On 18-Jan-2005 Chris wrote:
> I have been compiling ports using -O2 since I started using FreeBSD
> back in 2003 and only port that has had issues with this is lang/ezm3
> in FreeBSD 5.2.1 it needed -O.
I fixed that at the beginning of November.
John
__
On 30-Sep-2003 Damon Anton Permezel wrote:
> Recent SUP.
> Installed on Dell PowerEdge 4600.
> Getting tons of:
>
> Sep 30 12:23:16 zige /kernel: bge0: gigabit link up
> Sep 30 12:23:46 zige last message repeated 98 times
Check your dmesg output, and I bet you'll see that bge0 is sharing an
IRQ w
0 is reconized as valid, any device number above that fails
> to be recognized as a configured device.
It's a bug that I fixed in the -stable branch on 19 August. If you
update "src/sys/dev/md/md.c" to revision 1.8.2.2 and rebuild your
kernel, your problem should go
Broadcom chips? In particular: what kind of CPU,
and was the bus PCI or PCI-X?
John
--
John Polstra
John D. Polstra & Co., Inc.Seattle, Washington USA
"Disappointment is a good sign of basic intelligence." -- Chögyam Trungpa
To Unsubscribe: send mai
find the address of its maintainer in the FreeBSD Handbook
(section A.6.7 currently) and let him know.
Thanks,
John
--
John Polstra
John D. Polstra & Co., Inc.Seattle, Washington USA
"Disappointment is a good sign of basic intelligence." -- Chögyam Trungpa
oks like something is wrong with its src-all
> mirror.
Could you please report this to cvsup14's maintainer,
<[EMAIL PROTECTED]>?
By the way, all of the CVSup mirror site maintainers are listed in the
FreeBSD Handbook here:
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/cvsup.
used by a non-corrected bit error in memory
> (cosmic ray at an inopportune time while the rtld was doing its
> thing?).
Yes, I'm betting it's that or a corrupted file.
John
--
John Polstra
John D. Polstra & Co., Inc.Seattle, Washington USA
&qu
irror?
I removed cvsup2 from the list of mirrors over a month ago, but I
forgot to have its DNS record deleted. I have asked our hostmaster to
do that now.
We'll have a replacement for cvsup2 soon, maybe even this week.
John
--
John Polstra
John D. Polstra & Co., Inc.
In article <[EMAIL PROTECTED]>,
Maxime Romano <[EMAIL PROTECTED]> wrote:
> I've been getting the exact same problems at three diffrent locations. You
> should use cvsup.ca.freebsd.org.
Did that actually fix it for you? If so, cvsup.ca.freebsd.org is
seriously screw
In article <[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
> Try another mirror
No, that's not the problem. It's on the user's end.
John
--
John Polstra
John D. Polstra & Co., Inc.Seattle, Washington USA
"Disappointme
rst, the
supfile specifies a prefix of "/usr", but the error messages
indicate that the prefix is really "prefixes/FreeBSD.cvs". Second,
it is complaining about "*,v" files, which you wouldn't be getting
if you had "tag=." in your supfile.
This looks
ill pumping out
> source with the S1G bug. A cvsup started at 9:25 PDT shows
>
> Connected to cvsup4.FreeBSD.org
> Server software version: REL_16_1
Confirmed. I have removed this site from the list in the Handbook,
and have asked the freebsd.org hostmaster to delete its
vsup bug makes the dates come out 31 Dec 69 instead of 10 Sep
> > 01.
> >
> If this "69" is 2069, that is definitely the problem!
It was a pretty strange bug, but I don't think it would have been able
to set the modtime to anything except the epoch+0, i.e., 1970-01-
quiring Modula's
> additional packages to exist in the system? How to include these
> modules 'static'?
You can build the port statically-linked by setting "STATIC=yes" in
the environment before building.
John
--
John Polstra
have any
way myself to build binaries for -3.x. However, you can build the
port from source.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington USA
"Disappointment is a good sign
the section REFUSE
FILES in cvsup(1).) If so, please post it. Also please post your
supfile. I'm sure there's a simple explanation for the problem.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.
he way it is
configured. I doubt that any of us can help solve this problem. Only
your firewall administrator can fix it.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington U
he command line. If you don't, it will
eventually die with a "Bad address" error.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington USA
"Disappointment is a
.) The
ports tree is relatively small, so it won't take too awfully long.
