gmane.os.netbsd.general

wi and WPA-PSK

Vlad Markov — 2008-11-29T04:15:06

I flashed a Netgear MA401 Wireless NIC so that I cold use it via my WPA-PSK wireless router on a Windows box. It has been working there nicely for a few months now. I want to use it on my NetBSD box also. From dmesg: wi0 at pcmcia0 function 0: wi0: 802.11 address 00:09:5b:23:3c:0e wi0: using RF:PRISM2.5 MAC:ISL3873B(PCMCIA) wi0: Intersil Firmware: Primary (1.1.1), Station (1.7.4) wi0: 11b rates: 1Mbps 2Mbps 5.5Mbps 11Mbps That looks good to me. from ifconfig: wi0: flags=8863 mtu 1500 ssid utopia nwkey ***** powersave off bssid 00:1e:e5:6f:ad:4d chan 1 address: 00:09:5b:23:3c:0e media: IEEE802.11 autoselect (DS2) status: active inet 0.0.0.0 netmask 0xff000000 broadcast 255.255.255.255 inet6 fe80::209:5bff:fe23:3c0e%wi0 prefixlen 64 scopeid 0x3 Doesn't "nwkey" mean its trying to use WEP? It looks like WPA is not supported by the NetBSD wi driver. Am I correct? If I am correct, is there any chance WPA will be implemented for wi in NetBSD 5.0? Vlad

netbsd 5 & lii0

jimerickso — 2008-11-29T00:54:33

is there an irq conflict when using the lii0 interface on the eee 900 16g in netbsd 5.0? ifconfig shows the interface as being up but with the media showing "none" and the status showing "no carrier". networking is configured properly using dhcp. i can not initialize the interface. shouldn't it at least show a carrier? the cabling, interface, and router are all functional when using freebsd, openbsd and netbsd 4.0.1 on this eee 900 16g. thank you for any help you can offer.

build netbsd5 fail

Water NB — 2008-11-28T17:48:03

I checkout the latest src tonight and build distribution # link rump_cd9660/rump_cd9660 /usr/world/386t5/bin/i386--netbsdelf-gcc -B/usr/world/386o5/destdir.i386/usr/lib -B/usr/world/386o5/destdir.i386/usr/lib/ -Wl,-nostdlib -Wl,--wrap=malloc -Wl,--fatal-warnings -o rump_cd9660 mount_cd9660.o rump_cd9660.o pathadj.o fattr.o -lrumpfs_cd9660 -lp2k -lukfs -lrump -lrumpuser -lpuffs -lutil -lpthread -L/usr/world/386o5/destdir.i386/usr/lib --Wl,-rpath-link,/usr/world/386o5/destdir.i386/lib -L/usr/world/386o5/destdir.i386/lib -Wl,-rpath-link,/usr/world/386o5/destdir.i386/usr/lib -L/usr/world/386o5/destdir.i386/usr/lib /usr/world/386t5/lib/gcc/i386--netbsdelf/4.1.3/../../../../i386--netbsdelf/bin/ld: cannot find -lrumpfs_cd9660 *** Failed target: rump_cd9660

Mozilla Sunbird and TCL-Plugin 3.0

Volker Wolfram — 2008-11-24T15:10:58

Hello, please let me ask a question. I'm interested in the following programs: Mozilla Sunbird in an newer version with german language support and TCL/TK Plugin 3.0 . Is there anyone who can tell me, if is there any work on these programs in the pkgsrc tree? Any comments are very welcome! Thank you. Volker

NetBSD 4.0.1

Rick Kelly — 2008-11-24T15:21:09

Is it possible to upgrade from 3.1 to 4.0.1 without sendmail getting deleted?

firefox on netbsd 3.1

Rick Kelly — 2008-11-24T02:21:03

Has anyone built firefox 3.04 or 2.18 on NetBSD 3.1 i386? Mozilla has stopped working with /. and a few other sites.

