nd, and everything goes as fast as sequential reads.
I gave the example with a very slow device, but the effect is quite
noticeable even with internal hard drives. Of course, SSD drives do not
cause this problem.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
to lower the frequency. But
if your laptop is very old, it probably can not do that; in that case,
"killall -STOP ffmpeg" would probably work.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
h stable names.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
the block device, not a corruption of the
filesystem: fsck will not really help you.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le sextidi 6 thermidor, an CCXXIII, Erwan David a écrit :
> I thiink you should dd the .iso to the stick, not cp
Both work, and dd will give terrible performance without extra options.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
mputer is on the
UEFI boot menu or the GRUB menu.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ted in BIOS compatibility ("lecacy") mode rather than with a UEFI
bootloader.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
e as the external reference clock as is done by both ntp and
> ntpdate, approx computes an estimate of *rate* of ticking of the local
> clock.
Unless I am mistaken, nowadays the kernel does that by itself, provided the
NTP client uses adtime() or a variant and not a brutal settimeofday().
Regards
UB instead of
LILO and be done with that kind of trouble.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le duodi 2 nivôse, an CCXXIV, John Hasler a écrit :
> "Unused" is the problem with this design. The space is unused on by
> accident. There is no standard (even an informal one) that reserves it.
This is theoretical. In practice, it works pretty well, because people (at
least competent sysadmins
Le duodi 2 nivôse, an CCXXIV, John Hasler a écrit :
> Think about multiboot home user systems.
Multiboot or no multiboot, once the system(s) is/are installed, nobody
writes outside the partitions.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ecause they have decided they have the right to
do so, or because they have a bug.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ng half of the
> RAID, let the RAID resync, repeated the whole procedure with the other
> disk, installed GRUB to its new home, rebooted while hoping/praying and
> ... tadaaa ... the system came right up.
The same thing would have worked with without GPT. Especially if you use a
b
egacy BIOS to boot.
Then you confirm what I wrote: you could have done exactly the same with
legacy MBR partitions, GPT was not needed for that operation.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le septidi 7 nivôse, an CCXXIV, Pascal Hambourg a écrit :
> You are confusing GPT (partition table format) and UEFI (firmware and
> boot interface).
Quite the contrary, I am trying to de-confuse by using exactly the correct
terms.
> Can you tell more ? AFAIK, there is no equivalent to the "BIOS b
Le septidi 7 nivôse, an CCXXIV, Pascal Hambourg a écrit :
> There was nothing confusing in Sven's message until you mentionned UEFI
> in response to "I love the GPT". Why did you start talking about UEFI ?
For the early stage of booting, there is no difference between MBR-style
partitions (it is m
Le septidi 7 nivôse, an CCXXIV, Pascal Hambourg a écrit :
> *Cough*
> At least it was supposed to. The more I mess with UEFI implementations,
> the more I feel dubious about its real advantages.
Six messages ago, I wrote "or would be if it was correctly implemented by
vendors"; you read it since y
Le septidi 7 nivôse, an CCXXIV, Pascal Hambourg a écrit :
> I repeat : the point is not MBR/MSDOS vs GPT. It is that the GRUB
> installer handles *specially* one type of partition which exists only in
> GPT style : "BIOS Boot" aka "bios_grub".
>
> Are you talking about the installer code run by gr
the state of hardware support we
had in the early 2000s, with a few differences.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
s for security.
> unless there's a way to
> protect against that in the ssh configuration file, which I did look
> for but have not found.
Search for "chroot" in sshd_config(5). Also, search the web for "chroot
sftp".
Regards,
--
Nic
rds,
--
Nicolas George
signature.asc
Description: Digital signature
for x264.
There are no such relations between audio and video codec. The only
considerations are quality and compatibility with the intended player(s).
> Check the video vs audio bitrate - try with fixed bitrate
This is not relevant, frame dropping or duplicating is done before the
geometry option in the X11 resources. You can try to
look at the output of xrdb -query, and compare to what happens on a blank
X11 server.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
tput of xrdb -query after removing my custom .Xresources is:
> *customization: -color
> Xpdf*fileFilterStyle: filter_hidden_files
Then it does not seem to be the cause of the problem.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
an use the resources of a remote computer without
actually annoying its console user by moving the mouse cursor for them, that
would make them try to put the red queen on the black ten, and this is a big
no, with a risk of BSOD.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ithout breaking the completion of globs.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
conversion is expensive.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ne with the same MD5.
Of course, that does not mean MD5 and SHA-1 should be used nowadays. New
theoretical attacks are found, keeping using hashes with known weaknesses is
stupid. And of course, to avoid malicious tampering, cryptographic
signatures would be much better than plain hashes.
rotection.
Signing hashes will get you a spanking from any cryptographer.
Cryptographic signatures must be applied on the file itself; it works
internally by signing a hash of the file, but the hash is done in a way that
prevents most attacks even with weak hashes.
Regards,
--
Nicolas George
sign
Le quintidi 5 ventôse, an CCXXIV, Thomas Schmitt a écrit :
> i wrote:
> > > The ISO checksums are provided more for transport verification than
> > > for the fight against intentional mainpulation.