Or, if you are careful and know what you're doing, you can probably
get things working by doing a systematic manual deletion of all
subdirectories named "patches" and "pkg" in your
vision 1.82 is not
a part of -stable; it is in -current only. -stable was branched at
revision 1.81, before revision 1.82 was created. For -stable, the
revision after 1.81 is 1.81.2.1. Next will come 1.81.2.2, etc.
> I wonder how many other things a
0600a8.
> #0 0x1804f358 in ?? ()Cannot access memory at address 0x180600a8.
>
> Anything I did wrong?
No, I forgot a step. :-) When you run yppush, set LD_BIND_NOW=1 in the
environment. (And I _hope_ that won't make it stop failing.)
John
--
John Polstra
file in one of the following places?
/usr/sup/refuse
/usr/sup/src-all/refuse
/usr/sup/src-all/refuse.cvs:RELENG_4
Are you sure that the server host you are using is keeping its
src-all collection up-to-date?
John
--
John Polstra
Chad R. Larson wrote:
>
> I'm assuming none of this affects those of us who are using CVSup to
> maintain a local copy of the repository, and using cvs for our local
> ports tree.
Correct.
John
--
John Polstra [EMAIL PROTECTED]
blem, that's definitely true.
Unfortunately, the fix for that problem provoked a completely
separate bug in CVSup. So cvs doesn't deserve all the blame. :-(
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.
g around for that. The real fix is going
to be in CVSup, and it's going to take me some time. Sorry, that's
just life. It's Monday, I'm barely recovering from the flu, and
I'm swamped with work from my real jobs. It'll be a while. So the
work-around is your bes
e refuse file to ignore
the jakarta-tomcat port. That probably hides the problem, but it
doesn't really fix anything. The breakage is still there in your
checkouts file. Therefore I don't recommend using the refuse file as
a work-around.
John
--
John Polstra
he information you've
provided doesn't tell me much. Could you please fire up the "script"
program, run through the above commands again, and send me the
"typescript" file that gets created? Also please send me the
relevant variables from your /etc/make.conf file
ot;The pattern is a standard file name pattern.
It is interpreted relative to the collection's prefix directory." So
you should have used "-i src/contrib/ntp/ntpd".
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., In
In article <[EMAIL PROTECTED]>,
Eriya Akasaka <[EMAIL PROTECTED]> wrote:
> /etc/fbtab ?
>
> >>>>> In article <[EMAIL PROTECTED]>,
> On Mon, 02 Apr 2001 18:49:54 -0700 (PDT)
> "xconsole broken in recent 4.3-RC?"
>
in PIO
mode. It also panics the same way in DMA mode.
The destination filesystem has soft-updates turned on. I have never
seen any problems with this machine's SCSI subsystem.
The dmesg output and kernel config file are at the end of this mail.
John
--
John Polstra
quot;-l" option in the cvsup man page to keep this from happening.
Or run it under lockf(1).
> Sorry if these are common questions - I searched the archives but
> couldn't find anything.
Searching the archives is fine. But don't forget, there's a great
big manual page t
ork pressures I may not find
time soon.
In any case, this appears to be a genuine bug that we (FreeBSD)
introduced.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington USA
&
apply to subsequent lines
in the supfile. Multiple *default lines may be present. New
values augment or override any defaults specified earlier in the
supfile. Values specified explicitly for a collection override
any default values.
The only thing th
in the mailing lists when we switched
to ELF, so pardon me if I refuse to let myself get drawn into another
long-winded rehash of it.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington
In article <[EMAIL PROTECTED]>,
Satoshi - Ports Wraith - Asami <[EMAIL PROTECTED]> wrote:
> * From: John Polstra <[EMAIL PROTECTED]>
>
> * If it contains a new libc, that seems like the real problem
> * to me. It's always risky to use new libs
In article <[EMAIL PROTECTED]>,
Satoshi - Ports Wraith - Asami <[EMAIL PROTECTED]> wrote:
> * From: John Polstra <[EMAIL PROTECTED]>
>
> * In article <[EMAIL PROTECTED]>,
> * Satoshi Asami <[EMAIL PROTECTED]> wrote:
> *
> * > It was brou
ity.
This isn't much to go on. What are the symptoms?
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington USA
"Disappointment is a good sign of basic intelligence.&quo
dgram ?
Userland and kernel are in sync. This is on an Alpha, though I
don't know whether that's significant or not. Is anybody else
seeing this problem?
John
--
John Polstra [EMAIL PROTECTED]
John D. Pol
ndefined reference to `setresuid'
> pp_hot.o(.text+0x16d4): undefined reference to `setresgid'
> *** Error code 1
Yep, I got the exact same problem here.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co.,
> (Or: cvsup7.freebsd.org has the tag missing)
It's not cvsup7's problem. The tag is missing on freefall.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington USA
&qu
ind one?
