Re: multi-link 802.11b through netgraph yields poor performance.
Also, before blaming netgraph, which may well be to blame, could it be that you have interference from some other source that's making things bad? The exactly every other packet being dropped does seem to be a big clue. I have ruled out interference as a contributor to these results. My next step was to remove netgraph from the equation - I simply set up the two channel/SSID pairs across the two laptops, and assigned one pair addresses 10.10.10.10 and 10.10.10.20 with a /24 netmask and the other pair 192.168.0.129 and 192.168.0.130 with a /25 netmask. Each pair could ping between themselves with 0% packet loss. Therefore, it seems that there are no real technical difficulties in running two pair 802.11b networks between two laptops, using two adaptors in each machine. So, I switched the master/slave order for the cards participating in the multi-link on each laptop. That is, I made wi1 the master and wi0 the slave on one side, and an1 the master and an0 the slave on the other. This was illuminating, as this configuration did not allow any functionality - not even the 50% packet loss. Instead, on the system with two Lucent cards, I got the familiar watchdog timeout errors: /kernel: wi1: wi_write_data device timeout /kernel: wi1: xmit failed /kernel: wi1: watchdog timeout ... /kernel: wi1: failed to allocate 1594 bytes on NIC /kernel: wi1: tx buffer allocation failed Basically the same old watchdog errors that I see quite frequently on Lucent cards. So, the point made earlier in this thread that the behavior I produced looked like the behavior of a ng_one2many setup with 50% of the link not working seems to be correct. However, because the first test I mentioned above (two wireless nets operating simultaneously across two systems without netgraph) succeeded, I am not simply dealing with a bad card or a down link. Instead, I think this may be a firmware issue. Specifically, different revisions of the Lucent firmware behave differently as regards promiscuous mode, etc. Because the card works in all other tests, and because the other Lucent card can successfully master the netgraph ng_one2many, I am tentatively concluding that some Lucent firmwares will not work in a ng_one2many setting. OR maybe it is just general `wi` flakiness - this suspect card produces wi_seek timeouts when given an IBSS to create (`wicontrol -q`) on Toshiba Libretto laptops, but not on other laptops (even other Toshibas). Therefore, it is possible that there is no ng_one2many / firmware problem, but rather, since Lucent cards (seem to) malfunction based on somewhat random and arbitrary conditions, perhaps this is just one more of those random and arbitrary conditions. Comments ? In the meantime I will retry this experiment with Cisco cards exclusively. - John Kozubik - [EMAIL PROTECTED] - http://www.kozubik.com To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: FSCK/current and dump errors
FYI: The error described below is fully fixed by moving from 1.10.2.4 to rev. 1.22 of traverse.c. It is the change from 1.21 - 1.22 which restores the ability to do a backup again. Thanks! Dw. On Sat, 29 Jun 2002 [EMAIL PROTECTED] wrote: Not sure if I should blame current - but see the errors below. I've tried an fsck and an fsck -f from single user mode on each of the affected disks (7 disk, mix of ide/scsi give this). FSCK comes through clean. Prior to running -CURRENT the disks where attached to a 2.0.8 machine; and the dump prior to the upgrade is good (and restore can fully read it). But right now a real dump to tape; or a dump to /dev/null give me the errors below. - Is there a more throurough consistency check I could do ? I've attached one of the SCSU disks to a 4.5 RELEASE machine; and fsck sees no errors; and there dump gives me similar errors. Any ideas, or should I just ignore this as -CURRENT madness :-) Thanks, Dw sendbackup: info BACKUP=/sbin/dump sendbackup: info RECOVER_CMD=/usr/bin/gzip -dc |/sbin/restore -f... - sendbackup: info COMPRESS_SUFFIX=.gz sendbackup: info end | DUMP: Date of this level 0 dump: Sat Jun 29 05:10:43 2002 | DUMP: Date of last level 0 dump: the epoch | DUMP: Dumping /dev/ad2s1f (/local2) to standard output | DUMP: mapping (Pass I) [regular files] | DUMP: mapping (Pass II) [directories] | DUMP: estimated 1874624 tape blocks. | DUMP: dumping (Pass III) [directories] | DUMP: dumping (Pass IV) [regular files] ? DUMP: read error from /dev/ad2s1f: Invalid argument: [block -645840818]: count=-1 ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840818]: count=-1 ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840817]: count=-1 ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840816]: count=-1 ...cut 150 lines... ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840749]: count=-1 ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840748]: count=-1 ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840747]: count=-1 ? DUMP: read error from /dev/ad2s1f: Invalid argument: [block -645840746]: count=-1 ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840746]: count=-1 ...cut 1500 lines... ? DUMP: read error from /dev/ad2s1f: Invalid argument: [sector -645840147]: count=-1 ? DUMP: More than 32 block read errors from 135025408 ? DUMP: This is an unrecoverable error. ? DUMP: fopen on /dev/tty fails: Device not configured | DUMP: The ENTIRE dump is aborted. sendbackup: error [/sbin/dump returned 3] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: dc0: failed to force tx and rx to idle state
On Friday, 5th July 2002, Martin Blapp wrote: This problem still persists. On my Laptop a ACCTON MiniPCI 100Mbit card does make this output. Then I loose my network connection. Only a ifconfig down/up of the interface helps. Are you running -current? I did a quick (but safe) hack in -stable hoping that people would test -current and report back. Nobody did. If you are running -current, I will be able to make the error message go away, but that will not solve any hang problem. The code really has no effect at all except to print a warning on some cards. The Linux driver for these cards has no wait for tx and rx idle at all, and just assumes the effect is instant. In practice this is so. Perhaps I should just remove this loop. If anyone is using -current with PNIC, Davicom, or Accton cards and the dc driver, please speak up. I'd like to squash this tiny bug. Stephen. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: multi-link 802.11b through netgraph yields poor performance.
Sorry to cross post this, I want it in the archives.
[discussion on using mulitilink acrsss wireless cards deleted]
I have done similar, using two IP channels and with mpd as the
one2many
basically, assign real IP addresses to the 4 cards, on 2 separate
10.x.x.x/30 networks then open ksocket mpd nodes for each network, making
2 parallel 'pipes'.
then run mpd using the netgraph link type, and set up Multilink.
Multilink will round-robin forthe links, but it will also stop using a
link htat appears to have failed so you have some
redundancey:
here are my configs for this:
firstly the script that sets up the ksockets.
