Re: Setting up default boot from a wd0d
On 06/24/11 16:51, Anand Buddhdev wrote: Hello list users, I have a virtual server for testing, on which I have installed OpenBSD 4.8. The installation is on wd0a, and there's just a single / partition. I then did a second installation, and setup OpenBSD 4.9 on wd0d, also with a single / partition. At this point, I can reboot the server, and at the boot prompt, I can manually boot into one or the other system as follows: boot boot hd0a:/bsd (for OpenBSD 4.8) boot boot hd0d:/bsd (for OpenBSD 4.9) Next, I wished to make the 4.9 installation the default. Therefore, I did the following: 1. Booted into 4.9 2. Logged in as root 3. cd /usr/mdec; ./installboot /boot ./biosboot wd0 4. echo 'boot hd0d:/bsd' /etc/boot.conf My expectation was that after a reboot, the OpenBSD booter would load up, look for wd0d:/etc/boot.conf, find the boot command in there, and boot automatically into 4.9. This is based on my reading of the boot.conf man page, section 5: 5. If the file */etc/boot.conf* exists on the filesystem *boot* was loaded from, open and parse it. This file may contain any commands *boot* accepts at the interactive prompt. Though default settings usually suffice, they can be changed here. However, this didn't happen. Instead, the server booted off hd0a:/bsd, and into 4.8. Is this a bug, or have I missed something? The output of installboot -v (-n) could be helpful. Then read and understand boot_i386(8). On i386 BIOS loads MBR, (which maybe loads another MBR, ) which loads the PBR (biosboot), which loads second stage bootloader (/boot), which loads kernel (/bsd). Your problem is possibly that the MBR still loads the PBR/biosboot from your wd0a disklabel partition. Thats because your MBR only knows one OpenBSD MBR partition, which starts with your wd0a disklabel partition. I see two possible solutions to achieve dualbooting: - use only one OpenBSD biosboot and boot bootloader and install both on wd0a. You can configure this bootloader with /etc/boot.conf on wd0a. - If you really, really want to load the bootloaders from wd0d, then create another MBR partition using fdisk having same size and location as your wd0d disklabel partition. Then you can dualboot by activating either your real OpenBSD partition, starting at wd0a or this fake partition starting at wd0d. Christopher
Re: OpenBSD's choosing of a boot device -- scd
Hello Ted, Ted Unangst ted.unan...@gmail.com writes: 2011/6/26 Fridiric Perrin f...@resel.fr: I'm currently toying with OpenBSD's boot process. I see in arch/i386/stand//libsa/dev_i386.c:bdevs that one can boot from a an scd device. However, `apropos scd' doesn't tell me anything Some kind of old CDROM that was never supported. I was supposed to have deleted all references to it eight years ago, but you found some that I missed. Now it's even more deleted. OK, I'll just ignore it. BTW, you can save 10 bytes or so of memory by reducing the size of bdevs[] now :-). -- FrC)dC)ric Perrin -- http://tar-jx.bz
openbsd hard disk information
Dear list member, i have installed OpenBSD on my desktop; every thing is ok, expect for disk information report. It is showed as wd0. I am confused because as far as i know it is a sata device. Why does it (OpenBSD) see it as an old wd. Thanks in advance.
Re: openbsd hard disk information
On Mon, 27 Jun 2011 08:31:43 -0300, Friedrich Locke wrote: Dear list member, i have installed OpenBSD on my desktop; every thing is ok, expect for disk information report. It is showed as wd0. I am confused because as far as i know it is a sata device. Yep, and the old one is pata. So they are both ATA devices Why does it (OpenBSD) see it as an old wd. It doesn't. It sees it as a new wd. man wd would have told you all that. Maybe you are an old man? 8=) *** NOTE *** Please DO NOT CC me. I am subscribed to the list. Mail to the sender address that does not originate at the list server is tarpitted. The reply-to: address is provided for those who feel compelled to reply off list. Thankyou. Rod/ --- This life is not the real thing. It is not even in Beta. If it was, then OpenBSD would already have a man page for it.
