Re: Numerous hard hangs on TWO different ASUS P4T-E w/P4 1.6G
On Fri, 7 Jun 2002, Frank Mayhar wrote: > I see very common short-term hangs, a few seconds to less than a minute. > The mouse and keyboard stop responding, X stops updating and everything > just pauses, the whole system (including the network). It then starts i've seen this happen on an Asus 1400R 1U rackmount server, P3 1Ghz, dual intel etherexpress pro (fxp devices), 4GB RAM. thru trial and error i tracked it down to apm causing it. disabling apm has not shown a recurrence of the hang. however in my case, when it hung, everything froze except the box still responded to pings. telnet, ssh or any other network service wasnt responding however, just icmp. Regards, /\_/\ "All dogs go to heaven." [EMAIL PROTECTED](0 0)http://www.alphaque.com/ +==oOO--(_)--OOo==+ | for a in past present future; do| | for b in clients employers associates relatives neighbours pets; do | | echo "The opinions here in no way reflect the opinions of my $a $b." | | done; done | +=+ To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Numerous hard hangs on TWO different ASUS P4T-E w/P4 1.6G
Dinesh Nair wrote: >>I see very common short-term hangs, a few seconds to less than a minute. >>The mouse and keyboard stop responding, X stops updating and everything >>just pauses, the whole system (including the network). It then starts > > i've seen this happen on an Asus 1400R 1U rackmount server, P3 1Ghz, dual > intel etherexpress pro (fxp devices), 4GB RAM. thru trial and error i > tracked it down to apm causing it. I can't disable apm in the BIOS (doesn't have an option for it) but it's not enabled in the kernel - we were getting spurious signal 11's with apm on SMP machines several years ago, and the mailing list consensus back then was "apm is not for SMP". So either it's something else, or my BIOS is lacking an option. Lars -- Lars Eggert <[EMAIL PROTECTED]> USC Information Sciences Institute smime.p7s Description: S/MIME Cryptographic Signature
kernel crash in nfs code...
Hi,
I recently had 3 kernel crashes on 2 different
machines in the same place. The machines have 3 GB
memory (but MAXMEM is set to 1 GB in kernel config
file). The machines are running freeBSD 4.4, Sept 16th
build. Enlcosed is the trace of the crash...
#0 dumpsys () at ../../kern/kern_shutdown.c:473
#1 0xc01bcc33 in boot (howto=256) at
../../kern/kern_shutdown.c:313
#2 0xc01bd000 in poweroff_wait (junk=0xc03b252c,
howto=-1069866929)
at ../../kern/kern_shutdown.c:581
#3 0xc0343a52 in trap_fatal (frame=0xf1782c5c,
eva=2714115144)
at ../../i386/i386/trap.c:956
#4 0xc0343725 in trap_pfault (frame=0xf1782c5c,
usermode=0, eva=2714115144)
at ../../i386/i386/trap.c:849