(Assumes all modules needed are loaded)
#!/bin/sh
# $FreeBSD: src/share/examples/netgraph/udp.tunnel,v 1.1 2000/01/28
00:44:30 archie Exp $
# This script sets up a virtual point-to-point WAN link between
# two subnets, using UDP packets as the ``WAN connection.''
# The two subnets might be non-routable addresses behind a
# firewall.
#
# Here define the local and remote inside networks as well
# as the local and remote outside IP addresses and UDP port
# number that will be used for the tunnel.
#
LOC_EXTERIOR_IP1=10.42.3.3
REM_EXTERIOR_IP1=10.42.5.1
UDP_TUNNEL_PORT1=4028
LOC_EXTERIOR_IP2=10.42.1.3
REM_EXTERIOR_IP2=10.42.4.1
UDP_TUNNEL_PORT2=4029
ngctl shutdown tee1:
ngctl shutdown tee2:
sleep 1
ngctl -f - DONE
mkpeer tee dummy left2right
name dummy tee1
mkpeer tee1: ksocket left inet/dgram/udp
name tee1:left ksock1
#
mkpeer tee dummy2 left2right
name dummy2 tee2
mkpeer tee2: ksocket left inet/dgram/udp
name tee2:left ksock2
DONE
#
# Bind the UDP socket to the local external IP address and port
# Connect the UDP socket to the peer's external IP address and port
#
cat DONE
msg ksock1: bind inet/${LOC_EXTERIOR_IP1}:${UDP_TUNNEL_PORT1}
msg ksock1: connect inet/${REM_EXTERIOR_IP1}:${UDP_TUNNEL_PORT1}
msg ksock2: bind inet/${LOC_EXTERIOR_IP2}:${UDP_TUNNEL_PORT2}
msg ksock2: connect inet/${REM_EXTERIOR_IP2}:${UDP_TUNNEL_PORT2}
DONE
ngctl -f - DONE
msg ksock1: bind inet/${LOC_EXTERIOR_IP1}:${UDP_TUNNEL_PORT1}
msg ksock1: connect inet/${REM_EXTERIOR_IP1}:${UDP_TUNNEL_PORT1}
DONE
sleep 2
ngctl -f - DONE
msg ksock2: bind inet/${LOC_EXTERIOR_IP2}:${UDP_TUNNEL_PORT2}
msg ksock2: connect inet/${REM_EXTERIOR_IP2}:${UDP_TUNNEL_PORT2}
DONE
netstat -finet -n
##end of script
And next the mpd.conf file:
default:
load vpn
vpn:
new -i ng1 vpn tunnel1 tunnel2
set iface disable on-demand
set iface addrs 108.106.78.53 192.168.150.85
set iface idle 0
set iface route 192.168.150.0/24
set ipcp yes vjcomp
set ipcp ranges 108.106.78.53/32 192.168.150.85/32
set bundle enable multilink
set bundle enable round-robin
set link tunnel1
set link yes acfcomp protocomp
set link no pap
set link no chap
set link keep-alive 2 15
set link tunnel2
set link yes acfcomp protocomp
set link no pap
set link no chap
set link keep-alive 2 15
open
and now the mpd.links file
tunnel1:
set link type ng
set ng node tee1:
set ng hook right
tunnel2:
set link type ng
set ng node tee2:
set ng hook right
##
the result of this is two tunnels (which can actually be routed over the
internet and encryted with IPSEC if required) implemented completely
within the kernel, (mpd uses the netgraph ppp node)
that handles packet re-ordering and link failure.
I actually use this to connect 2 sites via different ISPs
so that if one link has a failure, my VPNs are still active.
It is extensible to as many ISPs as I need.. without requiring
any BGP mess. (my sites do not access the internet through these links
just use it as a transport for the VPNs.)
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-hackers in the body of the message
Re: dc0: failed to force tx and rx to idle state
Hi, Are you running -current? I did a quick (but safe) hack in -stable hoping that people would test -current and report back. Nobody did. Yes, I'm running CURRENT. If you are running -current, I will be able to make the error message go away, but that will not solve any hang problem. Exactly. The code really has no effect at all except to print a warning on some cards. The Linux driver for these cards has no wait for tx and rx idle at all, and just assumes the effect is instant. In practice this is so. Perhaps I should just remove this loop. Martin To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Problems getting wireless card working with PCI adaptor
Hi all, Sorry to cross post this, but I'm having problems getting this card working under FreeBSD. It works under Windows XP and I've had reports of it working on RedHat 7.3. I'm trying to get a wireless card going in a desktop machine. The wireless card is a ORiNOCO Wireless LAN PC Card The PCI-PCMCIA controller has a texas instruments chip on it and appears to be made by Elan. This combo works under XP but does not work under FreeBSD 4.6 The pc card worked in the old ISA controller, but I recently had to replace my board and I could not find a motherboard with an ISA slot. When I boot the machine up, I get the following output: pccardd[46]: Card Lucent Technologies(WaveLAN/IEEE) [Version 01.01] [] matched Lucent Technologies (WaveLAN/IEEE) [(null)] [(null)] pccardd[46]: Using IO addr 0x240, size 64 pccardd[46]: Setting config reg at offs 0x3e0 to 0x41, Rest time = 50 ms pccardd[46]: Assigning I/O window 0, start 0x240, size 0x40 flags 0x5 pccardd[46]: Assign wi0, io 0x240-0x27f, mem 0x0, 0 byes, irq 5, flags 0 wi0 at port 0x240-0x27f irq 5 slot 0 on pccard0 wi0: 802.11 address: 00:02:2d:02:a6:13 wi0: using Lucent Technologies, WaveLAN/IEEE wi0: Licent Firmware: Station 7.28.01 pccardd[46]: wi0: Lucent Technlogies (WaveLAN/IEEE) inserted. pccardd[46]: pccardd started This information became available when I set debuglevel in pccard.conf to 4. I am then able to assign and IP address to wi0 and set other options like network, etc using wicontrol. However when I do anything network related (ping, traceroute, etc) I get the following message: wi0: watchdog timeout The settings under XP are as follows: ORiNOCO Wireless LAN PC Card IRQ 5 I/O Range FF40-FF7F Texas Instruments PCI-1211 CardBus Controller (says Elan on card) Memory Range EF004000 - EF004FFF Memory Range FEBFF000 - FEBF Memory Range FABFF000 - FEBFEFFF I/O Range FE00 - FEFF I/O Range FD00 - FDFF IRQ 5 Memory Range 000DF000 - 000D I've tried both compiling the wi driver into the kernel and using it as a kernel module. The problem happens the same. I've recompiled the kernel and I have the following line for my pcic device in my kernel configuration file: device pcic0 at pci? irq 0 port 0x3e0 iomem 0xdf This iomem seems to be one of the ones XP is reporting as being in use for this device. I've also tried having the above line with at isa replacing at pci. I have found that IRQ 5 is used by the onboard usb controller, but even if I disable in the bios I still get this message popping up. Any advice on how to fix this, or even whether or not this card is supported would be MUCH appreciated. I've tried a Belkin controller as well with even less result (couldn't even get the machine to find a pccard port). -- - Wayne Pascoe - http://www.penguinpowered.org.uk/wayne/ If someone eventually manages to bag a B-2, that's a cool US$1bn worth of scrap metal - missiles, on the other hand, are cheap. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
offtopic: low level format of IDE drive.