Re: openbsd hard disk information
On Mon, Jun 27, 2011 at 1:31 PM, Friedrich Locke friedrich.lo...@gmail.com wrote: Dear list member, i have installed OpenBSD on my desktop; every thing is ok, expect for disk information report. It is showed as wd0. I am confused because as far as i know it is a sata device. Why does it (OpenBSD) see it as an old wd. Probably your SATA drive is configured in Legacy mode from the BIOS: go check for it in your bios settings. Ciao, David
Re: openbsd hard disk information
On 27/06/2011, at 9:31 PM, Friedrich Locke wrote: Dear list member, i have installed OpenBSD on my desktop; every thing is ok, expect for disk information report. It is showed as wd0. I am confused because as far as i know it is a sata device. Why does it (OpenBSD) see it as an old wd. why not? all you should care about is that you can talk to the blocks on your disk, how you get to them is hopefully not important. the reason is sata disks still respond to ata commands, and a lot of sata controllers appear as largely compatible to a traditional ata controller supported by the legacy ata stack. just luck i guess :) dlg
Re: openbsd hard disk information
On Mon, Jun 27, 2011 at 01:44:39PM +0200, David Coppa wrote: On Mon, Jun 27, 2011 at 1:31 PM, Friedrich Locke friedrich.lo...@gmail.com wrote: Dear list member, i have installed OpenBSD on my desktop; every thing is ok, expect for disk information report. It is showed as wd0. I am confused because as far as i know it is a sata device. Why does it (OpenBSD) see it as an old wd. Probably your SATA drive is configured in Legacy mode from the BIOS: go check for it in your bios settings. Ciao, David But beware, if yu change the bios mode to ahci, the disk shows up as a sd(4) device. On that case, you'll have to edit your existing fstab. -Otto
Re: openbsd hard disk information
On 2011 Jun 27 (Mon) at 13:54:32 +0200 (+0200), Otto Moerbeek wrote: :On Mon, Jun 27, 2011 at 01:44:39PM +0200, David Coppa wrote: : : On Mon, Jun 27, 2011 at 1:31 PM, Friedrich Locke : friedrich.lo...@gmail.com wrote: : Dear list member, : : i have installed OpenBSD on my desktop; every thing is ok, expect for : disk information report. : It is showed as wd0. I am confused because as far as i know it is a sata device. : : Why does it (OpenBSD) see it as an old wd. : : Probably your SATA drive is configured in Legacy mode from the BIOS: : go check for it in your bios settings. : : Ciao, : David : :But beware, if yu change the bios mode to ahci, the disk shows up as a :sd(4) device. On that case, you'll have to edit your existing fstab. : : -Otto : Only if you don't use DUIDs in your fstab (default these days...) -- If time heals all wounds, how come the belly button stays the same?
Re: openbsd hard disk information
On Mon, Jun 27, 2011 at 02:00:26PM +0200, Peter Hessler wrote: On 2011 Jun 27 (Mon) at 13:54:32 +0200 (+0200), Otto Moerbeek wrote: :On Mon, Jun 27, 2011 at 01:44:39PM +0200, David Coppa wrote: : : On Mon, Jun 27, 2011 at 1:31 PM, Friedrich Locke : friedrich.lo...@gmail.com wrote: : Dear list member, : : i have installed OpenBSD on my desktop; every thing is ok, expect for : disk information report. : It is showed as wd0. I am confused because as far as i know it is a sata device. : : Why does it (OpenBSD) see it as an old wd. : : Probably your SATA drive is configured in Legacy mode from the BIOS: : go check for it in your bios settings. : : Ciao, : David : :But beware, if yu change the bios mode to ahci, the disk shows up as a :sd(4) device. On that case, you'll have to edit your existing fstab. : : -Otto : Only if you don't use DUIDs in your fstab (default these days...) In current, yet, but not in the latest release, iirc. -Otto
Re: Setting up default boot from a wd0d
On Mon, Jun 27, 2011 at 08:57:27AM +0200, Christopher Zimmermann wrote: On 06/24/11 16:51, Anand Buddhdev wrote: Hello list users, I have a virtual server for testing, on which I have installed OpenBSD 4.8. The installation is on wd0a, and there's just a single / partition. I then did a second installation, and setup OpenBSD 4.9 on wd0d, also with a single / partition. At this point, I can reboot the server, and at the boot prompt, I can manually boot into one or the other system as follows: boot boot hd0a:/bsd (for OpenBSD 4.8) boot boot hd0d:/bsd (for OpenBSD 4.9) Next, I wished to make the 4.9 installation the default. Therefore, I did the following: 1. Booted into 4.9 2. Logged in as root 3. cd /usr/mdec; ./installboot /boot ./biosboot wd0 4. echo 'boot hd0d:/bsd' /etc/boot.conf The boot loader in the MBR is barely intelligent enough to load the /boot program from the first FFS in the bootable OpenBSD partition. Which will be 'a' unless you have worked very hard to make a completely confused disk layout. So it is likely not loading the /boot you think it is, nor looking at the /etc/boot.conf file you think it is. Don't think too harshly of it. It has less than 512 bytes to live in. As Nick says: DON'T DO THIS! 