#5 0xc03432e3 in trap (frame={tf_fs = -208928752,
tf_es = 16, tf_ds = 16,
tf_edi = -243782356, tf_esi = -304356480, tf_ebp
= -243782448,
tf_isp = -243782520, tf_ebx = -1011973632,
tf_edx = -208871772,
tf_ecx = -304356480, tf_eax = -1580852480,
tf_trapno = 12, tf_err = 0,
tf_eip = -1071153738, tf_cs = 8, tf_eflags =
66194,
tf_esp = -1011973632, tf_ss = -304356480}) at
../../i386/i386/trap.c:448
#6 0xc0277db6 in nfs_removerpc (dvp=0xf38cdea4,
name=0xc3ae820c "",
namelen=1, cred=0x0, proc=0x0) at
../../nfs/nfs_vnops.c:1643
#7 0xc0277d9d in nfs_removeit (sp=0xc3ae8200) at
../../nfs/nfs_vnops.c:1622
#8 0xc024cd14 in nfs_inactive (ap=0xf1782d2c) at
../../nfs/nfs_node.c:245
#9 0xc01ea442 in vput (vp=0xf38ccc80) at
vnode_if.h:815
#10 0xc01e871a in lookup (ndp=0xf1782ecc) at
../../kern/vfs_lookup.c:588
#11 0xc01e7ef8 in namei (ndp=0xf1782ecc) at
../../kern/vfs_lookup.c:153
#12 0xc01f0ac3 in vn_open (ndp=0xf1782ecc, fmode=1,
cmode=0)
at ../../kern/vfs_vnops.c:137
---Type to continue, or q to quit---
#13 0xc01eca44 in open (p=0xeddbe380, uap=0xf1782f80)
at ../../kern/vfs_syscalls.c:996
#14 0xc0343cfd in syscall2 (frame={tf_fs = 135266351,
tf_es = 47,
tf_ds = -1079836625, tf_edi = 4, tf_esi =
673140944,
tf_ebp = -1079901748, tf_isp = -243781676,
tf_ebx = 673058916,
tf_edx = 0, tf_ecx = 10, tf_eax = 5, tf_trapno =
22, tf_err = 2,
tf_eip = 672961720, tf_cs = 31, tf_eflags = 659,
tf_esp = -1079901792,
tf_ss = 47}) at ../../i386/i386/trap.c:1155
#15 0xc0334115 in Xint0x80_syscall ()
MESSAGE LOG
---
Fatal trap 12: page fault while in kernel mode
Jun 7 05:05:49 10_10_101_102 /kernel: fault virtual
address = 0xa1c61c48
Jun 7 05:05:49 10_10_101_102 /kernel: fault code
= supervisor read, page not present
Jun 7 05:05:49 10_10_101_102 /kernel: instruction
pointer= 0x8:0xc0277db6
Jun 7 05:05:49 10_10_101_102 /kernel: stack pointer
= 0x10:0xf1782c9c
Jun 7 05:05:49 10_10_101_102 /kernel: frame pointer
= 0x10:0xf1782cd0
Jun 7 05:05:49 10_10_101_102 /kernel: code segment
= base 0x0, limit 0xf, type 0x1b
Jun 7 05:05:49 10_10_101_102 /kernel: = DPL 0, pres
1, def32 1, gran 1
Jun 7 05:05:49 10_10_101_102 /kernel: processor
eflags = interrupt enabled, resume, IOPL = 0
Jun 7 05:05:49 10_10_101_102 /kernel: current process
= 784 (lpfsd)
Jun 7 05:05:49 10_10_101_102 /kernel: interrupt mask
= none
Jun 7 05:05:49 10_10_101_102 /kernel: trap number
= 12
Jun 7 05:05:49 10_10_101_102 /kernel: panic: page
fault
Uptime: 7m6s
I was looking at the core and it the dvp->v_mount
field in frame 6 is corrupted. Haven't seen this b4,
but its quite puzzling to see 3 crashes. Also, the
uptime shows that it happened very quicky.
any ideas whats goin on here ??
thanks
nitin
PS: please cc: me on this as I am not in the mailing
list.
__
Do You Yahoo!?
