P;P0QP5P6, QP0P=P:QP8P8, P=P5P?QP5PP4PP;P8PP0Q QP8P;P0, P0QP1P8QQP0P6 * P#QP;PP2P8Q P?P;P0QP5P6P0 (P2P0P;QQP=P-QP8P=P0P=QPP2QP5 QQP;PP2P8Q P2P=P5QP=P5QPQP3PP2QQ P:PP=QQP0P:QPP2) * P!P0P=P:QP8P8 P8 QP5P:P;P0PP0QP8P8. P$PQQ-PP0P6PQ P8 PQP1P8QQP0P6. * PP0P8P1PP;P5P5 QP8P?P8QP=QP5 PQP8P1P:P8, P4PP?QQP:P0P5PQP5 P?QP8 P7P0P:P;QQP5P=P8P8 P2/Q P:PP=QQP0P:QP0 QPQQP8P9QP:P8PP8 PQP3P0P=P8P7P0QP8QPP8. * B+PQP8P7P8QP=PP5 QP?QP0P2P;P5P=P8P5B; P2P=P5QP=P5QPQP3PP2QP P:PP=QQP0P:QPP. PQP3P0P=P8P7P0QP8Q P:5QQP5QP:P8Q P?P5QP5P3PP2PQPP2. PQP=PP2Q P4P5P;PP2PP3P Q QP8P:P5QP0 P8 P?QPQPP:PP;P0 * P PP;Q, QQP=P:QP8P8 P8 PQPP1P5P=P=PQQP8 P4P5P;PP2PP3P Q QP8P:P5QP0. * PPP4P3PQPP2P:P0 P8 P?QPP2P5P4P5P=P8P5 P:5QQP5QP:P8Q P?P5QP5P3PP2PQPP2. * PP5QP=P8P:P0 P8 Q QP0P?Q P?QPP2P5P4P5P=P8Q P?P5QP5P3PP2PQPP2. * PP0QP8PP=P0P;QP=P-P?QP8QPP;PP3P8QP5QP:P8P5 PQPP1P5P=P=PQQP8 P2P5P4P5P=P8Q P?P5QP5P3PP2PQPP2. PQQP3P;QP9 QQPP;: P2PP?QPQQ P8 PQP2P5QQ QQP0QQP=P8P:PP2. 17 - 18 QP5P2QP0P;Q 2010 P3PP4P0. (2 P4P=Q Q 10.00 P4P 18.00) PP5QQP P?QPP2P5P4P5P=P8Q: PPQP:P2P0, P. PPQP0P=P8QP5QP:P8P9 QP0P4, P#QP5P1P=P0Q
Re: routing and pf at 10Gbps
On 12 Feb 2010 at 11:44, Aaron Mason wrote: On Fri, Feb 12, 2010 at 9:48 AM, System Administrator ad...@bitwise.net wrote: On 11 Feb 2010 at 23:15, Dirk Mast wrote: Daniel Ouellet wrote: On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Best, Daniel http://www.youtube.com/watch?v=DF7MroTLDfU Actually I was hoping that if it is nothing more than sarcasm Henning would give a hint -- I'm old enough to remember earlier generations of i386 architecture where poorly designed graphics card would affect the entire bus performance to slow down all kinds of I/O (disk, lan, etc.) That's why you see very few servers with video cards. Even well-designed cards can rob the system of precious, precious I/O. Same goes for sound cards (which, from what I've heard, used to create havoc by not lowering its IRQ after each request), floppy drives, anything not needed for the system to function basically. They might not have physical add-in cards, but all i386/amd64 servers have graphics hardware attached to some interconnect bus. Otherwise they would not be able to paste those Microsoft Windows stickers. And on many the only way to turn off the on-board (often inferior) graphics hardware is to insert an add-in card... -- Aaron Mason - Programmer, open source addict I've taken my software vows - for beta or for worse
Clase - Taller: Trastornos de la Alimentación
Escuela Sistimica Argentina presenta: Clase -Taller Trastornos de la Alimentacisn === y Supervisisn de casos clmnicos === Miircoles 17 de Febrerode 18.00 a 19.30 y de 19.30 a 21.00 hs. Coordina: Lic. Graciela Piatti Actividad aranceladaSe entregaran certificados. La reserva de vacantes puede ser realizada vma mail o telefsnicamente. Informes e inscripcisn: Fray J. S. M. Oro 1843 (C1414DBC) Cap. Fed. Tel/ Fax: 4774-2875/6112 - 4899-1053i...@escuelasistemica.com.ar
Re: routing and pf at 10Gbps