One of my FreeBSD development boxes had a hernia last week when it lost power while writing to disk. The drive wrote out garbage to a track. I want to reformat the drive, (low level) but the bios doesn't have any support to do this (In the past That is how I did this). The machiine has 1 CD drive and no floppy.. anyone with any ideas as to how one can reformat a hard drive feel free to lend me a clue.. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
: :One of my FreeBSD development boxes had a hernia last week when it lost :power while writing to disk. The drive wrote out garbage to a track. : :I want to reformat the drive, (low level) but the bios doesn't have any :support to do this (In the past That is how I did this). :The machiine has 1 CD drive and no floppy.. : :anyone with any ideas as to how one can reformat a hard drive feel free to :lend me a clue.. Two things: (1) dd if=/dev/zero of=raw_device bs=32k This will force the drive to reassign broken sectors. Run this command twice. If the second run of the command is successful and does not stall (looking at running 'iostat 10' output will tell you whether it stalled) then you are golden. Use the base device for the dd output file, e.g. like '/dev/ad0'. Do not specify a slice or a partition. (2) If the command fails for any reason other then hitting the end of the media, or if it stalls on the second go-around, throw the drive away and buy a new one. If that does work then use fdisk -IB to reinitialize the slice table and disklabel to initialize the disklabel, or use sysinstall to reinitialize the tables. -Matt To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Julian Elischer [EMAIL PROTECTED] wrote: One of my FreeBSD development boxes had a hernia last week when it lost power while writing to disk. The drive wrote out garbage to a track. I want to reformat the drive, (low level) but the bios doesn't have any support to do this (In the past That is how I did this). The machiine has 1 CD drive and no floppy.. anyone with any ideas as to how one can reformat a hard drive feel free to lend me a clue.. Boot from a fixit CD, and use dd to zero out the whole disk, e.g.: dd if=/dev/zero of=/dev/ad0c JN To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Mon, Jul 08, 2002 at 02:08:27PM -0700, Matthew Dillon wrote: ... :power while writing to disk. The drive wrote out garbage to a track. : :I want to reformat the drive, (low level) but the bios doesn't have any for what matters, sometimes I managed to recover the disk by just dd'ing a zero block to the broken sectors using dd oseek=nn where nn is the sector number where read fails. cheers luigi To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
kern/40003: Panic on boot w/4.6 and 4.6-stable (ATA problems)
Hi, Just thought I'd check here, I haven't had much luck on -stable. Is this information helpful or do I need to get my remoted gdb stuff working to give useful info? It would be nice to see any ata issues ironed out before the point release. There still seem to be a decent number of people having the used to work, but now it panics problem... Thanks, Charles -- Forwarded message -- Hello, I filed a PR on this, but I've been having trouble scrounging up parts to build a null-modem cable so I can do a remote GDB trace on it. Since it appears there'll be a 4.6.1, and whatever's broken in ata is still kind of broken, here's what I can get from DDB (panic is before root is mounted): ad0: READ command timeout tag-0 serv-0 - restting ata0: restting devices .. ata0-slave: ATA identify retries exceeded done ad0: 1916MB Maxtor 72004 AP [3893/16/63] at ata0-master BIOSDMA Fatal trap 12: page fualt while in kernel mode fault virtual address = 0x6 fault code = supervisor read, page not present instruction pointer = 0x8:0xc013e865 stack pointer = 0x10:0xc0363fbc code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres1, def32 1, gran 1 processor eflags= interrupt enabled, resume, IOPL = 0 current process = 0 (swapper) interrupt mask = bio trap number = 12 panic: page fault syncing disks... done Uptime: 15s Automatic reboot in 15 seconds - press akey on the console to abort Debugger(manual escape to debugger) Stopped at Debugger+0x34: movb$0,in_Debugger.429 db tr Debugger(c029aa49) at Debugger+0x34 scgetc(c02e0fc0,3,c02d9b60,1,84) at scgetc+0x37e sccngetch(2,c0363e80,c01825e1,c02c9380,c036e90) at sccngetch+0xf3 sccncheckc(c02c9380,c0363e90,c0169790,1186a0,c044ae60) at sccncheckc+0xa cncheckc(186a0) at cncchecckc+0x29 shutdown_panic(0,100) at shutdown_panic+0x34 boot(100,10,c0363f64,c0363ef8,c026239f) at boot+0x314 panic(c029ff8c,c029fa6f,c02e1a00,c02c8de0,0) at panic+0x79 trap_fatal(c0363f,6,c02e1a00,c,0) at trap_fatal+0x32b trap_pfault(c0363f64,0,6,68c040,0) at trap_pfault+0x101 trap(10,c0440010,10,c075b850,0) at trap+0x34f calltrap() at calltrap+0x11 --- trap 0xc, eip = 0xc013e865, esp = 0xc0363fa4, ebp = 0xc0363fbc --- ad_attach(c075b850) at ad_attach+0x5d ata_boot_attach(0) at ata_boot_attach+0x12a run_interrupt_driven_config_hooks(0,360c00,368000,0,c0120fc0) at run_interrupt_driven_config_hooks+ox1a mi_startup(0,0,0,0,0) at mi_startup+0x68 begin() at begin+0x47 db To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Julian Elischer wrote: One of my FreeBSD development boxes had a hernia last week when it lost power while writing to disk. The drive wrote out garbage to a track. I want to reformat the drive, (low level) but the bios doesn't have any support to do this (In the past That is how I did this). The machiine has 1 CD drive and no floppy.. anyone with any ideas as to how one can reformat a hard drive feel free to lend me a clue.. The track needs to be reformatted. Generally, this requires a vendor-specific tool to tunnel commands to the drives firmware. Who is the drive manufacturer? In general, Wester Digital provides these tools in the technical support section of its web site. I don't know about other vendors, but I would expect them to provide them as well, since this must be supported by the firmware to get the disk low level formatted in the first place. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
this is not a 'reformat' what I want to do is an old-fashionned refomat/verify where the controller writes new track headers etc. On Mon, 8 Jul 2002, John Nielsen wrote: Julian Elischer [EMAIL PROTECTED] wrote: One of my FreeBSD development boxes had a hernia last week when it lost power while writing to disk. The drive wrote out garbage to a track. I want to reformat the drive, (low level) but the bios doesn't have any support to do this (In the past That is how I did this). The machiine has 1 CD drive and no floppy.. anyone with any ideas as to how one can reformat a hard drive feel free to lend me a clue.. Boot from a fixit CD, and use dd to zero out the whole disk, e.g.: dd if=/dev/zero of=/dev/ad0c JN To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Luigi Rizzo wrote: On Mon, Jul 08, 2002 at 02:08:27PM -0700, Matthew Dillon wrote: ... :power while writing to disk. The drive wrote out garbage to a track. : :I want to reformat the drive, (low level) but the bios doesn't have any for what matters, sometimes I managed to recover the disk by just dd'ing a zero block to the broken sectors using dd oseek=nn where nn is the sector number where read fails. If you have a power failure during writing, you can actually screw up the low level format of the disk. It sounds like the disk you dd'ed only had the high level format screwed up. Julian got struck by lightning; perhaps he will now stick to disks with built-in lightning rods (e.g. not succeptible to this failure), e.g. SCSI. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