> Nicolas George wrote:
> > If that were true, CRC32 would be
And of course, all the hype nowadays is about quantum-resistant
algorithms.
> I could imagine that PGP is easier to surpass than that.
It is not a matter of surpassing anything. Cryptographic signatures protect
against all the same attacks as hashes, plus a whole bunch of new attacks.
Regar
Le quintidi 5 ventôse, an CCXXIV, David Wright a écrit :
> Any faster ones that you recommend from the lists below? (I've rolled
> my own implementation of fdupes (which uses MD5) in python.)
Nobody can recommend anything without knowing the intended use.
Regards,
--
Nic
Le quintidi 5 ventôse, an CCXXIV, Thomas Schmitt a écrit :
> Only as far as use cases for Debian ISO image hashs are concerned.
> No hash collisions among all Debian ISOs (or better all ISOs in the
> world) is a valuable property.
??? I have no idea what you are talking about.
> If the SHA512SUMS
Le quintidi 5 ventôse, an CCXXIV, David Wright a écrit :
> 1) I do what fdupes does, ie identify files (in a benevolent
>environment) using the MD5 signature to detect duplicate
>contents.
You did not specify the average size of files nor how sure you want to be.
If the files are large, I
Le quintidi 5 ventôse, an CCXXIV, Seeker a écrit :
> If you take security out of the equation, simple true or false.
>
> 1. A corrupted download is better able to be detected when using MD5 than it
> is with CRC32.
>
> 2. A corrupted download is better able to be detected when using SHA than it
>
lisions for SHA1 as soon as the files are larger than 20 bytes.
The same applies to SHA-2 with the corresponding size.
The collisions are not known, and very unlikely, but "absolute" means
absolute, not "very likely".
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le quintidi 5 ventôse, an CCXXIV, Thomas Schmitt a écrit :
> You have unsurpassable objections against variants which might not
> much weaken the strength of PGP ?
> Not even willing to consider the constraints of such variants ?
I have no idea what you are trying to express.
> Despite leading de
his is quite orthogonal to the issues you quoted, which were about
signing not only the files' individual contents but also the files' metadata
and the global information about what files are in the collection. Signing
the hash file protects against that but is wide open for collision attacks.
A proper protocol can protect both.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le sextidi 6 ventôse, an CCXXIV, Richard Hector a écrit :
> Fair enough. Got a link to someone else's explanation?
Sorry, I do not. But I gave a rather lengthy explanation myself in the part
you trimmed.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le quintidi 5 ventôse, an CCXXIV, Christian Seiler a écrit :
> But if you say what Debian is doing is a mistake, then this _is_ what
> you are talking about.
I am quite sure of what I am talking about and what I am not talking about.
> This is decisively not true when we are talking about signing
Le quintidi 5 ventôse, an CCXXIV, David Wright a écrit :
> Well, md5 beats md4
There is something wrong in your library.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ould be better to include parts that the attacker can not control
before the message, not just after.)
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
e: public key not found
>
> Could the exit status of gpg be causing mutt to give up?
In this kind of situation, I use strace (with the -f option, and a large
value for -s) in order to know exactly how gpg is invoked, then I invoke it
separately to see the error messages.
HTH.
R
put is a
file, "tar xf" will detect the compression.
Furthermore, dpkg provides tools to handle deb packages:
dpkg --fsys-tarfile /path/to/file.deb | tar x
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
t public humiliation that this will work.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ou did try. Can you copy-paste your
attempt?
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le septidi 7 germinal, an CCXXIV, deloptes a écrit :
> He is mixing up Perl Harbor 1941
Using regexps in a war is fighting dirty for sure.
Regards,
--
Nicolas George
quested optional modules.
You can reproduce the same process invoking grub-mkimage explicitly.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le quartidi 4 floréal, an CCXXIV, The Wanderer a écrit :
> This means that the mailing-list software _could_ technically work
> around this behavior by modifying the Message-ID of the received message
> before it sends that message out to list members.
>
> That strikes me as a dreadful idea from a
Le quartidi 4 floréal, an CCXXIV, The Wanderer a écrit :
> If the list software did this modification for _all_ messages, not just
> ones from Gmail addresses, I don't see how it would break threading
It breaks it for the *senders*: they would have the message in their "sent"
archive with the mess
Le quartidi 4 floréal, an CCXXIV, The Wanderer a écrit :
> I'm not sure this would actually manifest the way you're describing it -
> but I'm not positive it wouldn't, either, and I don't care to invest the
> brainpower in working it out for certain right now.
Actually, it is even worse than I sug
that you want to receive but never archive (for example notifications),
since you may never teach the filter to recognize them as ham.
OTOH, feeding enough ham is necessary to avoid false-positives. This
sub-thread was about false-negatives, IIRC.
Regards,
--
Nicolas George
signature.
ther similar libraries are off on version number.
>
> How does one handle such an issue?
Rebuild from source. The two versions are source-compatible.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
card is a Marvell
> 88E8056 is that it only works in 32?
What makes you say that? A quick look in the logs of the kernel shows that
this card is supported natively on all relevant architectures since long
ago (~may 2007).