On Fri, Jan 22, 2010 at 08:12:29PM +, Mike Williams wrote: Hey all, I was hoping there are some heavy PF users here, who wouldn't mind sharing some of their experiences? So I've watched Hennings talk about PF performance, read the PDF, but I haven't actually seen anyone saying they can, and do, PF at 10Gbps. Can it? If so, what actual hardware can? Or more precisely, what hardware could sustain our expected usage? We've got a big project in it's earliest stages which would require very basic firewalling at multi-gigabit-per-second. Probably in the region of 3Gbps (yes yes, PPS is the real killer), with peaks for software releases much higher. No NAT, just routing (bgpd/ospfd), and simple limits on what ports are available. I can't imagine needing more than 200-300 rules. I'm actually a Linux guy, and I'm pretty confident that netfilter simply won't keep up, and while we've not personally used OpenBSD in anger yet, there is plenty of time to get acquainted. So, at the edges I'm imagining a large hardware router, handing off to OpenBSD to sub-route, VLAN, PF, to the actual servers, and then a few 10s of Mbps of IPSec stuff back to base. The traffic patterns expected are very approximately: 5Mbps DNS 30Mbps of HTTP requests that elicit a sub-500byte response. 200,000,000 hits per day. 300Mbps of normal HTTP. 2-3Gbps of several hundred KB, to many-MB, files over HTTP. 20Mbps of stuff over IPSec. syslog, ssh, snmp, etc. Nearer the core will have much more complex PF rules, but only on a few hundred Mbps, so easy for modest hardware. Performance, cheapness, quality. You should choose only two of these. Do not play with totally-software routers, buy Juniper. -- MINO-RIPE
Re: routing and pf at 10Gbps
Performance, cheapness, quality. You should choose only two of these. Do not play with totally-software routers, buy Juniper. http://praetorianprefect.com/archives/2010/01/juniper-kernel-crash-scapy-code/
Re: routing and pf at 10Gbps
On 12/02/2010, at 11:24 AM, Claudio Jeker wrote: On Thu, Feb 11, 2010 at 03:07:28PM -0500, Daniel Ouellet wrote: On 2/11/10 2:46 PM, Henning Brauer wrote: disk i/o is irrelevant. you will need a very very very fast opengl capable graphics card with loads of memory of course. ??? I am sure I am missing something big here, but Fast Video Card with OpenGL for router? Are you trying to look live every packets routed here? If I may asked Henning, please give me a clue stick as that part I really do not understand what so ever. No bunt intended, I just do not understand that at all, please help me get it? What Video have to do with routing? Henning, I told you, we should not talk about unfinsihed projects. We planned to announce this in exactly 7 weeks. Anyway, to late, the cat is out of the bag. So Henning and Oga are working at offloading pf into the graphic card cores by using the DRI interface. The shader will evaluate the ruleset and packets in parallel and use the graphic memory for the state table. Additionally if the speed of one card is not enough you can use SLI or crossfire to use multiple cards in parallel. -- :wq Claudio It is just a 3-line diff Gosh. They must be hellish lines if it takes 7 weeks to write just three of them ... paulm
Re: routing and pf at 10Gbps
On Thu, Feb 11, 2010 at 10:41:04PM -0800, Bryan Irvine wrote: Performance, cheapness, quality. You should choose only two of these. Do not play with totally-software routers, buy Juniper. http://praetorianprefect.com/archives/2010/01/juniper-kernel-crash-scapy-code/ http://ptresearch.blogspot.com/2010/01/juniper-junos-remote-kernel-crash-flaw.html Software releases built on or after January 28, 2009 have already fixed the issue. Every OS can have it's vulnerabilities. OpenBSD is very secure, but I'm sure it cannot route (even without pf) 10Gbit/s traffic. -- MINO-RIPE