Network interface locked up under 5.0_BETA

Matthias Scheler — 2008-11-23T23:13:24

Hello, the ethernet interface of my NetBSD/amd64 5.0_BETA server locked up after a few hours of uptime. The machine was no longer reachable via network and I had to login on the console. Pinging another machine failed with "no buffer space available". But "sysctl" reported that ony about 1100 of 4096 mbuf-s were in use. I had to "ifconfig bge0 down" followed by "ifconfig bge0 up" to get the network working again. Has anybody else seen something similar? Kind regards

listing available 802.11 networks

Steven M. Bellovin — 2008-11-21T16:40:56

What is the preferred way to list available 802.11 networks? Sometimes, I can use 'ifconfig wpi0 list scan'; however, even when it works it seems to drop my association with an AP. Other systems seem to be able to display a list of available networks while associating with one particular one; can NetBSD? --Steve Bellovin, http://www.cs.columbia.edu/~smb

Webcams

Andrew Ball — 2008-11-20T14:51:09

Hello, What "webcams" are NetBSD people using these days? Are there USB 2 or FireWire cameras that work well with NetBSD? - Andy Ball.

building new version of config

rvargh jo — 2008-11-20T10:31:21

I have installed NetBSD 4.0.1 on a X86 box. I tired to build a 5.0 kernel source inside the box . But when i run config i get the following error # config GENERIC ../../../../conf/files:5: your sources require a newer version of config(1) -- please rebuild it. ../../../../ipkdb/files.ipkdb:5: syntax error ../../../../dev/raidframe/files.raidframe:6: syntax error ../../../../conf/files:1194: syntax error ../../../../conf/files:1198: syntax error ../../../../conf/files:1223: syntax error ../../../../conf/files:1225: syntax error ../../../../dev/pad/files.pad:3: syntax error *** Stop. I downloaded the new config.tar.gz from /pub/NetBSD/NetBSD-5_BETA/tar_files/src/ But when I try to make source of config.tar.gz, it complained: mkheaders.c:58:24: error: crc_extern.h: No such file or directory. I am tring to update kernel with current source, so I real need the newer config.

tooling around with the kernel source, had a question about cvs

Michael Litchard — 2008-11-19T21:47:20

What is the cvs command to take a diff of any given change and the original source?

Format/Partition new disk (i386/4.0.1)

Mark Thomas — 2008-11-18T23:24:43

For some reason I can not get this new disk to format/partition correctly, at least it doesn't _look_ correct. I ran fdisk -u then disklabel. What bothers me is why are the fsize, bsize and cpg/sgs all zero's? That can not be correct. Also, does anyone have a simple example on how to format, label and 'newfs' a complete IDE drive? I rarely have to setup a new disk but when I do it is always an issue for me, I just don't get it I guess. I have read the Guide many times over and still get confused every time I read it. thanks for any and all help ================================================== # fdisk wd1 Disk: /dev/rwd1d NetBSD disklabel disk geometry: cylinders: 969021, heads: 16, sectors/track: 63 (1008 sectors/cylinder) total sectors: 976773168 BIOS disk geometry: cylinders: 1023, heads: 255, sectors/track: 63 (16065 sectors/cylinder) total sectors: 976773168 Partition table: 0: NetBSD (sysid 169) start 63, size 976773105 (476940 MB, Cyls 0-60801/80/63) PBR is not bootable: All bytes are identical (0x00) 1: 2: 3: Drive serial number: 129 (0x00000081) # disklabel wd1 # /dev/rwd1d: type: ESDI disk: WDC WD5001ABYS-1 label: fictitious flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 16 sectors/cylinder: 1008 cylinders: 969021 total sectors: 976773168 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # microseconds track-to-track seek: 0 # microseconds drivedata: 0 4 partitions: # size offset fstype [fsize bsize cpg/sgs] a: 976773105 63 4.2BSD 0 0 0 # (Cyl. 0*- 969020) c: 976773105 63 unused 0 0 # (Cyl. 0*- 969020) d: 976773168 0 unused 0 0 # (Cyl. 0 - 969020) wd1 at atabus1 drive 1: wd1: drive supports 16-sector PIO transfers, LBA48 addressing wd1: 465 GB, 969021 cyl, 16 head, 63 sec, 512 bytes/sect x 976773168 sectors wd1: 32-bit data port wd1: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)

tuning squid (maybe off topic)