'a' is the root. Period. Feel free to experiment otherwise but such configurations will not ever be supported and will work by accident if at all. Ken My expectation was that after a reboot, the OpenBSD booter would load up, look for wd0d:/etc/boot.conf, find the boot command in there, and boot automatically into 4.9. This is based on my reading of the boot.conf man page, section 5: 5. If the file */etc/boot.conf* exists on the filesystem *boot* was loaded from, open and parse it. This file may contain any commands *boot* accepts at the interactive prompt. Though default settings usually suffice, they can be changed here. However, this didn't happen. Instead, the server booted off hd0a:/bsd, and into 4.8. Is this a bug, or have I missed something? The output of installboot -v (-n) could be helpful. Then read and understand boot_i386(8). On i386 BIOS loads MBR, (which maybe loads another MBR, ) which loads the PBR (biosboot), which loads second stage bootloader (/boot), which loads kernel (/bsd). Your problem is possibly that the MBR still loads the PBR/biosboot from your wd0a disklabel partition. Thats because your MBR only knows one OpenBSD MBR partition, which starts with your wd0a disklabel partition. I see two possible solutions to achieve dualbooting: - use only one OpenBSD biosboot and boot bootloader and install both on wd0a. You can configure this bootloader with /etc/boot.conf on wd0a. - If you really, really want to load the bootloaders from wd0d, then create another MBR partition using fdisk having same size and location as your wd0d disklabel partition. Then you can dualboot by activating either your real OpenBSD partition, starting at wd0a or this fake partition starting at wd0d. Christopher
Re: Setting up default boot from a wd0d
On 27 June 2011 04:46, Nick Holland n...@holland-consulting.net wrote: Hello Nick, I then did a second installation, and setup OpenBSD 4.9 on wd0d, also with a single / partition. bad. Why is this bad? What's wrong with having 2 OpenBSD installations on a single disk? it's an unsupported configuration -- OpenBSD boots from the 'a' partition. Anything else...you are pretty much on your own, and your results will probably not be portable across platforms. Things may break. You say this is a virtual server...why are you trying to multiboot a virtual server? Build a new VM! Or maybe bolt a new disk onto an old VM, if you want easy access to the old files. It's a virtual server because I'm using it for testing. My scenario is that I have a small Soekris board running an older OpenBSD (installed on a CF card). I'd like to upgrade to a newer OpenBSD without overwriting the current setup (to use as a fallback in case something is wrong with the new install). This is why I did the second installation in wd0d. If I manually boot 4.9 from wd0d, everything just works. Why do you say that this is an unsupported configuration? I know that the convention is to use 'a' for / and the other partitions for other things, but it can't be a crime to use a 'd' partition as the root for another OpenBSD installation. I can't think of any good reason to have different versions on the same machine, 'specially a past version. Some people want to pretend to develop (on -current) and use -release/-stable for production, but if you are developing, eat your own dog food, and USE -current, you aren't doing anyone any favor by busting -current and using something else. The closest justification I can find is on an amd64 system, to test both amd64 and i386 code. I just explained that I'd like to install a new version on an existing CF card, so that I can fall back to the old version in case anything goes wrong. Actually, I know of a way to make this work, which involves moving the new 'd' partition into the place of 'a', and moving the old 'a' partition into the place of 'd'. This works, and allows for auto booting into the new installation while still preserving the old one for fallback. However, it involves the ugly step of moving partitions around inside the disklabel. Surely the ability to boot off another partition would be elegant, no? If you really want to have multiple versions of OpenBSD on one physical i386/amd64 machine, use some kind of boot manager program which hides inactive partitions. You can do this manually using bsd.rd and fdisk, changing the desired partition to type 'a6' and flagged active, and the unwanted partition as some other type and NOT flagged active. However, I've tried this, I know exactly what needs to be done, and I've made a lot of mistakes trying to do what I know needed to be done. Mistakes result in unbootable disks to corrupt file systems (big lesson: boot from bsd.rd, don't try to use fdisk to change the ID of a running system. That's the corrupted FS). Ok, so just to be clear: you're suggesting that I create *two* OpenBSD slices on a disk using fdisk, and then hide the one I don't want to boot from?
Re: Setting up default boot from a wd0d
On 2011 Jun 27 (Mon) at 14:32:47 +0200 (+0200), Anand Buddhdev wrote: :My scenario is that I have a small Soekris board running an older OpenBSD :(installed on a CF card). : :I'd like to upgrade to a newer OpenBSD without overwriting the current setup :(to use as a fallback in case something is wrong with the new install). Buy two CF cards. This way, you can even install and test without touching the currently running system. -- One way to stop a runaway horse is to bet on him.