this is not a 'reformat' what I want to do is an old-fashionned refomat/verify where the controller writes new track headers etc. You want to follow terry's advice then. Ken To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Julian Elischer wrote: One of my FreeBSD development boxes had a hernia last week when it lost power while writing to disk. The drive wrote out garbage to a track. I want to reformat the drive, (low level) but the bios doesn't have any support to do this (In the past That is how I did this). The machiine has 1 CD drive and no floppy.. anyone with any ideas as to how one can reformat a hard drive feel free to lend me a clue.. All of the manufacturers have a program that will do that. Many of them even produce a bootable floppy. Check their support web page. Kent To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message . -- Kent Stewart Richland, WA http://users.owt.com/kstewart/index.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Montag, 8. Juli 2002 23:46, Kent Stewart wrote: Julian Elischer wrote: The machiine has 1 CD drive and no floppy.. ^^ All of the manufacturers have a program that will do that. Many of them even produce a bootable floppy. Check their support web page. which leaves the no floppy problem.. Perhaps you could use the bootable floppy from the manufacturer as the el torito boot image for a CD-R you create on another machine.. Robert To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Mon, 8 Jul 2002, Terry Lambert wrote: Julian got struck by lightning; perhaps he will now stick to disks with built-in lightning rods (e.g. not succeptible to this failure), e.g. SCSI. This is an urban ledgend.. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Multi CDR burn
I hate to bother the list, but after lots of searching I haven't found this out - maybe I missed something obvious : I have 8 scsi CDRs in a netserver 5 case, need to burn a number of CDs. I haven't seen anything in the ports that will do multi CDRs. So does anyone know of a program. On the other hand, if I just mirror the CDRs would that do the trick? Thank you, Keith To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Julian Elischer wrote: On Mon, 8 Jul 2002, Terry Lambert wrote: Julian got struck by lightning; perhaps he will now stick to disks with built-in lightning rods (e.g. not succeptible to this failure), e.g. SCSI. This is an urban ledgend.. That SCSI disks don't use random inter-record gap placement like IDE does, particularly on the multimedia drives that ignore the need for thermal recalibration, so as to not make video capture jittery? Or that SCSI II directly supports low level track formatting, without requiring a magic tool? 8-) 8-). -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Mon, Jul 08, 2002 at 02:53:20PM -0700, Julian Elischer wrote: On Mon, 8 Jul 2002, Terry Lambert wrote: Julian got struck by lightning; perhaps he will now stick to disks with built-in lightning rods (e.g. not succeptible to this failure), e.g. SCSI. This is an urban ledgend.. No - it's SCSI Specs. A SCSI Disk is required to savely finish the started sector even on powerloss. If all drives fullfill this requirement is another story. -- B.Walter COSMO-Project http://www.cosmo-project.de [EMAIL PROTECTED] Usergroup [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Mon, 8 Jul 2002, Julian Elischer wrote: On Mon, 8 Jul 2002, Terry Lambert wrote: Julian got struck by lightning; perhaps he will now stick to disks with built-in lightning rods (e.g. not succeptible to this failure), e.g. SCSI. This is an urban ledgend.. Which part? :) -- We have known freedom's price. We have shown freedom's power. And in this great conflict, ... we will see freedom's victory. - George W. Bush, President of the United States State of the Union, January 28, 2002 Do YOU Yahoo!? To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Robert Klein wrote: On Montag, 8. Juli 2002 23:46, Kent Stewart wrote: Julian Elischer wrote: The machiine has 1 CD drive and no floppy.. ^^ All of the manufacturers have a program that will do that. Many of them even produce a bootable floppy. Check their support web page. which leaves the no floppy problem.. Perhaps you could use the bootable floppy from the manufacturer as the el torito boot image for a CD-R you create on another machine.. So mount it temporarily on another machine. You don't even need them in the assigned bios tables. When I first started using FreeBSD, I tried the dangerous option and the system would not get past checking the initial hardware. I had to change the HD assignment to none and ll format the disk. Kent Robert To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message . -- Kent Stewart Richland, WA http://users.owt.com/kstewart/index.html To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
One of my FreeBSD development boxes had a hernia last week when it lost power while writing to disk. The drive wrote out garbage to a track. I want to reformat the drive, (low level) but the bios doesn't have any support to do this (In the past That is how I did this). The machiine has 1 CD drive and no floppy.. anyone with any ideas as to how one can reformat a hard drive feel free to lend me a clue.. Modern drives have low level formatting done at the factory due to drives having multiple zones (different sectors/track in each zone) and other horrible things done to sqeeze out many more bits of storage. They even retired the FORMAT opcode from ATA standard! I think your best bet may be to see if you can find a windows program that will reinit the disk. Your disk's vendor may provide such a utility, usually mislabelled DiscWizard or something for free. I am ready for Millipede :-) To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Request for submissions: FreeBSD Bi-Monthly Development Status Report
This is a solicitation for submissions for the May 2002 - June 2002 FreeBSD Bi-Monthly Development Status Report. All submissions are due by July 19, 2002. Submissions should be made by filling out the template found at: http://www.FreeBSD.org/news/status/report-sample.xml Submissions must then be e-mailed to the following address: [EMAIL PROTECTED] For automatic processing. Reports must be submitted in the XML format described, or they will be silently dropped. Submissions made to other e-mail addresses will be ignored. Status reports should be submitted once per project, although project developers may choose to submit additional reports on specific sub-projects of substantial size. Status reports are typically one or two short paragraphs, but the text may be up to 20 lines in length. Submissions are welcome on a variety of topics relating to FreeBSD, including development, documentation, advocacy, and development processes. Prior status reports may be viewed at: http://www.FreeBSD.org/news/status/ Robert N M Watson FreeBSD Core Team, TrustedBSD Projects [EMAIL PROTECTED] Network Associates Laboratories