Re: Large UIDs (>65536)
> Good point :) I just got a little freaked by these friendly warning messages > from pwd_mkdb: > > "/etc/pw.Z26392" 15 lines, 319 characters > chpass: updating the database... > pwd_mkdb: 14 > recommended max uid value (65535) > chpass: done > > I know the many UNIXes still have a cap at 65535 (is Linux one of them?), and > I interpeted that as an error message, not a warning. I was silly :) Actually, it's wire protocols that are the issue; NFSv2 in particular. -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ [EMAIL PROTECTED] \\ and he'll hate you for a lifetime. \\ [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Large UIDs (>65536)
Good point :) I just got a little freaked by these friendly warning messages from pwd_mkdb: "/etc/pw.Z26392" 15 lines, 319 characters chpass: updating the database... pwd_mkdb: 14 > recommended max uid value (65535) chpass: done I know the many UNIXes still have a cap at 65535 (is Linux one of them?), and I interpeted that as an error message, not a warning. I was silly :) Yann On Fri, 09 Jun 2000, you wrote: > > Hi, > > > > I was just wondering if anyone has considered allowing larger UIDs > > (unsigned long) on a FreeBSD system? What would this require changing? > > From what I can tell, the code is typedefed so if you go recompile > > everything, you should be ok. Am I missing anything here? > > How about this line from : > > typedefu_int32_t uid_t; /* user id */ > > It pays to do your own research - then you only get to look silly in > front of yourself. 8) To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Large UIDs (>65536)
> Hi, > > I was just wondering if anyone has considered allowing larger UIDs (unsigned > long) on a FreeBSD system? What would this require changing? From what I > can tell, the code is typedefed so if you go recompile everything, you should > be ok. Am I missing anything here? How about this line from : typedefu_int32_t uid_t; /* user id */ It pays to do your own research - then you only get to look silly in front of yourself. 8) -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ [EMAIL PROTECTED] \\ and he'll hate you for a lifetime. \\ [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: IOCTL to the character driver is failing
> [moved to -hackers; this is an in-depth technical question] > > On Friday, 9 June 2000 at 0:28:04 -0700, NandaKumar P.K. wrote: > > Hi, > > > > I am in the process of debugging my character driver > > interface to the RAID controller card. Lots of my > > IOCTLs i sent from the user code works with the driver > > but some IOCTLs are not reaching the driver. I found > > that one of the IOCTL that failed is having a size of > > data as 8096 bytes. Does FreeBSD has got any size > > restriction in IOCTL data size ? I know that there is > > no problem with Linux and Windows NT. > > Indeed, you're limited to 4096 bytes. From sys/ioccom.h: > > #define IOCPARM_MASK0x1fff /* parameter length, at most 13 bits */ > #define IOCPARM_MAX PAGE_SIZE /* max size of ioctl, mult. of >PAGE_SIZE */ > > >From sys/param.h: > > #define PAGE_SHIFT12 /* LOG2(PAGE_SIZE) */ > #define PAGE_SIZE (1< > Even that is probably too big. What are you trying to transfer that's > so big? NB. if you want to move more than 4k, you should pass a pointer to the buffer in userspace and use copyin/copyout. -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ [EMAIL PROTECTED] \\ and he'll hate you for a lifetime. \\ [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Large UIDs (>65536)
Hi, I was just wondering if anyone has considered allowing larger UIDs (unsigned long) on a FreeBSD system? What would this require changing? From what I can tell, the code is typedefed so if you go recompile everything, you should be ok. Am I missing anything here? Yann To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Problem mouting NFS exports from multi-homed servers
:The one problem standing in the way of my being able to implement this :solution is a very specific problem with mounting NFS exports from :multi-homed servers on our network. We have this problem both from the :FreeBSD box itself and from the "NAT'ed" clients on the 10.x.x.x networks it :serves. There are two solutions to this problem, both involving fixing the NFS server. The problem is that the FreeBSD NFS server would respond to UDP NFS packets using a different source IP then they were sent to. The solution is to either use TCP NFS mounts, or to use the -h option to nfsd (see 'man nfsd') to force nfsd to bind to and respond to UDP packets using the same interface IP. 'man nfsd' should give you enough information to fix this on your NFS server. You should not have to mess with the clients at all. I'm pretty sure I backported this feature to 3.x. -Matt To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Non-promiscuous tcpdump on 4.0-STABLE doesn't see outgoing traffic
tcpdump -p (interface in non-promiscuous mode) shows incoming and outgoing traffic in 3.4-STABLE (as expected). tcpdump -p does *not* show outgoing traffic in 4.0-STABLE. Incoming is fine. Is this intended? Steinar Haug, Nethelp consulting, [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
quota/mount commands inconsistency
Hi,
recently, I have tried to setup quotas as usual. the first
things I tries where the "rq" option then the "quota" option.
ok, the "quota" option isn't supported under FreeBSD, as well
as the "rq" option while the later is a little documented in
fstab(5) :
#define FSTAB_RQ"rq"/* read/write with quotas */
so, I use the FreeBSD "userquota" and "groupquota" options.