aucat(1) mixing streams from different users
Hi, I start mpd and aucat with default settings using rc scripts. aucat thus runs as user _sndio and mpd runs as _mpd. Access to sndio between mpd and programs running as a different user are mutually exclusive, so when mpd is playing music nothing else can access the sound device. According to the aucat(1) manpage: [...] the server can be started by the super-user, [...] but for privacy reasons only one user may have connections to it at a given time. I was wondering if this is intended behaviour because aucat here does not run as root but as _sndio. Best regards, Jona -- Worse is better Richard P. Gabriel
Re: Setting up default boot from a wd0d
On 27 June 2011 14:40, Peter Hessler phess...@theapt.org wrote: On 2011 Jun 27 (Mon) at 14:32:47 +0200 (+0200), Anand Buddhdev wrote: :My scenario is that I have a small Soekris board running an older OpenBSD :(installed on a CF card). : :I'd like to upgrade to a newer OpenBSD without overwriting the current setup :(to use as a fallback in case something is wrong with the new install). Buy two CF cards. This way, you can even install and test without touching the currently running system. Hi Peter, Yes, that's also another option we can (and do) use for some systems. I'm just amazed that the OpenBSD boot loader can't be configured (by default) to boot off another partition. Anyway, from all the answers here (some of which are rather preachy and evangelical: don't do this), I've gathered that what I am asking for isn't possible. That's ok. I'll just fall back to some of the other options we have (new CF card, swapping partitions in the disklabel), all of which work. Thanks everyone for taking the time to answer my question. Regards, Anand
Re: aucat(1) mixing streams from different users
On Mon, Jun 27, 2011 at 02:42:42PM +0200, Jona Joachim wrote: Hi, I start mpd and aucat with default settings using rc scripts. aucat thus runs as user _sndio and mpd runs as _mpd. Access to sndio between mpd and programs running as a different user are mutually exclusive, so when mpd is playing music nothing else can access the sound device. According to the aucat(1) manpage: [...] the server can be started by the super-user, [...] but for privacy reasons only one user may have connections to it at a given time. I was wondering if this is intended behaviour because aucat here does not run as root but as _sndio. This is ok, the _sndio user is to avoid running the server with root privileges, it's not involved in the authentication process. You can find slightly more details about how it works in the AUTHENTICATION section of the sndio(7) man page. The simpler -- and most natural imho -- would be configure mpd to use unix domain sockets (instead of TCP) and to run it as your user id instead of _mpd. If you can't, you can cheat by copying mpd's ~/.aucat_cookie in your home directory (it must have mode 0600) this way aucat will consider _mpd and you are the same person. After all, I guess you run _mpd for you ;) -- Alexandre
Re: Setting up default boot from a wd0d
On 06/27/2011 08:32 AM, Anand Buddhdev wrote: On 27 June 2011 04:46, Nick Hollandn...@holland-consulting.net wrote: Hello Nick, I then did a second installation, and setup OpenBSD 4.9 on wd0d, also with a single / partition. bad. Why is this bad? What's wrong with having 2 OpenBSD installations on a single disk? it isn't supported. The developers make (and document) certain assumptions about how a system is to be used. You violated those assumptions. That simple. it's an unsupported configuration -- OpenBSD boots from the 'a' partition. Anything else...you are pretty much on your own, and your results will probably not be portable across platforms. Things may break. ... My scenario is that I have a small Soekris board running an older OpenBSD (installed on a CF card). I'd like to upgrade to a newer OpenBSD without overwriting the current setup (to use as a fallback in case something is wrong with the new install). This is why I did the second installation in wd0d. If I manually boot 4.9 from wd0d, everything just works. Why do you say that this is an unsupported configuration? I know that the convention is to use 'a' for / and the other partitions for other things, but it can't be a crime to use a 'd' partition as the root for another OpenBSD installation. I say it is unsupported because it is. I get to hang around with the developers (a very special privilege, I must tell you). I get yelled at if I suggest something that isn't fully supported. It's not a legal crime to do things wrong with OpenBSD (hey, OpenBSD defends your right to do anything you want with it far more than most other OSs), but it isn't just a convention -- it's how you do things properly. If you do things improperly and things don't work as you imagine, you are 100% on your own, and we will assume things WILL BREAK. Ok, so your goal is rollback from a failed upgrade. There are better ways of doing this, a few of them: 1) Alternate flash cards. 2) image the existing disk to another machine (flash cards are relatively small, relatively fast to do this), revert the image if needed. 3) don't sweat it -- practice on a different machine, and just do your upgrade. Soekris are popular machines in the OpenBSD world, and the upgrade process works well. Likelihood of a problem when just using it as intended is very low. All these are supported processes. Now, you have to look at the risks of having the system fail BECAUSE you are doing unsupported things with it. I've played with multibooting OpenBSD systems. Can be done, but FAR more likely to blow out your system trying to protect it from upgrade failures than an upgrade failure itself is to occur. I.e., by trying to avoid one unlikely problem, you are much more likely to create downtime by creating new problems that never should have existed. Some people call that job security, I call that bad system administration. Simple systems have simple problems. Complex systems blow their guts all over the place and take hours or days to repair. ... Actually, I know of a way to make this work, which involves moving the new 'd' partition into the place of 'a', and moving the old 'a' partition into the place of 'd'. This works, and