Al - Image Hosting Services — 2008-11-18T21:32:10

Hi, This may be a little bit off topic for this list, but I figure someone on here would know. I installed squid on my servers (running NetBSD 4). Right now, I am using it as a web accelerator. It seems to work pretty well, although a little bit overloaded. I found some performance tuning how to on the web that made a difference, but I was wondering if anyone had any suggestions? Also, I was wondering if squid was built without arp-acl aufs carp coss diskd icmp ipf-transparent pam-helper pf-transparent snmp ssl unlinkd, if that could help and if someone could give me some insight on what they do? Also, I noticed that squid seems to be compiled to use 8192 file descriptors. This doesn't make sense to me, because that is all the file descriptors that are set on the server. Is there a way to change this value? Best Regards, Al

2 different flash plugin for firefox

Cem Kayali — 2008-11-18T19:01:25

Hello! I would like to enable gnash plugin as default, while enable Adobe flash plugin for certain users's firefox brpwsers... If someone has advise how to do this, i appreciate. I'm using NetBSD-current / amd64. Regards, Cem

gcc predefined labels

David Dudley — 2008-11-17T22:43:28

Does the gcc compiler predefine a label like __netbsd__ to let you know what system you're running on? How do you list the predefined labels? David BEGIN:VCARD VERSION:2.1 X-GWTYPE:USER FN:David Dudley TEL;WORK:826-1265 ORG:;MIS TEL;PREF;FAX:880-3741 EMAIL;WORK;PREF;NGW:DavidDu< at >cctexas.com N:Dudley;David TITLE:Functional Analyst END:VCARD

Please help - dialin-server?

Nino on NetBSD 4.0 — 2008-11-16T19:26:46

Dear NetBSD-users, I would like to configure a dial-in server, and I would like to know 1. what exactly are the configuration files and 2. what exactly should be in there. I want to remotely control my computer without ethernet. Please no RTFM-responses, I have read enough FM in the last three weeks, and I just had no success. I would not ask you if I thought it is something easy to figure out for me. I must admit, I tried it out on Xubuntu 8.04, NOT on NetBSD. But in fact, I want that finally it works on NetBSD. I thought Xubuntu should be easier, however, far from it - I had no success. What I want and what I did: I want to be able to dial in to my NetBSD machine and control it through my Xubuntu or Windows machine. For the sake of simplicity, let us assume the user is "person" and password is "mypass" on both machines. For the connection, two cellphones are used (Sony Ericsson D750i and W200i). In Linux on connection ach one appears as the devices /dev/ttyACM0 and /dev/ttyACM1. On NetBSD, they are recognised on connection as ucom0 and ucom1 as well as umodem0 and umodem1. I tried dialling out to my university (which runs AIX) with each of them, and it works, both in Linux and in Windows. Moreover, I tried dialling from Windows to Linux, on Linux Minicom was running and could automatically take the call when told ATS0=1. ATA worked fine, too. So, what do I write in which file? - At this stage, I do not want routing or anything complicated, I just want a connection between computer A dialling to computer B, there are on no other networks. B takes the call automatically, A then gets to login prompt and shell. A is running Windows, Linux or FreeBSD, B is running NetBSD. What do I have to do? - I want to first ask you, instead of first messing up my configuration files until nothing is recognisable (as I already did on Xubuntu). Thank you in advance for your kind help with this, I very much appreciate it. And yes, I DO know that chapter 23.3. of the NetBSD guide touches the topic. Kind regards, Nino

csup and softdeps

Christian Baer — 2008-11-16T13:26:39

Hello all you out there! Today I tried something rather unorthodox (for NetBSD anyway): I wanted to update a 4.0 to current using csup (as apposed to the usual cvs). Under NetBSD this seams to be a rather unusual method (Is there some reason for this?) but under FreeBSD it is the method of choice. Since I came from FreeBSD, I'm kinda used to the way cvsup (later csup) works. To speed up the whole thing I mounted /usr with softdeps. That however broke (and I have no clue how) csups download. I kept getting protocol errors, no matter what user I tried on the system or what server I used to dl from (I tried cvsup.se.netbsd.org and cvsup.uk.netbsd.org). After mounting /usr without softdeps again, csup works just fine. Does anyone know why this is happening? Note: The hardware is a Sun U60 with 2GB of RAM and two SCA-HDDs. Regards Chris