here is my configuration (/disk0 is my fallback root partition):
# egrep 'disk|\B/\B' /etc/fstab
/dev/ad0s3 /disk0 ufs rw,userquota,groupquota,noauto 1 1
/dev/da0s1a / ufs rw,userquota,groupquota 1 1
/dev/da1s1c /disk2 ufs rw,userquota,groupquota 1 2
/dev/da2s1a /disk1 ufs rw,userquota,groupquota 1 2
/dev/da3s1a /disk4 ufs rw,userquota,groupquota 1 2
# df -k
Filesystem 1K-blocks UsedAvail Capacity Mounted on
/dev/da0s1a 1904559 1255135 49706072%/
/dev/da1s1c 2031922 1372446 49692373%/disk2
/dev/da2s1a 1904559 16625078968895%/disk1
/dev/da3s1a 1904559 1751524 671 100%/disk4
as I already says, the "rq" option seems to no be supported any
more while mounting an ufs filesystem and the userquota and
groupquota options aren't checked as well to turn quota on while
an ufs filesystem is mounted. you have to turn on quota manually
using quotaon. huh!
# mount -o rq /dev/ad0s3a /disk0
mount: -o rq: option not supported
(well)
# mount -v /disk0
/dev/ad0s3 on /disk0 (ufs, local, soft-updates, writes: sync 2 async 0, reads: s
ync 1 async 0)
(humm... quota are not enabled at mount time!)
# mount | grep /disk1
/dev/da2s1a on /disk1 (ufs, local, with quotas, writes: sync 34 async 110, reads: sync
43 async 5)
(while there are at boot time)
# cat /tmp/c.c
#include
#include
int
main () {
char *path = "/disk0";
struct statfs buf;
if (statfs (path, &buf) < 0) {
perror ("statfs");
return (1);
}
printf ("%s: %s", path, (buf.f_flags & MNT_RDONLY) ? "ro" : "rw");
/* ufs stuffs */
if (buf.f_flags & MNT_SYNCHRONOUS)printf (",%s", "sync");
if (buf.f_flags & MNT_NOEXEC) printf (",%s", "noexec");
if (buf.f_flags & MNT_NOSUID) printf (",%s", "nosuid");
if (buf.f_flags & MNT_NODEV) printf (",%s", "nodev");
if (buf.f_flags & MNT_UNION) printf (",%s", "union");
if (buf.f_flags & MNT_ASYNC) printf (",%s", "async");
if (buf.f_flags & MNT_NOATIME)printf (",%s", "noatime");
if (buf.f_flags & MNT_NOCLUSTERR) printf (",%s", "noclusterr");
if (buf.f_flags & MNT_NOCLUSTERW) printf (",%s", "noclusterw");
if (buf.f_flags & MNT_NOSYMFOLLOW)printf (",%s", "nosymfollow");
if (buf.f_flags & MNT_SUIDDIR)printf (",%s", "suiddir");
if (buf.f_flags & MNT_SOFTDEP)printf (",%s", "soft-updates");
/* misc stuffs */
if (buf.f_flags & MNT_LOCAL) printf (",%s", "local");
if (buf.f_flags & MNT_QUOTA) printf (",%s", "quota");
if (buf.f_flags & MNT_ROOTFS) printf (",%s", "rootfs");
if (buf.f_flags & MNT_USER) printf (",%s", "user");
if (buf.f_flags & MNT_IGNORE) printf (",%s", "ignore");
/* nfs stuffs */
if (buf.f_flags & MNT_EXPORTED) {
printf (" nfs: %s", (buf.f_flags & MNT_EXRDONLY) ? "ro" : "rw");
if (buf.f_flags & MNT_DEFEXPORTED)printf (",%s", "world");
if (buf.f_flags & MNT_EXPORTANON) printf (",%s", "maproot");
if (buf.f_flags & MNT_EXKERB) printf (",%s", "kerb");
if (buf.f_flags & MNT_EXPUBLIC) printf (",%s", "webnfs");
}
printf ("\n");
return (0);
}
# make /tmp/c
cc -O -pipe /tmp/c.c -o /tmp/c
# /tmp/c
/disk0: rw,soft-updates,local
# quotaon -a
(oops, panic! checking for core dump...savecore: no core dump)
(reboot, then same things until quotaon -a)
# quotaon /disk0
# mount | grep /disk0
/dev/da0s1a on / (ufs, local, with quotas, soft-updates, writes: sync 173 async 2939,
reads: sync 1779 async 180)
(ok... quota are now enabled)
# /tmp/c
/disk0: rw,soft-updates,local,quota
# umount /disk0
(goto first mount :)
also, if a filesystem isn't mounted at boot time and quotacheck is
requested, quota files are created w/ hole in the mount point of
that filesystem. other quota commands just create an empty file.
so, quotacheck lacks to check if the filesystem is mounted before
to do anything as well as every other quota commands.
# quotacheck -a
(doesn't complain about unmounted /disk0 which is ok)
# ls -l /disk0
total 80
-rw-r- 1 root operator 2097120 Jun 5 22:09 quota.group
-rw-r- 1 root operator 2097120 Jun 5 22:09 quota.user
kernel panic seems to arrive when multiple quotaon -a are done
and/or doing q
Re: Comments on Athlon [motherboards] sought..