allows for auto booting into the new installation while still preserving the old one for fallback. You just think you know how this is working. Read faq14, 'specially the section on how i386/amd64 boots, pay special attention to the PBR. You are using PARTS of one system, parts of another. Maybe that's good enough...sometimes. However, it involves the ugly step of moving partitions around inside the disklabel. Surely the ability to boot off another partition would be elegant, no? Maybe (not likely), but the developers have spent their time on more real issues, like making sure upgrades go smoothly and the system can boot properly and Just Work on many different kinds of machines, rather than trying to support bad solutions to non-problems. What you propose has complexities you don't seem to appreciate. Making it all work properly and perfectly on all the buggy BIOSs of the PC world is non-trivial...and when you get done with it, so what?. If you really want to have multiple versions of OpenBSD on one physical i386/amd64 machine, use some kind of boot manager program which hides inactive partitions. You can do this manually using bsd.rd and fdisk, changing the desired partition to type 'a6' and flagged active, and the unwanted partition as some other type and NOT flagged active. However, I've tried this, I know exactly what needs to be done, and I've made a lot of mistakes trying to do what I know needed to be done. Mistakes result in unbootable disks to corrupt file systems (big lesson: boot from bsd.rd, don't try to use fdisk to change the ID of a running system. That's the corrupted FS). Ok, so just to be clear: you're suggesting that I create *two* OpenBSD slices on a disk
Re: aucat(1) mixing streams from different users
On 2011-06-27, Alexandre Ratchov a...@caoua.org wrote: On Mon, Jun 27, 2011 at 02:42:42PM +0200, Jona Joachim wrote: Hi, I start mpd and aucat with default settings using rc scripts. aucat thus runs as user _sndio and mpd runs as _mpd. Access to sndio between mpd and programs running as a different user are mutually exclusive, so when mpd is playing music nothing else can access the sound device. According to the aucat(1) manpage: [...] the server can be started by the super-user, [...] but for privacy reasons only one user may have connections to it at a given time. I was wondering if this is intended behaviour because aucat here does not run as root but as _sndio. This is ok, the _sndio user is to avoid running the server with root privileges, it's not involved in the authentication process. You can find slightly more details about how it works in the AUTHENTICATION section of the sndio(7) man page. The simpler -- and most natural imho -- would be configure mpd to use unix domain sockets (instead of TCP) and to run it as your user id instead of _mpd. If you can't, you can cheat by copying mpd's ~/.aucat_cookie in your home directory (it must have mode 0600) this way aucat will consider _mpd and you are the same person. After all, I guess you run _mpd for you ;) I see, this is the new authentication mechanism kicking in :) Thanks for the explanation, now that I know what's causing it, it's easy to fix. Best regards, Jona -- Worse is better Richard P. Gabriel
parameter scope in bourne vs ksh functions
Hi,
ksh(1) states this:
Functions defined with the function reserved word are treated differently
in the following ways from functions defined with the () notation:
[...]
o Parameter assignments preceding function calls are not kept in the
shell environment (executing Bourne-style functions will keep
assignments).
This does not work for me:
$ i=foo
$ function fun { echo $i; }
$ fun
foo
$ i=bar
$ fun
bar
$ function fun2 { echo $j; }
$ fun2
$ j=foo
$ fun2
foo
$
Have I got something wrong there?
Christopher
Re: parameter scope in bourne vs ksh functions
On Mon, Jun 27, 2011 at 11:00 AM, Christopher Zimmermann
madro...@zakweb.de wrote:
Hi,
ksh(1) states this:
Functions defined with the function reserved word are treated differently
in the following ways from functions defined with the () notation:
[...]
o Parameter assignments preceding function calls are not kept in the
shell environment (executing Bourne-style functions will keep
assignments).
This does not work for me:
$ i=foo
$ function fun { echo $i; }
That doesn't count as an assignment preceding a function call.
Compare with what happens running the test below.
function f1 {
echo $i;
}
f2() {
echo $i;
}
i=foo
i=1 f1
f1
i=bar
i=2 f2
f2
4.8 and 4.9 Panic on Supermicro P8SCi
Hi, At the end of May, there was discussion on the list regarding the panic related to interdrm when installing 4.8 or 4.9 on Supermicro P8SCi mobos. At the time, it looked like a fix was going to be checked in. I was hoping that a new ISO might be posted with this fix, but I don't see anything later than March 2nd on the ftp site. Does anyone know if there is a patched installer available? Thanks, --Bill
Re: 4.8 and 4.9 Panic on Supermicro P8SCi
On 06/27/2011 02:03 PM, listmail wrote: Hi, At the end of May, there was discussion on the list regarding the panic related to interdrm when installing 4.8 or 4.9 on Supermicro P8SCi mobos. At the time, it looked like a fix was going to be checked in. I was hoping that a new ISO might be posted with this fix, but I don't see anything later than March 2nd on the ftp site. Does anyone know if there is a patched installer available? Thanks, --Bill pub/OpenBSD/snapshots/arch is what you are looking for on your favorite mirror. And maybe a read-through of http://www.openbsd.org/faq/faq5.html (this will probably answer the questions you are about to ask in response to the above...) Nick.
Re: parameter scope in bourne vs ksh functions
On 06/27/11 17:49, Ted Unangst wrote:
On Mon, Jun 27, 2011 at 11:00 AM, Christopher Zimmermann
madro...@zakweb.de wrote:
Hi,
ksh(1) states this:
Functions defined with the function reserved word are treated differently
in the following ways from functions defined with the () notation:
[...]
o Parameter assignments preceding function calls are not kept in the
shell environment (executing Bourne-style functions will keep
assignments).
This does not work for me:
$ i=foo
$ function fun { echo $i; }
That doesn't count as an assignment preceding a function call.