assessing a wireless link

James K. Lowden — 2008-11-15T21:15:15

Can someone help me interpret netstat -p tcp? I've noticed odd behavior in the wi0 interface in my Soekris box. Web pages loading on my laptop frequently pause at the "same" point: there's a hesitation after the progress bar is about an inch long. Also remote X sometimes feels like it's underwater, running on a 19.2 kbps line. I thought I'd see if I could determine if the wireless link is at fault and, if so, which end is complaining the loudest. Below is the output from "netstat -p tcp -I wi0". I'm not sure how to interpret what I'm seeing, or whether another comand might be more illuminating. My question is, basically, what is the signal/noise ratio? How many packets are being lost, and is that figure good or bad? Could it reasonably be expected to improve using different hardware, or is what I'm seeing par for the 802.11 course? I'm inclined to think the hardware is fine because the ip statistics look good. But if I knew what I was talking about, I wouldnt' be asking here. ;-) Much appreciation for your insight. == snip == $ netstat -p tcp -I wi0 tcp: 150923 packets sent 92968 data packets (25932648 bytes) 1640 data packets (327864 bytes) retransmitted 51397 ack-only packets (56737 delayed) 0 URG only packets 0 window probe packets 52 window update packets 4876 control packets 0 send attempts resulted in self-quench 181240 packets received 104731 acks (for 25621999 bytes) 4833 duplicate acks 0 acks for unsent data 57677 packets (6183672 bytes) received in-sequence 565 completely duplicate packets (38781 bytes) 0 old duplicate packets 1 packet with some dup. data (88 bytes duped) 13733 out-of-order packets (0 bytes) 0 packets (0 bytes) of data after window 0 window probes 313 window update packets 903 packets received after close 1 discarded for bad checksum 0 discarded for bad header offset fields 0 discarded because packet too short 352 connection requests 15304 connection accepts 15532 connections established (including accepts) 15660 connections closed (including 466 drops) 1 embryonic connection dropped 0 delayed frees of tcpcb 80886 segments updated rtt (of 76309 attempts) 2018 retransmit timeouts 82 connections dropped by rexmit timeout 0 persist timeouts (resulting in 0 dropped connections) 0 keepalive timeouts 0 keepalive probes sent 0 connections dropped by keepalive 2003 correct ACK header predictions 10982 correct data packet header predictions 32778 PCB hash misses 143 dropped due to no socket 0 connections drained due to memory shortage 126 PMTUD blackholes detected 1558 bad connection attempts 15327 SYN cache entries added 0 hash collisions 15304 completed 0 aborted (no space to build PCB) 21 timed out 0 dropped due to overflow 0 dropped due to bucket overflow 2 dropped due to RST 0 dropped due to ICMP unreachable 0 delayed free of SYN cache entries 292 SYN,ACKs retransmitted 192 duplicate SYNs received for entries already in the cache 0 SYNs dropped (no route or no space) 0 packets with bad signature 0 packets with good signature $ netstat -I wi0 -p ip ip: 4604045 total packets received 0 bad header checksums 0 with size smaller than minimum 0 with data size < data length 0 with length > max ip packet size 0 with header length < data size 0 with data length < header length 0 with bad options 0 with incorrect version number 0 fragments received 0 fragments dropped (dup or out of space) 0 fragments dropped (out of ipqent) 0 malformed fragments dropped 0 fragments dropped after timeout 0 packets reassembled ok 572856 packets for this host 0 packets for unknown/unsupported protocol 4015491 packets forwarded (0 packets fast forwarded) 4861 packets not forwardable 0 redirects sent 0 packets no matching gif found 551368 packets sent from this host 0 packets sent with fabricated ip header 0 output packets dropped due to no bufs, etc. 0 output packets discarded due to no route 0 output datagrams fragmented 0 fragments created 0 datagrams that can't be fragmented 0 datagrams with bad address in header = pins == --jkl