On Saturday, 10 June 2000 at 11:02:48 +0930, Greg Lehey wrote: > On Friday, 9 June 2000 at 19:06:09 +0200, Wilko Bulte wrote: >> On Fri, Jun 09, 2000 at 10:29:45AM +0930, Greg Lehey wrote: >>> On Thursday, 8 June 2000 at 17:33:13 -0600, Wes Peters wrote: Wilko Bulte wrote: > > I'm considering buying an Athlon based machine. Before shelling out the > $ (well, fl ) I'd like to know what experiences have with Athlon and > FreeBSD. And obviously which mom boards to prefer or keep away from. EPoX == trash. Avoid like the plague. >>> >>> Buy EPoX. They're good. >> >> Greg, what model EPox do you have? > > It's a 7KX. Do they make any others for Athlon? Correction, I see it written here as EP-7KXA. Greg -- Finger [EMAIL PROTECTED] for PGP public key 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
Re: Comments on Athlon [motherboards] sought..
On Friday, 9 June 2000 at 19:06:09 +0200, Wilko Bulte wrote: > On Fri, Jun 09, 2000 at 10:29:45AM +0930, Greg Lehey wrote: >> On Thursday, 8 June 2000 at 17:33:13 -0600, Wes Peters wrote: >>> Wilko Bulte wrote: I'm considering buying an Athlon based machine. Before shelling out the $ (well, fl ) I'd like to know what experiences have with Athlon and FreeBSD. And obviously which mom boards to prefer or keep away from. >>> >>> EPoX == trash. Avoid like the plague. >> >> Buy EPoX. They're good. > > Greg, what model EPox do you have? It's a 7KX. Do they make any others for Athlon? Greg -- Finger [EMAIL PROTECTED] for PGP public key 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
Re: IOCTL to the character driver is failing
[moved to -hackers; this is an in-depth technical question] On Friday, 9 June 2000 at 0:28:04 -0700, NandaKumar P.K. wrote: > Hi, > > I am in the process of debugging my character driver > interface to the RAID controller card. Lots of my > IOCTLs i sent from the user code works with the driver > but some IOCTLs are not reaching the driver. I found > that one of the IOCTL that failed is having a size of > data as 8096 bytes. Does FreeBSD has got any size > restriction in IOCTL data size ? I know that there is > no problem with Linux and Windows NT. Indeed, you're limited to 4096 bytes. From sys/ioccom.h: #define IOCPARM_MASK0x1fff /* parameter length, at most 13 bits */ #define IOCPARM_MAX PAGE_SIZE /* max size of ioctl, mult. of PAGE_SIZE */ >From sys/param.h: #define PAGE_SHIFT 12 /* LOG2(PAGE_SIZE) */ #define PAGE_SIZE (1
Re: Comments on Athlon [motherboards] sought.. BAD ASUS Story
Mike Smith wrote: > > > "Nicole Harrington." wrote: > > > > > > Sad to say.. I have another bad experience with the NEW asus K7A MB. It will > > > not allow a Mylex AccellRaid 150 to break out of the bootup sequence to > > > be configured. :( > > > > That may be as well due to the bugs in Mylex soft. > > In this case, that's not correct. Or at least, if it is, Adaptec, AMI, > ICP Vortex, Compaq, IBM, Symbios and just about every other vendor of > add-in cards with boot vector interests have a similar set of bugs. Then I was wrong. Sorry. Just I have seen too much problems with the Mylex configuration utility which tends to hang at various places to suspect it first. -SB To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought.. BAD ASUS Story
> "Nicole Harrington." wrote: > > > > Sad to say.. I have another bad experience with the NEW asus K7A MB. It will > > not allow a Mylex AccellRaid 150 to break out of the bootup sequence to > > be configured. :( > > That may be as well due to the bugs in Mylex soft. In this case, that's not correct. Or at least, if it is, Adaptec, AMI, ICP Vortex, Compaq, IBM, Symbios and just about every other vendor of add-in cards with boot vector interests have a similar set of bugs. > I have used different > Mylex cards on a few occasions (the last one was testing SCSI drivers > for my employer) and _always_ had some problems with > them. Sometimes upgrading the Mylex drivers and/or configuration > utility helps, sometimes does not. I hate Mylex terribly and would > strongly recommend staying as far away from Mylex as possible. I have a lot of Mylex hardware around here (obviously enough), and I can't say their stuff is any worse than anyone else's. -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ [EMAIL PROTECTED] \\ and he'll hate you for a lifetime. \\ [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought.. BAD ASUS Story
"Nicole Harrington." wrote: > > Sad to say.. I have another bad experience with the NEW asus K7A MB. It will > not allow a Mylex AccellRaid 150 to break out of the bootup sequence to > be configured. :( That may be as well due to the bugs in Mylex soft. I have used different Mylex cards on a few occasions (the last one was testing SCSI drivers for my employer) and _always_ had some problems with them. Sometimes upgrading the Mylex drivers and/or configuration utility helps, sometimes does not. I hate Mylex terribly and would strongly recommend staying as far away from Mylex as possible. -SB To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: uidinfo has many race conditions.