To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Julian Elischer wrote: this is not a 'reformat' what I want to do is an old-fashionned refomat/verify where the controller writes new track headers etc. The thing is, just about all IDE drives more than a few GB or so do 'track writing' and have no fixed sectoring or sector positioning. ie: each time you write a single sector to a track, it does a read-modify-write of *THE ENTIRE TRACK*. This is why we have to have write caching turned on for IDE drives to get decent performance. Without it, it essentially rewrites the entire track over and over and over again because it cannot fill its write buffer in order to write a contiguous block to completely replace what was there before. ie: each track is one giant physical sector with multiple logical sectors inside it. The really annoying thing is that most newer scsi drives do this too. The sad thing is that this makes softdep almost completely useless, because the basic assumption is that sectors that were not explicitly written to will not be touched. The problem is that this isn't the case, even with write caching turned off. Writing a single sector causes the drive to completely rebuild the track and all the sectors on it... in a different relative postition to what was there before. The resulting power off midwrite can cause an absolute mess in sectors *adjacent* to where soft updates was carefully writing to. This means that the 'power off failsafe' file system idea isn't possible with these drives. The only thing that can deal with this sort of failure mode is being willing to resort to 'newfs and restore' or a log structured file system (can you say LFS?). Get a UPS if you value the data. :-] Back to the topic for a moment.. In theory, dd if=/dev/zero of=disk bs=64k is as good as it gets for a low level format, on these drive. With write caching turned on, you are causing every single bit on the disk to be written to, including the metadata. And the dd if=/dev/disk of=/dev/null is the read verify. Some drives can have write verify turned on (I know of certain maxtor models) where the drive will read back the data and rewrite the entire track if necessary. Take the above with a grain of salt, I've never actually worked at a drive manufacturer. The only thing for sure is that all hard drives suck. :-) Cheers, -Peter -- Peter Wemm - [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] All of this is for nothing if we don't go to the stars - JMS/B5 To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
: Julian got struck by lightning; perhaps he will now stick to disks : with built-in lightning rods (e.g. not succeptible to this failure), : e.g. SCSI. : : This is an urban ledgend.. : :No - it's SCSI Specs. :A SCSI Disk is required to savely finish the started sector even :on powerloss. :If all drives fullfill this requirement is another story. : :-- :B.Walter COSMO-Project http://www.cosmo-project.de No, it's an urban legend. Someone actually buttonholed a Seagate engineer a couple of years back and he said with absolute certainty that a Seagate drive would lose up to two sectors, but not more then that. I've had direct experience with this. Seagate drives will indeed lose up to two sectors if you are writing during a power loss... and this is *GOOD* for the industry. Quantum drives (more direct experience on my part) have been known to lose whole tracks and even multiple tracks. -Matt To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Mon, Jul 08, 2002 at 04:40:04PM -0700, Matthew Dillon wrote: : Julian got struck by lightning; perhaps he will now stick to disks : with built-in lightning rods (e.g. not succeptible to this failure), : e.g. SCSI. : : This is an urban ledgend.. : :No - it's SCSI Specs. :A SCSI Disk is required to savely finish the started sector even :on powerloss. :If all drives fullfill this requirement is another story. : :-- :B.Walter COSMO-Project http://www.cosmo-project.de No, it's an urban legend. Someone actually buttonholed a Seagate engineer a couple of years back and he said with absolute certainty that a Seagate drive would lose up to two sectors, but not more then that. After searching I've found the text I remembered in a german book describing SCSI :( But I have not found it in the ansi docs. Seems like I was just wrong and this is indeed a legend. Sorry for the missleading statement. I've had direct experience with this. Seagate drives will indeed lose up to two sectors if you are writing during a power loss... and this is *GOOD* for the industry. Quantum drives (more direct experience on my part) have been known to lose whole tracks and even multiple tracks. Not loosing unrelated data realy is a big difference. -- B.Walter COSMO-Project http://www.cosmo-project.de [EMAIL PROTECTED] Usergroup [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Mon, 8 Jul 2002, Peter Wemm wrote: The thing is, just about all IDE drives more than a few GB or so do 'track writing' and have no fixed sectoring or sector positioning. ie: each time you write a single sector to a track, it does a read-modify-write of *THE ENTIRE TRACK*. This is why we have to have write caching turned on for IDE The sad thing is that this makes softdep almost completely useless, because the basic assumption is that sectors that were not explicitly written to will not be touched. The problem is that this isn't the case, even with write caching turned off. Writing a single sector causes the drive to completely rebuild the track and all the sectors on it... in a different So, this basically means that even a journalling filesystem wouldn't be much safer... how about battery backed up controllers - would those provide protection? (I suspect not, but maybe they're more sophisticated than I thought.) Mike Silby Silbersack To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Monday, 8 July 2002 at 14:46:29 -0700, Kent Stewart wrote: Julian Elischer wrote: One of my FreeBSD development boxes had a hernia last week when it lost power while writing to disk. The drive wrote out garbage to a track. I want to reformat the drive, (low level) but the bios doesn't have any support to do this (In the past That is how I did this). The machiine has 1 CD drive and no floppy.. anyone with any ideas as to how one can reformat a hard drive feel free to lend me a clue.. I had this happen to me during some power fail testing about 18 months ago with an IBM IDE disk (forget the model number). On one such power fail, I lost something like 200 sectors. All of the manufacturers have a program that will do that. Many of them even produce a bootable floppy. Check their support web page. I went looking for format utilities and didn't find anything. Finally I stuck the disk in an old 486 with a format utility in the BIOS, and that worked (fortunately the damage was below the 504 MB boundary :-). While looking at these format programs, I gained the distinct impression that they didn't really format. The description was too vague to make it clear just what they did do, though. Quite possibly it's the same as dd if=/dev/zero, and it just relocates the logical sectors. Greg -- See complete headers for address and phone numbers To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
[hackers] Re: multi-link 802.11b through netgraph yields poorperformance.