Compare with what happens running the test below.
function f1 {
echo $i;
}
f2() {
echo $i;
}
i=foo
i=1 f1
f1
i=bar
i=2 f2
f2
$ function f1 { echo $i; }
$ f2 () { echo $i; }
$ i=foo
$ i=1 f1
1
$ echo $i
foo
$ i=1 true
$ echo $i
foo
$ i=1 f2
1
$ echo $i
1
Ok, I got it. But wtf? Thats creepy! Is there any rationale behind this
strange bourne behaviour?!?
Re: splassert: assertwaitok: want -1 have 1
On 2011-05-21, Ed D. li...@rensseltucky.com wrote:
I'm posting this because I haven't yet found
a resolution ot this problem.
From what I understand, this is fixed with this commit in -current
-
PatchSet 3072
Date: 2011/04/17 21:44:27
Author: stsp
Branch: HEAD
Tag: (none)
Log:
in_ifinit() is not prepared to be called from interrupt context so add
addresses from process context by scheduling a work queue.
Fixes an assertwaitok panic. ok henning
Members:
if_spppsubr.c:1.84-1.85
Index: src/sys/net/if_spppsubr.c
diff -u src/sys/net/if_spppsubr.c:1.84 src/sys/net/if_spppsubr.c:1.85
--- src/sys/net/if_spppsubr.c:1.84 Tue Jan 11 15:42:05 2011
+++ src/sys/net/if_spppsubr.c Sun Apr 17 20:44:27 2011
@@ -1,4 +1,4 @@
-/* $OpenBSD: if_spppsubr.c,v 1.84 2011/01/11 15:42:05 deraadt Exp $
*/
+/* $OpenBSD: if_spppsubr.c,v 1.85 2011/04/17 20:44:27 stsp Exp $ */
/*
* Synchronous PPP/Cisco link level subroutines.
* Keepalive protocol implemented in both Cisco and PPP modes.
@@ -47,6 +47,7 @@
#include sys/syslog.h
#include sys/malloc.h
#include sys/mbuf.h
+#include sys/workq.h
#if defined (__OpenBSD__)
#include sys/timeout.h
@@ -410,8 +411,7 @@
HIDE void sppp_qflush(struct ifqueue *ifq);
int sppp_update_gw_walker(struct radix_node *rn, void *arg, u_int);
void sppp_update_gw(struct ifnet *ifp);
-HIDE void sppp_set_ip_addrs(struct sppp *sp, u_int32_t myaddr,
- u_int32_t hisaddr);
+HIDE void sppp_set_ip_addrs(void *, void *);
HIDE void sppp_clear_ip_addrs(struct sppp *sp);
HIDE void sppp_set_phase(struct sppp *sp);
@@ -3024,19 +3024,38 @@
addlog(\n);
}
+struct sppp_set_ip_addrs_args {
+ struct sppp *sp;
+ u_int32_t myaddr;
+ u_int32_t hisaddr;
+};
+
HIDE void
sppp_ipcp_tlu(struct sppp *sp)
{
+ struct ifnet *ifp = sp-pp_if;
+ struct sppp_set_ip_addrs_args *args;
+
+ args = malloc(sizeof(*args), M_TEMP, M_NOWAIT);
+ if (args == NULL)
+ return;
+
+ args-sp = sp;
+
/* we are up. Set addresses and notify anyone interested */
- u_int32_t myaddr, hisaddr;
- sppp_get_ip_addrs(sp, myaddr, hisaddr, 0);
+ sppp_get_ip_addrs(sp, args-myaddr, args-hisaddr, 0);
if ((sp-ipcp.flags IPCP_MYADDR_DYN)
(sp-ipcp.flags IPCP_MYADDR_SEEN))
- myaddr = sp-ipcp.req_myaddr;
+ args-myaddr = sp-ipcp.req_myaddr;
if ((sp-ipcp.flags IPCP_HISADDR_DYN)
(sp-ipcp.flags IPCP_HISADDR_SEEN))
- hisaddr = sp-ipcp.req_hisaddr;
- sppp_set_ip_addrs(sp, myaddr, hisaddr);
+ args-hisaddr = sp-ipcp.req_hisaddr;
+
+ if (workq_add_task(NULL, 0, sppp_set_ip_addrs, args, NULL)) {
+ free(args, M_TEMP);
+ printf(%s: workq_add_task failed, cannot set
+ addresses\n, ifp-if_xname);
+ }
}
HIDE void
@@ -4689,15 +4708,27 @@
}
/*
+ * Work queue task adding addresses from process context.
* If an address is 0, leave it the way it is.