* Alfred Perlstein <[EMAIL PROTECTED]> [000609 16:45] wrote: > hi, > > Is it just me or does the fact that uidinfo structures (see > kern/kern_proc.c) are allocated with M_WAITOK after finding them > fails and then inserted into the uidhash structure a race condition? > > Index: kern_proc.c > === Yes, I know i forgot to put the created ones back into the list, I was just a bit flusteres after reading over the code. I'll have some more later. -Alfred To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
uidinfo has many race conditions.
hi,
Is it just me or does the fact that uidinfo structures (see
kern/kern_proc.c) are allocated with M_WAITOK after finding them
fails and then inserted into the uidhash structure a race condition?
There's also a problem with sbsize checking because of races going on
here, what needs to happen is that the changeXXsize/count functions
need to know what they are chenging and doing them without races.
I should have some diffs up soon that address this.
here's fixing the uidinfo stuff, but it still doesn't fix that
the sbsize checks before setting the size, which needs to be
done atomically. I think the solution is to have chgsbsize
actually check the rlimit instead of polling then setting.
chblob is for my own stuff.
Index: kern_proc.c
===
RCS file: /home/ncvs/src/sys/kern/kern_proc.c,v
retrieving revision 1.63
diff -u -u -r1.63 kern_proc.c
--- kern_proc.c 2000/02/08 19:54:15 1.63
+++ kern_proc.c 2000/06/09 23:40:46
@@ -56,6 +56,9 @@
MALLOC_DEFINE(M_SUBPROC, "subproc", "Proc sub-structures");
static void pgdelete __P((struct pgrp *));
+static struct uidinfo *uifind(uid_t uid);
+static struct uidinfo *uicreate(uid_t uid);
+static int uifree(struct uidinfo *uip);
/*
* Structure associated with user cacheing.
@@ -65,6 +68,7 @@
uid_t ui_uid;
longui_proccnt;
rlim_t ui_sbsize;
+ rlim_t ui_blobsize;
};
#defineUIHASH(uid) (&uihashtbl[(uid) & uihash])
static LIST_HEAD(uihashhead, uidinfo) *uihashtbl;
@@ -99,6 +103,57 @@
}
/*
+ * find/create a uidinfo struct for the uid passed in
+ */
+static struct uidinfo *
+uifind(uid_t uid)
+{
+ struct uihashhead *uipp;
+ struct uidinfo *uip;
+
+ uipp = UIHASH(uid);
+ LIST_FOREACH(uip, uipp, ui_hash)
+ if (uip->ui_uid == uid)
+ break;
+
+ return (uip);
+}
+
+static struct uidinfo *
+uicreate(uid_t uid)
+{
+ struct uidinfo *uip, *norace;
+
+ MALLOC(uip, struct uidinfo *, sizeof(*uip), M_PROC, M_WAITOK);
+ /*
+* if we M_WAITOK we must look afterwards or risk redundant entries
+*/
+ norace = uifind(uid);
+ if (norace != NULL) {
+ FREE(uip, M_PROC);
+ return (norace);
+ }
+ uip->ui_uid = uid;
+ uip->ui_proccnt = 0;
+ uip->ui_sbsize = 0;
+ uip->ui_blobsize = 0;
+ return (uip);
+}
+
+static int
+uifree(struct uidinfo *uip)
+{
+
+ if (uip->ui_sbsize == 0 && uip->ui_proccnt == 0 && uip->ui_blobsize == 0) {
+ LIST_REMOVE(uip, ui_hash);
+ FREE(uip, M_PROC);
+ return (1);
+ }
+ return (0);
+}
+
+
+/*
* Change the count associated with number of processes
* a given user is using.
*/
@@ -108,35 +163,25 @@
int diff;
{
register struct uidinfo *uip;
- register struct uihashhead *uipp;
- uipp = UIHASH(uid);
- LIST_FOREACH(uip, uipp, ui_hash)
- if (uip->ui_uid == uid)
- break;
+ uip = uifind(uid);
if (uip) {
uip->ui_proccnt += diff;
if (uip->ui_proccnt < 0)
panic("chgproccnt: procs < 0");
- if (uip->ui_proccnt > 0 || uip->ui_sbsize > 0)
- return (uip->ui_proccnt);
- LIST_REMOVE(uip, ui_hash);
- FREE(uip, M_PROC);
- return (0);
+ return (uifree(uip) == 1 ? 0 : uip->ui_proccnt);
}
if (diff <= 0) {
if (diff == 0)
return(0);
panic("chgproccnt: lost user");
}
- MALLOC(uip, struct uidinfo *, sizeof(*uip), M_PROC, M_WAITOK);
- LIST_INSERT_HEAD(uipp, uip, ui_hash);
- uip->ui_uid = uid;
+ uip = uicreate(uid);
uip->ui_proccnt = diff;
- uip->ui_sbsize = 0;
return (diff);
}
+
/*
* Change the total socket buffer size a user has used.
*/
@@ -146,12 +191,8 @@
rlim_t diff;
{
register struct uidinfo *uip;
- register struct uihashhead *uipp;
- uipp = UIHASH(uid);
- LIST_FOREACH(uip, uipp, ui_hash)
- if (uip->ui_uid == uid)
- break;
+ uip = uifind(uid);
if (diff <= 0) {
if (diff == 0)
return (uip ? uip->ui_sbsize : 0);
@@ -159,20 +200,38 @@
}
if (uip) {
uip->ui_sbsize += diff;
- if (uip->ui_sbsize == 0 && uip->ui_proccnt == 0) {
- LIST_REMOVE(uip, ui_hash);
- FREE(uip, M_PROC);
- return (0);
- }
- return (uip->ui_sbsize);
+ return (uifree(uip) == 1 ? 0 : uip->ui_sbsize);
}
- MALLOC(uip, struct uidinfo *, sizeof(*uip), M_PROC, M_WAITOK);
- LIST_INSERT_HEAD(uipp, uip, ui_hash);
Re: Comments on Athlon [motherboards] sought..