Yahoo! - Official partner of 2002 FIFA World Cup
http://fifaworldcup.yahoo.com
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
a hopefully simple question
Sorry if I'm asking on the wrong list, but would someone please write a simple program to do something like: beep 200 300 which would make a 200 hz tone for 300 ms? or at least tell me where to look for the information i need? To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Adding diffs to commit-mail on the fly
On Sat, Jun 01, 2002 at 04:17:54AM +0300, Giorgos Keramidas wrote: > The idea was resting at the back of my mind, and tonight I hacked this > Perl script that seems to mostly do what I want, when I pipe a single There's is also the `cvsmail' port that does that. :0 bf * ^Sender:.*owner-(freebsd-)?cvs-(committers|all)@.*FreeBSD.ORG | cvsmail To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: a hopefully simple question
In the last episode (Jun 08), tyler spivey said: > Sorry if I'm asking on the wrong list, but > would someone please write a simple program > to do something like: > beep 200 300 > which would make a 200 hz tone for 300 ms? > or at least tell me where to look for the information i need? man speaker man spkrtest -- Dan Nelson [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Improving GNU make compatibility in BSD make (+ patch)
On Fri, May 31, 2002 at 11:46:03PM -0500, Dan Nelson wrote: > Automake avoids the issue entirely by simply listing the dependencies > itself, so > > > envuidgid: envuidgid.o > > $(CC) $(LDFLAGS) -o $@ $^ > > becomes > > > $(CC) $(LDFLAGS) -o $@ envuidgid.o This is not always possible. If you have a VPATH, then you need make(1) to expand $^ to the full path it used to find the dependancy. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: Improving GNU make compatibility in BSD make (+ patch)
On Thu, May 30, 2002 at 07:42:28PM -0700, Jos Backus wrote: > The GNU make info file says: > > `$^' > The names of all the prerequisites, with spaces between them. For > prerequisites which are archive members, only the member named is > used (*note Archives::). A target has only one prerequisite on > each other file it depends on, no matter how many times each file > is listed as a prerequisite. So if you list a prerequisite more > than once for a target, the value of `$^' contains just one copy > of the name. I tried to add support for this a while back, but never got it working right. Nor did Will who worked on our make file a while. I'll have to see if your patch does everything $^ should do. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
MIB support for network devices in FreeBSD?
Hi all, (pls Cc: me on any response, not subscribed to either list) Can't find any references to this in the archives. What's the status of MIB support for network interfaces in FreeBSD? Is it deprecated, optional, "would be nice"? Reason for asking is that a dockapp I use has stopped displaying real-time stats since a recent upgrade. This is because the author has switched to using the interface MIB to collect statistics in the latest version[1]. Which in itself seems like a nice approach, except that it doesn't display anything for my 802.11 card, or the loopback device. Digging around a bit with grep (in 4.6-RC) shows that only 3 devices in /usr/src/sys/dev seem to have implemented full MIB support, a couple more increment error counters, and ~20 more NIC drivers (including most all of my NICs) don't apparently implement any support for MIB-based counters whatsoever[2]. Anyway, should the author of the utility be advised: a) Nice approach, little premature b) Nice approach, very premature c) Wrong approach, revert to previous if you want it to work with FreeBSD d) Other? Also, I'd be interested to know which MIBs we intend to support, and from which draft/RFC it's drawn from, if anyone happens to know or can point me at some *BSD docs? TIA. Regards, AS [1] He added FreeBSD support to a previously Solaris/Linux-only app, but doesn't apparently have a FreeBSD box to test on himself, asks for feedback on his webpage. [2] It's obviously not entirely that simple, 'coz I can't find any reference to the xl driver incrementing mib counters /either/, yet the monitor dockapp seems to work for the built-in xl device in my laptop... msg34905/pgp0.pgp Description: PGP signature
Re: MIB support for network devices in FreeBSD?