John == John Kozubik [EMAIL PROTECTED] writes: Also, before blaming netgraph, which may well be to blame, could it be that you have interference from some other source that's making things bad? The exactly every other packet being dropped does seem to be a big clue. John I have ruled out interference as a contributor to these results. John So, the point made earlier in this thread that the behavior I John produced looked like the behavior of a ng_one2many setup with John 50% of the link not working seems to be correct. However, John because the first test I mentioned above (two wireless nets John operating simultaneously across two systems without netgraph) John succeeded, I am not simply dealing with a bad card or a down John link. Naw (I know this one). It's the same reason that netgraph bridging doesn't work on wireless cards. The firmware on your card prevents you from sending a mac address that is not your own as the source mac address. This is an attempt by the wireless chipset cabal to prevent you from building your own access point. (You can't be a bridge if you can't transmit arbitrary mac addresses) For most cards, there's no workaround. For a few cards, it's rhumored that you can hack the firmware (DLink is one I've heard mentioned). If you hacked netgraph to set the source mac address to the source mac address of each card, everything would likely work. On second thought, tho, you might need to hack arp (et. al.) to do the right thing (or only have arp repsonses go out the primary interface). Dave. -- |David Gilbert, Velocet Communications. | Two things can only be | |Mail: [EMAIL PROTECTED] | equal if and only if they | |http://daveg.ca | are precisely opposite. | =GLO To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Julian Elischer wrote: According to the specs I had access to at Whistle they were pretty much the same low level device with different interface logic. The ATA drives I have seen had a format capacity just like their scsi cousins, just hard to find. Actually, if you read through the thread, you will see the format track was taken out of the ATA specification. You have to know a magic incantation to talk to the drive firmware (hence my vendor tool recomendation, and my question Who is the manufacturer?). -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: swap huge mem systems
On Mon, Jul 08, 2002 at 09:30:04PM -0400, Chuck Robey wrote: Nowadays, what with the price of fast memory at such low levels, I'm buying more memory than I really need, just because it's *so* cheap, the price has gone up before, and it's possible (maybe likely) that next year's popular new app will need the memory. I'm probably not alone in doing this. It's causing me to wonder about how much swap to allocate. I used to follow the rule that I dedicate twice as much disk memory to swap as I have RAM. Now, with my new system, I'm getting a gig of RAM, but it seems ridiculous to dedicate 2G of disk to swap. Under these conditions, what's the real bottom limit (if you have one gig of RAM) for how much swap you can get away with? One Gig? Less? Minimal amount of swap possible: No swap at all of course. Minimal swap if you want to be able to catch core dumps: Physical RAM size + 64K Minimal amount of swap you need: Depends on what you are doing, doesn't it? You don't need to configure any swap at all if you think your RAM is going to be large enough for everything you do. OTOH, considering how cheap disk space is these days, why worry about a gigabyte or two? Some day you might be running some extremly memory hungry application(s) and then you might neeed that extra swap. Running out of swap is No Fun, and should be avoided if practical. Personally, in your situation I would probably configure enough swap to be able to catch a core dump and not much more, i.e. slightly more than 1G swap. -- Insert your favourite quote here. Erik Trulsson [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: [hackers] Re: multi-link 802.11b through netgraph yields poorperformance.
For most cards, there's no workaround. For a few cards, it's rhumored that you can hack the firmware (DLink is one I've heard mentioned). Thank you. I have heard of and witnessed these problems in the past with Lucent cards. Do you know if Cisco Aironet cards exhibit the same behavior ? I am considering conducting this experiment again with only `an` cards. It would be very nice if there were a firmware tool for these cards for FreeBSD, however I have heard that is a non-trivial project. Comments ? - John Kozubik - [EMAIL PROTECTED] - http://www.kozubik.com To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: swap huge mem systems
On Tue, 9 Jul 2002, Erik Trulsson wrote: On Mon, Jul 08, 2002 at 09:30:04PM -0400, Chuck Robey wrote: Nowadays, what with the price of fast memory at such low levels, I'm buying more memory than I really need, just because it's *so* cheap, the price has gone up before, and it's possible (maybe likely) that next year's popular new app will need the memory. I'm probably not alone in doing this. It's causing me to wonder about how much swap to allocate. I used to follow the rule that I dedicate twice as much disk memory to swap as I have RAM. Now, with my new system, I'm getting a gig of RAM, but it seems ridiculous to dedicate 2G of disk to swap. Under these conditions, what's the real bottom limit (if you have one gig of RAM) for how much swap you can get away with? One Gig? Less? Minimal amount of swap possible: No swap at all of course. Minimal swap if you want to be able to catch core dumps: Physical RAM size + 64K Minimal amount of swap you need: Depends on what you are doing, doesn't it? You don't need to configure any swap at all if you think your RAM is going to be large enough for everything you do. OTOH, considering how cheap disk space is these days, why worry about a gigabyte or two? Some day you might be running some extremly memory hungry application(s) and then you might neeed that extra swap. Running out of swap is No Fun, and should be avoided if practical. Personally, in your situation I would probably configure enough swap to be able to catch a core dump and not much more, i.e. slightly more than 1G swap. Probably a good compromise, it just feels silly to go for a G of swap when I will probably never use more than 256M (and that not very often). I mostly compile, edit, and web-browse. Thanks for the confirmation (I suspected this answer, but I feel better now about it). Chuck Robey | Interests include C Java programming, FreeBSD, [EMAIL PROTECTED] | electronics, communications, and signal processing. New Year's Resolution: I will not sphroxify gullible people into looking up fictitious words in the dictionary. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: swap huge mem systems
On Monday 08 July 2002 09:30 pm, Chuck Robey wrote: | Nowadays, what with the price of fast memory at such low levels, I'm | buying more memory than I really need, just because it's *so* cheap, the | price has gone up before, and it's possible (maybe likely) that next | year's popular new app will need the memory. I'm probably not alone in | doing this. It's causing me to wonder about how much swap to allocate. | | I used to follow the rule that I dedicate twice as much disk memory to | swap as I have RAM. Now, with my new system, I'm getting a gig of RAM, | but it seems ridiculous to dedicate 2G of disk to swap. Under these | conditions, what's the real bottom limit (if you have one gig of RAM) for | how much swap you can get away with? One Gig? Less? The bottom limit is zero. I ran with zero swap for a while; the only problem is that if an app goes nuts and starts allocating unlimited memory it gets *all* the memory before you can possibly intervene, so now I allocate some swap space just so that I can see problems in xosview before they happen. If any swap is ever allocated, then I know something's wrong and I have time to intervene before the system is completely locked up. I allocate 256M of swap. In fact, I think that a pretty good formula for a workstation is probably swap = MIN(2*RAM, 256M) unless you have really massive applications for multiple users or something. The only big drawback that I know of with this scheme is that if your system panics you can't get a kmem dump because there's not enough space to hold it. | | --- |- Chuck Robey | Interests include C Java programming, FreeBSD, | [EMAIL PROTECTED] | electronics, communications, and signal processing. | | New Year's Resolution: I will not sphroxify gullible people into looking | up fictitious words in the dictionary. | --- |- | | | To Unsubscribe: send mail to [EMAIL PROTECTED] | with unsubscribe freebsd-hackers in the body of the message -- Brian, the man from Babble-On . . . . [EMAIL PROTECTED] (personal) http://www.babbleon.org http://www.eff.org http://www.programming-freedom.org To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
RE: offtopic: low level format of IDE drive.