Sure. But even the work-around mentioned in the man page doesn't
really solve the whole problem. I think it has something to do with
M3's use of the WM_TAKE_FOCUS window manager protocol.
John
--
John Polstra [EMAIL PROTECTED]
g to take longer, and it would take longer even if you had the
server all to yourself.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington USA
"Disappointment is a good sign of ba
let me tell you that I don't think
the cost of the MD5 calculations is at all significant. In my CVSup
profiling tests the MD5 calculation has never even made a blip on the
radar screen. Calculating an MD5 signature is very cheap. It takes
less CPU time than t
your repository, write the edited version to a
temp file, check the MD5 signature, and then move the temp file to the
right place again.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, W
nge once he realised it was not the correct way :-(
It has been fixed now. The file in question is back in its original
collection, and the src-crypto-rsa collection is gone.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc
allowing your outbound connection to the
CVSup server. Try a few different servers to make sure.
If this is the problem, only your firewall administrator can help
you solve it.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra &
nothing works if it is broken. (In particular, you won't be
able to run the compiler to build a new one.) Make a backup copy
named something like "/usr/libexec/ld-elf.so.1.good" first, so you can
"mv" it back into place if something goes wrong. The basic utilities
In article <[EMAIL PROTECTED]>,
Max Khon <[EMAIL PROTECTED]> wrote:
>
> the fix was MFC'ed on Jan 28.
> make sure that you have src/libexec/rtld-elf/lockdflt.c,v 1.3.2.2
Oops, I just realized my previous reply was wrong. You are right
about JDK -- the fix in lockdflt.c should make it work. Th
In article <[EMAIL PROTECTED]>,
Max Khon <[EMAIL PROTECTED]> wrote:
> hi, there!
>
> On 2 Feb 2000, Soren Dayton wrote:
>
> > Whenever using jdk 1.1.8 (from ports), I find that I get:
> >
> > ld-elf.so.1: assert failed: /usr/src/libexec/rtld-elf/lockdflt.c:55
> > ld-elf.so.1: assert failed: /u
e:p6 /home/bob> host umd1.umd.edu
> umd1.umd.edu is a nickname for haven.umd.edu
> haven.umd.edu has address 128.8.10.6
> haven.umd.edu has address 128.8.10.6
Yep, I just noticed the same thing yesterday. It's broken, I'd say.
(Same story in -current).
John
--
John Pols
Vadim Belman wrote:
> >> Makes sense. I was just going by the examples in
> >> /etc{/defaults}/make.conf. If COPY is more correct, should
> >> /etc{/defaults}/make.conf be changed?
>
> John> Yes, I think so.
>
> Wouldn't it slow down the installworld?
No, it wouldn't change anything
ut it on the
collection line:
src-all umask=077
John
---
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.Seattle, Washington USA
"No matter how cynical I get, I just can't keep up."
hile other systems act normally
> (loading delta's and other stuff)...
Make sure you run cvsup as the same user and with the same umask
each time. You can specify the umask in the supfile if you want to
be sure about that.
John
--
John Polstra
Brad Knowles wrote:
>
> In -CURRENT, I would say that this could probably be committed,
> if John feels safe. I am not yet convinced that it should be
> committed to -STABLE, although things do look good so far.
Just to clarify, I committed it to -current already this morning.
John
T
a working FreeBSD system,
or worse? Have you tried repeating the problem with other varieties
of wine? We are here to help you, but you'll have to do your part.
John
--
John Polstra [EMAIL PROTECTED]
John D. Polstra & Co., Inc.
In article <[EMAIL PROTECTED]>,
Gary Palmer <[EMAIL PROTECTED]> wrote:
>
> Since best.com got assimilated, cvsup2.freebsd.org (aka burka.rdy.com)
> is off verio.net.
The trouble with cvsup2 is that it has a limit of only 6 clients at
a time. It's usually maxxed out
Frank Mayhar wrote:
> John Polstra wrote:
>>
>> Use "make upgrade" to get from 2.2.8-stable to 3.0-release. Then
>> use "make world" to get from 3.0-release to 3.x-stable.
>
> "Make upgrade" doesn't exist in 3.0-release. The
e tip on the cited website; I'll
> check it out.
Use "make upgrade" to get from 2.2.8-stable to 3.0-release. Then
use "make world" to get from 3.0-release to 3.x-stable.
John
--
John Polstra [EMAIL PROTECTED]
John D
58 matches
Mail list logo