On Sat, 2002-06-08 at 22:01, Andy Sparrow wrote: > > Hi all, > > (pls Cc: me on any response, not subscribed to either list) > > Can't find any references to this in the archives. > > What's the status of MIB support for network interfaces in FreeBSD? Is it > deprecated, optional, "would be nice"? > > Reason for asking is that a dockapp I use has stopped displaying real-time > stats since a recent upgrade. This is because the author has switched to using > the interface MIB to collect statistics in the latest version[1]. > > Which in itself seems like a nice approach, except that it doesn't display > anything for my 802.11 card, or the loopback device. > > Digging around a bit with grep (in 4.6-RC) shows that only 3 devices in > /usr/src/sys/dev seem to have implemented full MIB support, a couple more > increment error counters, and ~20 more NIC drivers (including most all of my > NICs) > don't apparently implement any support for MIB-based counters whatsoever[2]. > > Anyway, should the author of the utility be advised: > > a)Nice approach, little premature > > b)Nice approach, very premature > > c)Wrong approach, revert to previous if you want it to work with FreeBSD > > d)Other? > > > Also, I'd be interested to know which MIBs we intend to support, and from > which draft/RFC it's drawn from, if anyone happens to know or can point me at > some *BSD docs? > > TIA. > > Regards, > > AS > > [1] He added FreeBSD support to a previously Solaris/Linux-only app, but > doesn't apparently have a FreeBSD box to test on himself, asks for feedback on > his webpage. > > [2] It's obviously not entirely that simple, 'coz I can't find any reference > to the xl driver incrementing mib counters /either/, yet the monitor dockapp > seems to work for the built-in xl device in my laptop... > > > I believe you can get this info if you add the net/snmp or net/snmp4 port. LER -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 972-414-9812 E-Mail: [EMAIL PROTECTED] US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
setrlimit and large maxssiz
I am not sure which is the right mailing list, so sorry about the
cross-posting:
I want to use a lot of memory in my program, so I set the following in
/boot/loader.conf:
kern.maxdsiz=2147483648
kern.maxssiz=2147483648
kern.dfldsiz=2147483648
Then I run this simple program:
#include
#include
#include
#include
int main() {
struct rlimit rlp;
getrlimit(RLIMIT_STACK,&rlp);
fprintf(stderr,"%lld %lld\n",rlp.rlim_cur,rlp.rlim_max);
rlp.rlim_cur = 512*1024*1024;
fprintf(stderr,"%lld %lld\n",rlp.rlim_cur,rlp.rlim_max);
setrlimit(RLIMIT_STACK,&rlp);
exit(0);
}
and it crashes like this:
2147483648 2147483648
536870912 2147483648
Bus error (core dumped)
Maybe I am expecting too much from the system. I have a dual Athlon MP
system with about 3G of RAM, and I want to be able to use a good portion
of this RAM in a single process. But I also want to use linuxthreads so
that I can take advantage of the two processors. But linuxthreads uses
setrlimit, and crashes in a similar way to my simple program.
I looked at the kernel code in /usr/src/sys/kern/kern_resource.c at the
function dosetrlimit, which I guess is where the action takes place, but
I have no idea what to make of it.
--
Stephen Montgomery-Smith
[EMAIL PROTECTED]
http://www.math.missouri.edu/~stephen
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
Re: MIB support for network devices in FreeBSD?
> I believe you can get this info if you add the net/snmp or net/snmp4
> port.
Hi Larry,
Thanks for replying.
Hmmm. I'm talking about code that uses a FreeBSD-specific sysctl to
interrogate the in-kernel if MIB counters, like this:
/* gather stats */
int
freebsd_sysctl_get(struct Devices*dev, unsigned long* ip,
unsigned long* op, unsigned long* ib, unsigned long* ob)
{
struct ifmibdata* drvdata = dev->drvdata;
int datamib[6];
int len;
*ip = *op = *ib = *ob = 0;
datamib[0] = CTL_NET;
datamib[1] = PF_LINK;
datamib[2] = NETLINK_GENERIC;
datamib[3] = IFMIB_IFDATA;
datamib[4] = 1; /* fill in later */
datamib[5] = IFDATA_GENERAL;
datamib[4] = IFMIB_IFCOUNT;
len = sizeof(struct ifmibdata);
if(sysctl(datamib, 6, drvdata, &len, NULL, 0) < 0)
return 1;
*ip = drvdata->ifmd_data.ifi_ipackets;
*op = drvdata->ifmd_data.ifi_opackets;
*ib = drvdata->ifmd_data.ifi_ibytes;
*ob = drvdata->ifmd_data.ifi_obytes;
return 0;
}
But these stats don't seem to be collected for at least some network card
drivers, presumably because those drivers aren't collecting those stats, e.g.
they don't #include , and thus don't allocate a mib structure or
increment any counters in that structure.
I can confirm that it definately doesn't work for the 'wi' and 'lo' drivers...
However, it definately seems to work for the xl driver...
Try this:
cd /usr/src/sys/dev; find . -exec grep mib {} /dev/null \;
'awi', 'ed' & 'ray' drivers seem to have the most complete implementations,
but 'fe' & 'xe' seem to have partial implementations (error counters only).