*/
HIDE void
-sppp_set_ip_addrs(struct sppp *sp, u_int32_t myaddr, u_int32_t hisaddr)
+sppp_set_ip_addrs(void *arg1, void *arg2)
{
- STDDCL;
+ struct sppp_set_ip_addrs_args *args = arg1;
+ struct sppp *sp = args-sp;
+ u_int32_t myaddr = args-myaddr;
+ u_int32_t hisaddr = args-hisaddr;
+ struct ifnet *ifp = sp-pp_if;
+ int debug = ifp-if_flags IFF_DEBUG;
struct ifaddr *ifa;
struct sockaddr_in *si;
struct sockaddr_in *dest;
+ int s;
+
+ /* Arguments are now on local stack so free temporary storage. */
+ free(args, M_TEMP);
+
+ s = splsoftnet();
/*
* Pick the first AF_INET address from the list,
@@ -4748,10 +4779,12 @@
if (debug error) {
log(LOG_DEBUG, SPP_FMT sppp_set_ip_addrs: in_ifinit
failed, error=%d\n, SPP_ARGS(ifp), error);
+ splx(s);
return;
}
sppp_update_gw(ifp);
}
+ splx(s);
}
/*
Re: parameter scope in bourne vs ksh functions
On Mon, Jun 27, 2011 at 2:44 PM, Christopher Zimmermann madro...@zakweb.de wrote: Ok, I got it. But wtf? Thats creepy! Is there any rationale behind this strange bourne behaviour?!? I would guess historical accident or some quirk of the implementation. Then once it was discovered, it became unfixable because of back compat. Fortunately, ksh is not strictly tied to bourne compatibility.
umount(8) by device + overloaded mountpoint is not catched
Hi, when umount(8)ing by device, umount fetches the mountpoint via getmntinfo(3), because unmount(2) only supports unmounting via mountpoint. This means it is simply impossible to unmount a specific filesystem from an overloaded mountpoint. I think umount(8) should detect this case, warn the user and give up. Instead it plays roulette and calls unmount(2) on the mountpoint. This should be easy to fix, but it will require some restructuring of the code. That's because the current code cannot discriminate between the unmount by special device and unmount by mountpoint cases in the getmntname() function where the evaluation of the statfs(2) structures from getmntinfo(3) happens. Regards, Christopher Here's an umount run in which it went wrong. $ mount /dev/wd0a on / type ffs (local) mfs:1701 on /tmp type mfs (asynchronous, local, nodev, nosuid, size=2096816 512-blocks) /dev/wd0d on /usr type ffs (local, nodev) /dev/wd0e on /usr/local type ffs (local, nodev) /dev/sd0d on /var type ffs (local, nodev, nosuid, softdep) /dev/sd0e on /home type ffs (local, nodev, nosuid, softdep) /dev/sd1i on /vol/sd1i type msdos (local, uid=0, gid=10, mask=0660, direxec) /dev/wd0i on /vol/sd1i type msdos (local, uid=0, gid=10, mask=0770) $ sudo umount /dev/sd1i $ mount /dev/wd0a on / type ffs (local) mfs:1701 on /tmp type mfs (asynchronous, local, nodev, nosuid, size=2096816 512-blocks) /dev/wd0d on /usr type ffs (local, nodev) /dev/wd0e on /usr/local type ffs (local, nodev) /dev/sd0d on /var type ffs (local, nodev, nosuid, softdep) /dev/sd0e on /home type ffs (local, nodev, nosuid, softdep) /dev/sd1i on /vol/sd1i type msdos (local, uid=0, gid=10, mask=0660, direxec) $
ZTE MF 180 USB modem
Hi, I have a friend in the Peace Corp in Morocco. He's using a ZTE MF 180 USB modem connecting with Maroc Telecom on his windows based Acer Netbook. I'm already sick of walking him through removing the latest windows 'sploit of the week, especially given that we do it through e-mail and across a lot of time zones :-) I'm thinking about making him a USB bootable OpenBSD installation with KDE to get him onto a secure platform but the modem has to work. Has anyone had any experience with the MF 180? Mr Google didn't turn up anything for OpenBSD. Thanks, Jeff
Raid configuration help.
Hello all, I'm new so if I do anything rude please let me know so I can adjust. I set up an openbsd file server using sftp, with a raid 1 configuration and failover with Carp. My plan was to use rsync to sync up the file sharing directories and I believe I got a little bit ahead of myself when I thought I could rsync the /etc folder to streamline the configuration of my servers. It's quite possible that one of my systems was broken from an earlier attempt at reinstalling (which synced a functional, yet broken configuration with a good system), at the time I thought the raid automatically fixed the attempt at reinstalling, but after looking at my fstab file I'm pretty certain the raid hadn't been functioning properly for a while. (very stable system though that fails so gracefully) Anyways, long story short, I have a lot of work on that system, I'm not hoping to recover the beast, I just want to copy my files off of it and transfer them to a working system, but it is booting read-only with /etc/nologin in place and I cannot figure out why. Please help. Zeb Packard
Sistema Compranet 5.0 Actualizaciones y uso