Narvi writes: > On Fri, 9 Jun 2000, Michael Bacarella wrote: > > > > > I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy >Asys > > > > > has on their praised K7V (mind you, I like Asus as such, excellent > > > > > experiences with them over the years). > > > > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > > > > slot, especially since I only heard good things about the board > > > > itself. I made an honest attempt but just ended up ordering it > > > > anyway. I've tried to figure out what that slot does but I've only come > > > > across market-speak. > > > > > > > > What does it do? > > > If AMR == Asus Media Slot, then it's a slot that is simultaneously PCI and > > > ISA, so you can have a PCi video card and ISA soundcard on one card. > > They're about a third of the size of a PCI slot, so I don't see how that > > would be conveniant. :) > But isn't it at the end of a PCI slot? If no, then they actually did come > with a next stupid thing that has the same abreviation. You are speaking about totally another thing then AMR AMR is industry standart slot and is placed on board instead PCI -- @BABOLO http://links.ru/ To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
Wilko Bulte wrote: > > On Fri, Jun 09, 2000 at 11:38:54AM -0700, Mike Smith wrote: > > > > > > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > > > slot, especially since I only heard good things about the board > > > itself. I made an honest attempt but just ended up ordering it > > > anyway. I've tried to figure out what that slot does but I've only come > > > across market-speak. > > > > > > What does it do? > > > > It conects to the AC97 modem codec in the VIA chipset; you put the modem > > line interface on a card and stick it in there. ie. it's a neat way to > > again reduce the cost of a WinModem. > > And a horrible way to loose valuable PCB real estate that could have housed a > PCI/ISA slot. A modem line interface on an L-shaped 'blech' with a cable > to the mom board would be better IMHO. Would allow the owner to throw > it away and put PCI/ISA expansion on the machine ;) Ah, the K7V already has as many PCI slots as you can drive off the existing bridge, and who wants ISA slots these days? Oh, there is the WaveLan ISA card I need... -- "Where am I, and what am I doing in this handbasket?" Wes Peters Softweyr LLC [EMAIL PROTECTED] http://softweyr.com/ To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought.. BAD ASUS Story
> > > Sad to say.. I have another bad experience with the NEW asus K7A MB. It will > not allow a Mylex AccellRaid 150 to break out of the bootup sequence to > be configured. :( Try putting "SCSI" above anything else in the "boot order" menu. I've been trying for some time now to work out how these stupid menus are sorted, and how they determine what constitutes "SCSI" as opposed to eg. an int19 device. I fear that it's not very deterministic. 8( > Ad that to the new BIOS that won't allow keyboardless reboots :( Bitch to ASUS about this, I guess. 8( -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ [EMAIL PROTECTED] \\ and he'll hate you for a lifetime. \\ [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought.. BAD ASUS Story
Sad to say.. I have another bad experience with the NEW asus K7A MB. It will not allow a Mylex AccellRaid 150 to break out of the bootup sequence to be configured. :( Ad that to the new BIOS that won't allow keyboardless reboots :( These came in a RED box. Nicole On 09-Jun-00 Wilko Bulte wrote: > On Thu, Jun 08, 2000 at 08:54:28PM +0200, Wilko Bulte wrote: >> I'm considering buying an Athlon based machine. Before shelling out the >> $ (well, fl ) I'd like to know what experiences have with Athlon and >> FreeBSD. And obviously which mom boards to prefer or keep away from. > > Anybody using the Abit KA7? If yes, do you like it? > > I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy > Asys > has on their praised K7V (mind you, I like Asus as such, excellent > experiences with them over the years). > > W/ > -- > Wilko Bulte FreeBSD, the power to serve http://www.freebsd.org > http://www.nlfug.nl > > [EMAIL PROTECTED] |\ __ /| (`\ http://www.unixgirl.com/ [EMAIL PROTECTED] | o_o |__ ) ) http://www.dangermouse.org/ // \\ ---(((---(((- -- Powered by Coka-Cola and FreeBSD -- -- Strong enough for a man - But made for a Woman -- -- OWNED? MS: Who's Been In/Virused Your Computer Today? -- --- To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Getting NIST mpeg2player to work?