They're relatively short, so here's one by way of illustration:
./xe/if_xe.c:#include
./xe/if_xe.c:scp->ifp->if_linkmib = &scp->mibdata;
./xe/if_xe.c:scp->ifp->if_linkmiblen = sizeof scp->mibdata;
./xe/if_xe.c: scp->mibdata.dot3StatsSingleCollisionFrames++;
./xe/if_xe.c: scp->mibdata.dot3StatsCollFrequencies[0]++;
./xe/if_xe.c: scp->mibdata.dot3StatsMultipleCollisionFrames++;
./xe/if_xe.c: scp->mibdata.dot3StatsCollFrequencies[scp->
tx_collisions-1]++;
./xe/if_xe.c: scp->mibdata.dot3StatsMultipleCollisionFrames += sent;
./xe/if_xe.c: scp->mibdata.dot3StatsExcessiveCollisions++;
./xe/if_xe.c: scp->mibdata.dot3StatsMultipleCollisionFrames++;
./xe/if_xe.c: scp->mibdata.dot3StatsCollFrequencies[15]++;
./xe/if_xe.c: scp->mibData.dot3StatsMissedFrames++;
./xe/if_xe.c: scp->mibdata.dot3StatsFrameTooLongs++;
./xe/if_xe.c: scp->mibdata.dot3StatsFCSErrors++;
./xe/if_xe.c: scp->mibdata.dot3StatsAlignmentErrors++;
./xe/if_xe.c: scp->mibdata.dot3StatsInternalMacReceiveErrors++;
./xe/if_xevar.h: struct ifmib_iso_8802_3 mibdata;
However, most of the drivers don't seem to have any code like this in them -
and at least some of those drivers don't work with the 3rd-party application
code above.
Should they? Or is this an older interface?
I'd like to advise the author so he can fix a very useful dockapp for me - or,
if this is the way to go, I'll badger Bill Paul to fix the 'wi' driver - oh,
wait a moment, maybe generating patches would be safer :)
Regards,
AS
msg34908/pgp0.pgp
Description: PGP signature
Re: MIB support for network devices in FreeBSD?
On Sat, 2002-06-08 at 22:51, Andy Sparrow wrote: > > > I believe you can get this info if you add the net/snmp or net/snmp4 > > port. > > Hi Larry, > > Thanks for replying. > > Hmmm. I'm talking about code that uses a FreeBSD-specific sysctl to > interrogate the in-kernel if MIB counters, like this: > [snip] > > However, most of the drivers don't seem to have any code like this in them - > and at least some of those drivers don't work with the 3rd-party application > code above. Hmm. I'm not a network developer, and have even less knowledge of what's going on in the interfaces stuff. > > Should they? Or is this an older interface? Haven't the foggiest. Have you looked in -CURRENT to see if it's different there? (and if so, see if it's planned to MFC it?) > > I'd like to advise the author so he can fix a very useful dockapp for me - or, > if this is the way to go, I'll badger Bill Paul to fix the 'wi' driver - oh, > wait a moment, maybe generating patches would be safer :) Probably Patches would work better, if their not in -CURRENT yet. But, I'm just a luser here. LER -- Larry Rosenman http://www.lerctr.org/~ler Phone: +1 972-414-9812 E-Mail: [EMAIL PROTECTED] US Mail: 1905 Steamboat Springs Drive, Garland, TX 75044-6749 To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
Re: MIB support for network devices in FreeBSD?
Andy Sparrow wrote: > But these stats don't seem to be collected for at least some network card > drivers, presumably because those drivers aren't collecting those stats, e.g. > they don't #include , and thus don't allocate a mib structure or > increment any counters in that structure. > > I can confirm that it definately doesn't work for the 'wi' and 'lo' drivers... > > However, it definately seems to work for the xl driver... Benchmark the driver. If it's fast, it doesn't collect the statistics. Just like "WITNESS": if your machine is fast, "WITNESS" is disabled. 8-). Bascially, most of these statistics are "manager fluff" (as opposed to "seat cushion fluuf" or "pocket fluff"). What you want is statistics that are managed universally; see "netstat -ibdt" for the usefult statistics kept per interface. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-hackers" in the body of the message