[IMAGE] Empresa Registrada ante la STPS Reg. COLG640205CP30005 Smguenos en Twitter@pmscapacitacion o bien en Facebook PMS de Mixico Conozca de la mano de un experto el correcto uso de este sistema !No deje pasar oportunidades de Negocio! Manejo Sptimo del Sistema Compranet 5.0 29 de Junio Guadalajara / 01 de Julio Monterrey Este taller esta orientado al uso del sistema Compranet 5.0, el participante podra realizar una bzsqueda de licitaciones, manejara correctamente el envmo de propuestas, conociendo al detalle el funcionamiento de esta nueva plataforma. !Beneficios que obtendra con este programa! Le aseguramos que con este programa usted adquirira los conocimientos necesarios para convertirse en un proveedor de gobierno altamente efectivo. Duracisn: 10 horas impartidas por nuestro experto en la materia Mtro. Alberto Ledesma G. ?A Quiin va Dirigido? Empresarios, Contratistas, Servidores Pzblicos y personas relacionadas con cualquier Proceso de Contrataciones y Licitaciones de Adquisiciones, Arrendamientos y Servicios. Solicite mas informes y analice este excelente evento. !Reservaciones! Solicite Mayores informes, Llamenos al (33) 8851-2365, (33) 8851-2741 Uno de nuestros asesores con gusto le atendera Responda esta invitacisn con sus datos para enviar el programa completo. Empresa: Nombre: Telifono: Email: Nzmero de Interesados: !Gracias! Copyright (C) 2010, PMS Capacitacisn Efectiva de Mixico S.C. Derechos Reservados. PMS de Mixico, El logo de PMS de Mixico son marcas registradas. ADVERTENCIA PMS de Mixico no cuenta con alianzas estratigicas de ningzn tipo dentro de la Repzblica Mexicana. NO SE DEJE ENGAQAR - DIGA NO A LA PIRATERIA. Todos los logotipos, marcas comerciales e imagenes son propiedad de sus respectivas corporaciones y se utilizan con fines informativos solamente. Este Mensaje ha sido enviado a misc@openbsd.org como usuario de Pms de Mixico o bien un usuario le refiris para recibir este boletmn. Como usuario de Pms de Mixico, en este acto autoriza de manera expresa que Pms de Mixico le puede contactar vma correo electrsnico u otros medios. Si usted ha recibido este mensaje por error, haga caso omiso de el y reporte su cuenta respondiendo este correo con el subject BAJACOMPRANET Unsubscribe to this mailing list, reply a blank message with the subject UNSUBSCRIBE BAJACOMPRANET Tenga en cuenta que la gestisn de nuestras bases de datos es de suma importancia y no es intencisn de la empresa la inconformidad del receptor. [demime 1.01d removed an attachment of type image/png which had a name of nlogo.png]
Sistema Compranet 5.0 Actualizaciones y uso
[IMAGE] Empresa Registrada ante la STPS Reg. COLG640205CP30005 Smguenos en Twitter@pmscapacitacion o bien en Facebook PMS de Mixico Conozca de la mano de un experto el correcto uso de este sistema !No deje pasar oportunidades de Negocio! Manejo Sptimo del Sistema Compranet 5.0 29 de Junio Guadalajara / 01 de Julio Monterrey Este taller esta orientado al uso del sistema Compranet 5.0, el participante podra realizar una bzsqueda de licitaciones, manejara correctamente el envmo de propuestas, conociendo al detalle el funcionamiento de esta nueva plataforma. !Beneficios que obtendra con este programa! Le aseguramos que con este programa usted adquirira los conocimientos necesarios para convertirse en un proveedor de gobierno altamente efectivo. Duracisn: 10 horas impartidas por nuestro experto en la materia Mtro. Alberto Ledesma G. ?A Quiin va Dirigido? Empresarios, Contratistas, Servidores Pzblicos y personas relacionadas con cualquier Proceso de Contrataciones y Licitaciones de Adquisiciones, Arrendamientos y Servicios. Solicite mas informes y analice este excelente evento. !Reservaciones! Solicite Mayores informes, Llamenos al (33) 8851-2365, (33) 8851-2741 Uno de nuestros asesores con gusto le atendera Responda esta invitacisn con sus datos para enviar el programa completo. Empresa: Nombre: Telifono: Email: Nzmero de Interesados: !Gracias! Copyright (C) 2010, PMS Capacitacisn Efectiva de Mixico S.C. Derechos Reservados. PMS de Mixico, El logo de PMS de Mixico son marcas registradas. ADVERTENCIA PMS de Mixico no cuenta con alianzas estratigicas de ningzn tipo dentro de la Repzblica Mexicana. NO SE DEJE ENGAQAR - DIGA NO A LA PIRATERIA. Todos los logotipos, marcas comerciales e imagenes son propiedad de sus respectivas corporaciones y se utilizan con fines informativos solamente. Este Mensaje ha sido enviado a misc@openbsd.org como usuario de Pms de Mixico o bien un usuario le refiris para recibir este boletmn. Como usuario de Pms de Mixico, en este acto autoriza de manera expresa que Pms de Mixico le puede contactar vma correo electrsnico u otros medios. Si usted ha recibido este mensaje por error, haga caso omiso de el y reporte su cuenta respondiendo este correo con el subject BAJACOMPRANET Unsubscribe to this mailing list, reply a blank message with the subject UNSUBSCRIBE BAJACOMPRANET Tenga en cuenta que la gestisn de nuestras bases de datos es de suma importancia y no es intencisn de la empresa la inconformidad del receptor. [demime 1.01d removed an attachment of type image/png which had a name of nlogo.png]