On Wed, May 31, 2000 at 01:48:40PM +0200, Wilko Bulte wrote: > Does anybody have the NIST mpeg2player working correctly? I'm using the > code that sos put on freebsd.dk in Jan. It compile OK, but: > > WKB /home/src/DVD/nist>mpeg2player > Illegal instruction (core dumped) > WKB /home/src/DVD/nist> > > and: > > (gdb) bt > #0 0x8050b19 in __static_initialization_and_destruction_0 () > #1 0x8051981 in global constructors keyed to RXwhite () > #2 0x804a78c in _start () > #3 0x8049f9d in _init () > (gdb) > > ?? > > Suggestions for another [working..] mpeg2player suitable to play DVDs > are also welcome. The configure script assumes -march=i686. Which does not work too well on a K6-2 :-( Changing it into i586 made it run OK. -- Wilko Bulte FreeBSD, the power to serve http://www.freebsd.org http://www.nlfug.nl To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Fri, Jun 09, 2000 at 11:38:54AM -0700, Mike Smith wrote: > > > > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > > slot, especially since I only heard good things about the board > > itself. I made an honest attempt but just ended up ordering it > > anyway. I've tried to figure out what that slot does but I've only come > > across market-speak. > > > > What does it do? > > It conects to the AC97 modem codec in the VIA chipset; you put the modem > line interface on a card and stick it in there. ie. it's a neat way to > again reduce the cost of a WinModem. And a horrible way to loose valuable PCB real estate that could have housed a PCI/ISA slot. A modem line interface on an L-shaped 'blech' with a cable to the mom board would be better IMHO. Would allow the owner to throw it away and put PCI/ISA expansion on the machine ;) -- Wilko Bulte FreeBSD, the power to serve http://www.freebsd.org http://www.nlfug.nl To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
> > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > slot, especially since I only heard good things about the board > itself. I made an honest attempt but just ended up ordering it > anyway. I've tried to figure out what that slot does but I've only come > across market-speak. > > What does it do? It conects to the AC97 modem codec in the VIA chipset; you put the modem line interface on a card and stick it in there. ie. it's a neat way to again reduce the cost of a WinModem. -- \\ Give a man a fish, and you feed him for a day. \\ Mike Smith \\ Tell him he should learn how to fish himself, \\ [EMAIL PROTECTED] \\ and he'll hate you for a lifetime. \\ [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Fri, 9 Jun 2000, Michael Bacarella wrote: > > > > > I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy Asys > > > > has on their praised K7V (mind you, I like Asus as such, excellent > > > > experiences with them over the years). > > > > > > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > > > slot, especially since I only heard good things about the board > > > itself. I made an honest attempt but just ended up ordering it > > > anyway. I've tried to figure out what that slot does but I've only come > > > across market-speak. > > > > > > What does it do? > > > > > > > If AMR == Asus Media Slot, then it's a slot that is simultaneously PCI and > > ISA, so you can have a PCi video card and ISA soundcard on one card. > > They're about a third of the size of a PCI slot, so I don't see how that > would be conveniant. :) > But isn't it at the end of a PCI slot? If no, then they actually did come with a next stupid thing that has the same abreviation. A switched away from asus some time before athlon came out. > Michael Bacarella > New York Connect Net > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
> > > I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy Asys > > > has on their praised K7V (mind you, I like Asus as such, excellent > > > experiences with them over the years). > > > > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > > slot, especially since I only heard good things about the board > > itself. I made an honest attempt but just ended up ordering it > > anyway. I've tried to figure out what that slot does but I've only come > > across market-speak. > > > > What does it do? > > > > If AMR == Asus Media Slot, then it's a slot that is simultaneously PCI and > ISA, so you can have a PCi video card and ISA soundcard on one card. They're about a third of the size of a PCI slot, so I don't see how that would be conveniant. :) Michael Bacarella New York Connect Net To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Fri, 9 Jun 2000, Michael Bacarella wrote: > > On Fri, 9 Jun 2000, Wilko Bulte wrote: > > > On Thu, Jun 08, 2000 at 08:54:28PM +0200, Wilko Bulte wrote: > > > I'm considering buying an Athlon based machine. Before shelling out the > > > $ (well, fl ) I'd like to know what experiences have with Athlon and > > > FreeBSD. And obviously which mom boards to prefer or keep away from. > > > > Anybody using the Abit KA7? If yes, do you like it? > > > > I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy Asys > > has on their praised K7V (mind you, I like Asus as such, excellent > > experiences with them over the years). > > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > slot, especially since I only heard good things about the board > itself. I made an honest attempt but just ended up ordering it > anyway. I've tried to figure out what that slot does but I've only come > across market-speak. > > What does it do? > If AMR == Asus Media Slot, then it's a slot that is simultaneously PCI and ISA, so you can have a PCi video card and ISA soundcard on one card. IIRC, that is. > Michael Bacarella > New York Connect Net > > To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Fri, Jun 09, 2000 at 01:59:58PM -0400, Michael Bacarella wrote: > > On Fri, 9 Jun 