Problem with CVS under cygwin - file make:

Donald T Hayford — 2008-11-15T15:30:06

If there is a better location for this question, please let me know. Thanks. I'm trying to retrieve NetBSD-current using CVS under cygwin. I use the following command lines: $ export CVS_RSH="ssh" $ export CVSROOT="anoncvs< at >anoncvs.NetBSD.org:/cvsroot" $ cvs checkout -A -P src Everything seems to be running ok, until I get the following error message: cvs [checkout aborted]: cannot make directory src/compat/lib/libm/make:: No such file or directory I think the problem is that the directory in question has a ":" at the end of the name (i.e., it is "src/compat/lib/libm/make:"), which I suppose Windows is objecting to. I scrounged through the web-based CVS repository, and there is, in fact, a directory with that name, though there is nothing in it that I could see. But I don't know how to make cvs skip that directory or to keep on going anyway. Any ideas on how to get around this? Thanks - Don

Xfree --> Xorg cookbook?

Paul Goyette — 2008-11-15T02:08:51

Has anyone written a step-by-step "cookbook" for upgrading X? I'm especially interested in things like: 1. What other files (like man.conf?) need to be updated to point to the new directories. Do I need to do anything special in /etc/rc.conf to make it start Xorg instead of Xfree? 2. What are all the various X config files that need to be moved from the old directories? Are there any new config files that need to be created? 3. Are there any special considerations or things in those config files that need to be editted? 4. Other than /usr/X11R6, what else should I delete when finished with the upgrade? 5. Any "gotchas" to watch out for in users' ~/.xsession files? Since -current is now switched to Xorg by default, I guess it's time for me to upgrade. But it's been so many years since I've actually done an X install, I'm totally certain that I don't even know what (or where) all the various files are. I vaguely remember /etc/X11/XF86config and then there's a file somewhere that describes the xserver and its transports, ... I know that this is not a voyage to "where no man has gone before," but it sure would be nice if one of those pioneers has actually recorded it all in the captain's log! TIA :-) ---------------------------------------------------------------------- | Paul Goyette | PGP DSS Key fingerprint: | E-mail addresses: | | Customer Service | FA29 0E3B 35AF E8AE 6651 | paul< at >whooppee.com | | Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette< at >juniper.net | ----------------------------------------------------------------------

default gateway on different subnet

Petar Bogdanovic — 2008-11-13T22:08:45

Hello, what is the proper NetBSD equivalent for the following two linux route-commands in a situation where one host has not just one network interface but also the problem that the default gateway is not on his subnet: ifconfig eth0 200.2.116.140 netmask 255.255.255.0 route add -net 200.2.117.0 netmask 255.255.255.240 dev eth0 route add default gw 200.2.117.1 My first guess was the following (vr0 = eth0): ifconfig vr0 inet 200.2.116.140 netmask 255.255.255.0 route add -net 200.2.117.0 -netmask 255.255.255.240 -link vr0 -iface route add default 200.2.117.1 but it didn't work. Unfortunately I also can't tell why exactly because the machine is located in a datacenter far away from here and I have to rely on a few notes made by busy technicians. What I do know is that the NetBSD-image always boots up successfully (according to eyewitness account ;). After digging through the archives I also found this, but it didn't work either: ifconfig vr0 inet 200.2.116.140 netmask 255.255.255.0 route add -host 200.2.117.1 -link vr0 -iface route add default -ifa 200.2.116.140 200.2.117.1 (http://mail-index.netbsd.org/netbsd-help/2006/12/15/0000.html) I tried to reproduce the scenario on a similar qemu-setup and found out that adding the other subnet fails with `Network unreachable' while adding a host works as expected (at least that's what `route show' told me). Any help is greatly appreciated, Petar P.S. That's how the linux routing table looks like, when the Debian rescue-image is loaded: # route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 200.2.117.0 0.0.0.0 255.255.255.240 U 0 0 0 eth0 200.2.116.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 0.0.0.0 200.2.117.1 0.0.0.0 UG 0 0 0 eth0