Howdy, -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Mike Silbersack Sent: Tuesday, 9 July 2002 10:53 To: Peter Wemm Cc: Julian Elischer; John Nielsen; [EMAIL PROTECTED] Subject: Re: offtopic: low level format of IDE drive. [snip] So, this basically means that even a journalling filesystem wouldn't be much safer... how about battery backed up controllers - would those provide protection? (I suspect not, but maybe they're more sophisticated than I thought.) That's right - a journalled filesystem doesn't help. Nor does battery backed up controllers. If the drive has come back and told the controller that it has written the data, then the controller - battery backed up or not - will mark those sectors as written. If it's a caching controller, then the cache entries for those sectors will be returned to the free list pool. The only way a journalled filesystem would help is if during replay, it checked that all the sectors matched prior to the checkpoint; ie write out all the sectors after the last checkpoint, then check sectors prior to the last checkpoint. That way, the journalled filesystem would know that the data had been written correctly. Does anyone have a detailed list of which SCSI drives do track writes rather than sector writes? Regards, Chris Knight Systems Administrator AIMS Independent Computer Professionals Tel: +61 3 6334 6664 Fax: +61 3 6331 7032 Mob: +61 419 528 795 Web: http://www.aims.com.au To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Matthew Dillon wrote: I've had direct experience with this. Seagate drives will indeed lose up to two sectors if you are writing during a power loss... and this is *GOOD* for the industry. You went to Men In Black II this weekend, didn't you? Do you have Conspiracy Theorist on your business cards? 8-). -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Chris Knight wrote: Does anyone have a detailed list of which SCSI drives do track writes rather than sector writes? All the broken ones. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: swap huge mem systems
Chuck Robey wrote: Nowadays, what with the price of fast memory at such low levels, I'm buying more memory than I really need, just because it's *so* cheap, the price has gone up before, and it's possible (maybe likely) that next year's popular new app will need the memory. I'm probably not alone in doing this. It's causing me to wonder about how much swap to allocate. I used to follow the rule that I dedicate twice as much disk memory to swap as I have RAM. Now, with my new system, I'm getting a gig of RAM, but it seems ridiculous to dedicate 2G of disk to swap. Under these conditions, what's the real bottom limit (if you have one gig of RAM) for how much swap you can get away with? One Gig? Less? Always have enough space reserved so that you can take a crashdump. You will be regretting it someday if you do not. Allowing for memory upgrades is a factor here too. Cheers, -Peter -- Peter Wemm - [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] All of this is for nothing if we don't go to the stars - JMS/B5 To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
On Mon, 8 Jul 2002, Terry Lambert wrote: Does anyone have a detailed list of which SCSI drives do track writes rather than sector writes? All the broken ones. Could you be a little more specific? :-) -- Kris Kirby, KE4AHR | TGIFreeBSD... 'Nuff said. [EMAIL PROTECTED] | IM: KrisBSD | HSV, AL. --- Fate, it seems, is not without a sense of irony. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Chris Knight wrote: Howdy, -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Mike Silbersack Sent: Tuesday, 9 July 2002 10:53 To: Peter Wemm Cc: Julian Elischer; John Nielsen; [EMAIL PROTECTED] Subject: Re: offtopic: low level format of IDE drive. [snip] So, this basically means that even a journalling filesystem wouldn't be much safer... how about battery backed up controllers - would those provide protection? (I suspect not, but maybe they're more sophisticated than I thought.) That's right - a journalled filesystem doesn't help. Nor does battery backed up controllers. If the drive has come back and told the controller that it has written the data, then the controller - battery backed up or not - will mark those sectors as written. If it's a caching controller, then the cache entries for those sectors will be returned to the free list pool. The only way a journalled filesystem would help is if during replay, it checked that all the sectors matched prior to the checkpoint; ie write out all the sectors after the last checkpoint, then check sectors prior to the last checkpoint. That way, the journalled filesystem would know that the data had been written correctly. Yes. Journalled filesystems are just normal file systems with a journal of recent *intentional* modifications to specific sectors. It cannot save you from modifications to *other* sectors as a side effect of writing. Does anyone have a detailed list of which SCSI drives do track writes rather than sector writes? You can find out by turning write caching on and off. camcontrol modepage daN. You want -m 8, the WCE bit. (write cache enable). I do not remember which -P args you need. If you see a HUGE difference in writing smallish blocks to disk between WCE on vs off, then you have a track write drive. A true sectored drive would have much less of a slowdown. I do not have a comparable set handy to get a better idea of what to expect. Really small writes cost scsi overhead though, so that adds to the slowdown. If I was to take a best guiess, I would expect a factor 10+ slowdown for track-write drives on 4K blocksize writes, vs factor 2-5 slowdown for a sectored drive. This is the slowdown factor when turning WCE off. Cheers, -Peter -- Peter Wemm - [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] All of this is for nothing if we don't go to the stars - JMS/B5 To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: swap huge mem systems
Erik Trulsson wrote: Minimal amount of swap possible: No swap at all of course. Minimal swap if you want to be able to catch core dumps: Physical RAM size + 64K Minimal amount of swap you need: Depends on what you are doing, doesn't it? You don't need to configure any swap at all if you think your RAM is going to be large enough for everything you do. Crash dumps good. Zero swap bad. Systems with zero swap lock up tight when they run out of memory; even 1M of swap makes this not happen. There appears to be a bug. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
Kris Kirby wrote: On Mon, 8 Jul 2002, Terry Lambert wrote: Does anyone have a detailed list of which SCSI drives do track writes rather than sector writes? All the broken ones. Could you be a little more specific? :-) Hard drive selection has always been an exclusion set, not an inclusion set. Annoying as that may be, it's no less true. Personally I like Hitachi hard drives; now that they are buying the DeathStar line from IBM, though... 8-(. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
RE: offtopic: low level format of IDE drive.