2000, Wilko Bulte wrote: > > > On Thu, Jun 08, 2000 at 08:54:28PM +0200, Wilko Bulte wrote: > > > I'm considering buying an Athlon based machine. Before shelling out the > > > $ (well, fl ) I'd like to know what experiences have with Athlon and > > > FreeBSD. And obviously which mom boards to prefer or keep away from. > > > > Anybody using the Abit KA7? If yes, do you like it? > > > > I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy Asys > > has on their praised K7V (mind you, I like Asus as such, excellent > > experiences with them over the years). > > Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR > slot, especially since I only heard good things about the board > itself. I made an honest attempt but just ended up ordering it > anyway. I've tried to figure out what that slot does but I've only come > across market-speak. Audio Modem Riser card slot. For audio/and/or/modem use. WinModem I suspect.. If yes, worse than useless. -- Wilko Bulte FreeBSD, the power to serve http://www.freebsd.org http://www.nlfug.nl To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Fri, 9 Jun 2000, Wilko Bulte wrote: > On Thu, Jun 08, 2000 at 08:54:28PM +0200, Wilko Bulte wrote: > > I'm considering buying an Athlon based machine. Before shelling out the > > $ (well, fl ) I'd like to know what experiences have with Athlon and > > FreeBSD. And obviously which mom boards to prefer or keep away from. > > Anybody using the Abit KA7? If yes, do you like it? > > I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy Asys > has on their praised K7V (mind you, I like Asus as such, excellent > experiences with them over the years). Ugh. I was suprised that I couldn't find an Asus K7V without a stupid AMR slot, especially since I only heard good things about the board itself. I made an honest attempt but just ended up ordering it anyway. I've tried to figure out what that slot does but I've only come across market-speak. What does it do? Michael Bacarella New York Connect Net To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Thu, Jun 08, 2000 at 08:54:28PM +0200, Wilko Bulte wrote: > I'm considering buying an Athlon based machine. Before shelling out the > $ (well, fl ) I'd like to know what experiences have with Athlon and > FreeBSD. And obviously which mom boards to prefer or keep away from. Anybody using the Abit KA7? If yes, do you like it? I appreciate the KA7 PCI/ISA combo slot instead of the useless AMR thingy Asys has on their praised K7V (mind you, I like Asus as such, excellent experiences with them over the years). W/ -- Wilko Bulte FreeBSD, the power to serve http://www.freebsd.org http://www.nlfug.nl To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Fri, Jun 09, 2000 at 10:29:45AM +0930, Greg Lehey wrote: > On Thursday, 8 June 2000 at 17:33:13 -0600, Wes Peters wrote: > > Wilko Bulte wrote: > >> > >> I'm considering buying an Athlon based machine. Before shelling out the > >> $ (well, fl ) I'd like to know what experiences have with Athlon and > >> FreeBSD. And obviously which mom boards to prefer or keep away from. > > > > EPoX == trash. Avoid like the plague. > > Buy EPoX. They're good. Greg, what model EPox do you have? -- Wilko Bulte FreeBSD, the power to serve http://www.freebsd.org http://www.nlfug.nl To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Fri, Jun 09, 2000 at 09:39:53AM +0100, Doug Rabson wrote: > On Thu, 8 Jun 2000, Wilko Bulte wrote: > > > I'm considering buying an Athlon based machine. Before shelling out the > > $ (well, fl ) I'd like to know what experiences have with Athlon and > > FreeBSD. And obviously which mom boards to prefer or keep away from. > > We have one at work and it works fine. We had to put some extra cooling on > the motherboard chipset though. What motherboard are you using? -- Wilko Bulte FreeBSD, the power to serve http://www.freebsd.org http://www.nlfug.nl To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Problem mouting NFS exports from multi-homed servers
I am looking to implement FreeBSD as a router/natd platform for five private 10.x.x.x/24 subnets to connect to the public world via a sixth NIC. Our immeadiate public address space is a protected network, so I am not concerned with any firewalling features. The one problem standing in the way of my being able to implement this solution is a very specific problem with mounting NFS exports from multi-homed servers on our network. We have this problem both from the FreeBSD box itself and from the "NAT'ed" clients on the 10.x.x.x networks it serves. The FreeBSD box is question has the hostname "snowspeeder" and its primary IP address is 128.222.25.177/24. It's 'uname -a' output is: FreeBSD snowspeeder.rtp.dg.com 3.4-RELEASE FreeBSD 3.4-RELEASE #3: Tue May 30 15:59:31 EDT 2000 [EMAIL PROTECTED]:/usr/src/sys/compile/router i386 There are several servers that exploit this problem, but I will provide one practical example. The server's primary hostname is "commtg3" and it runs DG/UX R4.20MU05. It's specific hostname and address info is as follows: commtg3 128.222.8.29/24 commtg3-thiin 128.222.25.1/24 Note that the "commtg3-thiin" interface is on the same segment as the FreeBSD box (snowspeeder). This server is known to users as "commtg3." When they issue any command to access it, they use its common name. Say I try to mount an NFS export on commtg3 that I do not have rights to: root@snowspeeder-/root$ mount commtg3:/usr/opt/sdk test nfs: can't access /usr/opt/sdk: Permission denied Just as we should expect. Now let's say we try to mount an export that does not exist: root@snowspeeder-/root$ mount commtg3:/usr/ack/bleh test1 nfs: can't access /usr/ack/bleh: No such file or directory Again, just like we should expect. Now with an export that both exists and that we have rights to: root@snowspeeder-/root$ mount commtg3:/usr/local test2 (roughly three minute pause) nfs server commtg3:/usr/local: not responding Now let's try the same NFS export, only specify the hostname for the interface on the same segment: root@snowspeeder-/root$ mount commtg3-thiin:/usr/local test3 root@snowspeeder-/root$ mount /dev/wd0s3a on / (ufs, local, writes: sync 95 async 3300) /dev/wd0s3f on /usr (ufs, local, writes: sync 41 async 8214) /dev/wd0s3e on /var (ufs, local, writes: sync 540 async 5797) procfs on /proc (procfs, local) commtg3-thiin:/usr/local on /root/test3 (nfs) And as you can see, that works just fine. Now we've put a sniffer on the 128.222.25.0/24 segment and what it looks like is happening is that the requests destined to the 128.222.8.29 address go out fine on the router and are received by commtg3 just fine on that segment, but that when commtg3 answers it looks at the source IP (128.222.25.177) then it replies back on its 128.222.25.1 interface (For which I can't blame it), but then snowspeeder rejects the response packets because they do not come back with the same source address as the origional destination address of the request. What I really don't undestand is how or why we get errors for such things as "permission denied" or "no such file dor directory," yet we can't complete a proper mount request. What I believe I need to do is figure out to make FreeBSD not be so picky about where the response to mount requests are coming from. I am running the bare minumum ipfw configuration that "man natd" says is neccessary for NAT: gneff@snowspeeder-/usr/home/gneff$ cat /etc/rc.firewall /sbin/ipfw -f flush /sbin/ipfw add divert natd all from any to any via sf0 /sbin/ipfw add pass all from any to any My rc.network file is unchanged from the v3.4-release distribution. Thank you in advance for any assistance you can offer. In the hopes that it may be helpful, I will paste my kernel configuration and my rc.conf files below. Regards, Glen - machine "i386" cpu "I686_CPU" ident GENERIC maxusers32 options NMBCLUSTERS=2048 options IPFIREWALL options IPDIVERT options INET#InterNETworking options FFS #Berkeley Fast Filesystem options FFS_ROOT#FFS usable as root device [keep this!] options MFS #Memory Filesystem options NFS #Network Filesystem options "CD9660"#ISO 9660 Filesystem options PROCFS #Process filesystem options "COMPAT_43" #Compatible with BSD 4.3 [KEEP THIS!] options SCSI_DELAY=15000#Be pessimistic about Joe SCSI device options UCONSOLE#Allow users to grab the console options FAILSAFE#Be conservative options USERCONFIG #boot -c editor options VISUAL_USERCONFIG #visual boot -c editor options KTRACE #ktrace(1) syscall trace support options SYSVSHM #SYSV-style shared me
Re: 4.0-Stable/XFree86-4.0/syscons switching crash
Kazutaka YOKOTA wrote: > > >I am running 4.0-S on a Compaq Presario laptop with a Trident Cyberblade > >VGA. I couldn't get this to work in anything other than 640x480 with > >XFree86-3.3.6, so I moved to XFree86-4.0... > > > >Anyway, XFree86 4.0 works for me, except for a couple of glitches. > >Much more distressing: if I switch out of X to a text mode console with > >Ctrl-Alt-Fn, and then switch back to X, the machine freezes up > >completely and has to be power-cycled. > > Does the machine respond to ping from another machine on the network? No. It's not pining for the fjords... ;-) -- Dr Graham WheelerE-mail: [EMAIL PROTECTED] Director, Research and Development WWW:http://www.cequrux.com CEQURUX Technologies Phone: +27(21)423-6065 Firewalls/VPN SpecialistsFax:+27(21)424-3656 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: rlist.h !!! help !!!
:I have several programs that use rlist.h (for various reasons ) and I :cannot find it in FreeBSD 4.0 ! (mainly for swap info ) : :Has it been droped ? I know that the kernel/sys/rlist.h has been droped :but does affect the /usr/include/sys/rlist.h ? : :If it has been dropped what do I use instead ??? : :thank you : : :nathan The rlist stuff was used to track swap in 3.x. In 4.x swap tracking was changed to a radix tree (sys/blist.h). Ummm... it's considerably more complex then the old rlist stuff, if you can avoid dependancies on it I would avoid dependancies on it. Since swap was the only thing that used rlists, when the 4.x swap was switched over to the new blist stuff I removed the rlist stuff entirely. -Matt Matthew Dillon <[EMAIL PROTECTED]> To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
RE: Comments on Athlon [motherboards] sought..
> > > EPoX == trash. Avoid like the plague. > > Buy EPoX. They're good. > > > ASUS K7v == good. > > ASUS K7v == slow. > > May 3 08:00:02 wantadilla /kernel: microuptime() went > backwards (65202.831743 -> 65202,804412) > In fact, I've seen the microuptime stuff doing bonnies on FreeBSD 4.0-release with my Asus K7v. I've cvsupped to 4.0-stable, rebuild and not seen the problem since. (and yes, I have run bonnie) I like my Asus. I've always used Abit before this. The Asus is just more complete. A few jumper caps and all the cabling you need. It comes with four USB ports. Too bad I don't have any USB hardware yet. Kees Jan == Everyone is responsible for his own actions, and (people tend to forget this) the effect they have on others. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Comments on Athlon [motherboards] sought..
On Thu, 8 Jun 2000, Wilko Bulte wrote: > I'm considering buying an Athlon based machine. Before shelling out the > $ (well, fl ) I'd like to know what experiences have with Athlon and > FreeBSD. And obviously which mom boards to prefer or keep away from. We have one at work and it works fine. We had to put some extra cooling on the motherboard chipset though. -- Doug Rabson Mail: [EMAIL PROTECTED] Nonlinear Systems Ltd. Phone: +44 20 8442 9037 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