Howdy, -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Peter Wemm Sent: Tuesday, 9 July 2002 12:44 To: [EMAIL PROTECTED] Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] Subject: Re: offtopic: low level format of IDE drive. [snip] Does anyone have a detailed list of which SCSI drives do track writes rather than sector writes? You can find out by turning write caching on and off. camcontrol modepage daN. You want -m 8, the WCE bit. (write cache enable). I do not remember which -P args you need. If you see a HUGE difference in writing smallish blocks to disk between WCE on vs off, then you have a track write drive. A true sectored drive would have much less of a slowdown. I do not have a comparable set handy to get a better idea of what to expect. Really small writes cost scsi overhead though, so that adds to the slowdown. If I was to take a best guiess, I would expect a factor 10+ slowdown for track-write drives on 4K blocksize writes, vs factor 2-5 slowdown for a sectored drive. This is the slowdown factor when turning WCE off. Excellent. This information should prove invaluable for future drive testing and acceptance. Thanks. Cheers, -Peter Regards, Chris Knight Systems Administrator AIMS Independent Computer Professionals Tel: +61 3 6334 6664 Fax: +61 3 6331 7032 Mob: +61 419 528 795 Web: http://www.aims.com.au To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: [hackers] Re: multi-link 802.11b through netgraph yields poorperformance.
In message: [EMAIL PROTECTED] David Gilbert [EMAIL PROTECTED] writes: : For most cards, there's no workaround. For a few cards, it's rhumored : that you can hack the firmware (DLink is one I've heard mentioned). All Prism 2 and 2.5 (and now 3) based cards can do this. At least with intersil's firmware. Support for it is already in FreeBSD. Warner To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: Problems getting wireless card working with PCI adaptor
In message: [EMAIL PROTECTED] Wayne Pascoe [EMAIL PROTECTED] writes: : I'm trying to get a wireless card going in a desktop machine. The : wireless card is a ORiNOCO Wireless LAN PC Card The PCI-PCMCIA ... : wi0: watchdog timeout Missing interrupts. : I have found that IRQ 5 is used by the onboard usb controller, but : even if I disable in the bios I still get this message popping up. Using a PCI expantion card requires you to use PCI routing. Can you post the complete dmesg? I need to know how we're setting things up and what you've posted so far isn't sufficient. : Any advice on how to fix this, or even whether or not this card is : supported would be MUCH appreciated. I've tried a Belkin controller as : well with even less result (couldn't even get the machine to find a : pccard port). Card is supported. Warner To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: [hackers] Re: multi-link 802.11b through netgraph yields poorperformance.
John == John Kozubik [EMAIL PROTECTED] writes: For most cards, there's no workaround. For a few cards, it's rhumored that you can hack the firmware (DLink is one I've heard mentioned). John Thank you. I have heard of and witnessed these problems in the John past with Lucent cards. Do you know if Cisco Aironet cards John exhibit the same behavior ? I am considering conducting this John experiment again with only `an` cards. John It would be very nice if there were a firmware tool for these John cards for FreeBSD, however I have heard that is a non-trivial John project. Comments ? Other than the 'wi' cards I have, I havn't found one that works yet. I have heard that some 'an' cards support sending arbitrary mac addresses. Dave. -- |David Gilbert, Velocet Communications. | Two things can only be | |Mail: [EMAIL PROTECTED] | equal if and only if they | |http://daveg.ca | are precisely opposite. | =GLO To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Pictausch??
Nachricht von Tina weitergeleitet von uns: Unser Funchat ist wieder online kannste ja mal reinschauen ich bin auch immer drin. ok bis später deine TINA To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: swap huge mem systems
Thus spake Terry Lambert [EMAIL PROTECTED]: Erik Trulsson wrote: Minimal swap if you want to be able to catch core dumps: Physical RAM size + 64K I've caught many core dumps with swap == RAM. Am I just getting lucky, or am I losing 64K of the image? Crash dumps good. I beg to differ. ;-) To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: offtopic: low level format of IDE drive.
In message [EMAIL PROTECTED], Terry Lambert writes: Julian Elischer wrote: According to the specs I had access to at Whistle they were pretty much the same low level device with different interface logic. The ATA drives I have seen had a format capacity just like their scsi cousins, just hard to find. Actually, if you read through the thread, you will see the format track was taken out of the ATA specification. You have to know a magic incantation to talk to the drive firmware (hence my vendor tool recomendation, and my question Who is the manufacturer?). That is understandable. For example, Western Digital has not supported a real format track command since as far back as 1991. The WDAC-280 documentation states that the drive supported a logical format command that would fill existing sectors with zeros. It also supported marking sectors bad to provide backward compatibility with older MFM drives. IIRC, the old IDE specification stated that vendor implementation of the IDE format track command was at the discretion of the vendor. The command could format a track, zero out a track, or do nothing. Using the vendor's supplied utilities to format and mark bad sectors is always recommended. -- Cheers, Phone: 250-387-8437 Cy SchubertFax: 250-387-5766 Team Leader, Sun/Alpha Team Email: [EMAIL PROTECTED] Open Systems Group, CITS Ministry of Management Services Province of BC FreeBSD UNIX: [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: swap huge mem systems
:Thus spake Terry Lambert [EMAIL PROTECTED]: : Erik Trulsson wrote: : Minimal swap if you want to be able to catch core dumps: Physical RAM : size + 64K : :I've caught many core dumps with swap == RAM. Am I just getting :lucky, or am I losing 64K of the image? : : Crash dumps good. : :I beg to differ. ;-) You only need as much as physical ram. e.g. 1G of ram, 1G on the dump device (which can be the swap partition). -Matt Matthew Dillon [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