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
ttp://www.normalesup.org/~george/comp/live_iso_usb/
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le primidi 11 floréal, an CCXXIV, Richard Owlett a écrit :
> I'll have to try it to understand what author means by saying "Requires
> cooperation from the distribution: as any procedure in fact." under cons.
It is the same principle as software RAID or LVM : the kernel has low-level
support for t
Le primidi 11 floréal, an CCXXIV, Brian a écrit :
> *All* Debian live images (the focus of this thread) are bootable from a
> USB stick by GRUB.
Are you sure you are talking about the same thing as me? I am talking about
booting the image copied as a file on the stick, not copied raw on the
stick.
Le primidi 11 floréal, an CCXXIV, Brian a écrit :
> I'm positive. :) Please see the final portion of Piyavkin's mail.
Then that is good news.
signature.asc
Description: Digital signature
egards,
--
Nicolas George
stemctl start postfix
> # systemctl status postfix
>
> and got several lines basically saying posfix.service was disabled.
Maybe reading these several lines in details rather than stopping at the
"basic" meaning would give more information.
Regards,
--
Nicolas George
c
>`-19868 qmgr -l -t unix -u
Looks like Postfix is running correctly.
Note that "systemctl start" is transient. If you want to make it start
automatically at boot, you have to use "systemctl enable".
Regards,
--
Nicolas George
is a mystery to me.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le tridi 3 fructidor, an CCXXV, Tom Browder a écrit :
> So "disabled" is normal?
Indeed. See:
https://docs.fedoraproject.org/en-US/Fedora/15/html/Deployment_Guide/ch-Services_and_Daemons.html#s3-services-configuration-enabling
Regards,
--
Nicolas George
signature.asc
Descrip
Le quintidi 5 fructidor, an CCXXV, Mario Castelán Castro a écrit :
> Wireless things do not solve the problem of having to cope with wires.
> They just replace this with the bigger problem of unauduitable firmware
> directly exposed to the attacker (via radio or sometimes infrared
> communication).
the police after robbing a bank" are two different problems,
for example.
The real question is, of course, which problem is the most severe for
the person who decides.
Regards,
--
Nicolas George
Le quintidi 5 fructidor, an CCXXV, Mario Castelán Castro a écrit :
> Wrong. Your ambiguous choice of words has hidden the difference.
That was YOUR own choice of words, showing how this discussion is
pointless.
Regards,
--
Nicolas George
Le nonidi 9 fructidor, an CCXXV, Dejan Jocic a écrit :
> 10. I'm sure that there is more
0. Think about against what risks you want to protect yourself.
Security is always a compromise with convenience. The only absolute
security is when you do nothing with no computer at all, but that is not
wha
The same thing with Perl's
Digest::b64digest would be much better.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le primidi 21 fructidor, an CCXXV, Pol Hallen a écrit :
> cat /tmp/file
Could we perhaps see the contents of this file?
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
Le primidi 21 fructidor, an CCXXV, Max a écrit :
> cat /tmp/file
>
> one
> two
> three
> four
This is not HTML, yet you pretend it is. Make it consistent. It has
nothing to do with Bash nor CGI.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
sing it means a possible ellipsis.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
uld have the new episode #n
immediately, followed by "previously on" and a complete rerun of episode
#(n-1), including the rerun of episode #(n-2) that was at its end.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
for example when it provides a plugin
for another software, it does so by providing a file. When the package
is upgraded, the normal handling of configuration files can be applied.
Modifying a monolithic configuration file would be much more fragile.
Regards,
--
Nicolas George
signature
rhkra...@gmail.com (2018-01-03):
> I hope the OP is still "listening".
If he is, he is probably enjoying the time wasted by his nonsensical
question.
Regards,
--
Nicolas George
signature.asc
Description: Digital signature
. And they have detection: please only do this on a computer
and network access when you will be the only one inconvenienced when
they block your access. It happened on a computer I co-administrate.
Regards,
--
Nicolas George
describes breaks Google's terms of use is the
same.
Now, if your interrogation is whether we think you would get away with
it. Well, as it is roughly equivalent to asking if I think you are
smarter than the people at Google, I will respectfully decline to
answer.
Regards,
--
Nicolas George
Michel Verdier (12023-09-24):
> This one is easy : bash read /etc/bash.bashrc and ~/.bashrc.
Only for interactive shells. There are no interactive shells in the
ancestry of your desktop environment if you log from a display manager.
If it was so easy, the OP would have found it.
--
Nico
es that will be of use for you are “bridge_ports none”
and “post-up ip link set dev wlan* master”.
Regards,
--
Nicolas George
an
use rsync or anything I want to do the transfer. Add Tinc to get a
stable reachable IP.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
of existing packages in Debian that could make my
work easier?
Thanks in advance.
--
Nicolas George
john doe (12023-10-06):
> I do not understand why you would want multiple repos, to me this looks
> like this would fit the bill for a Git branching workflow.
Please elaborate. How do you work around the fact that Git is terrible
at removing data with a single repository?
Regards,
--
N
. They only allow to keep the last commits, not decimate them.
> --interactive" it is possible to squash e.g. daily commits into weekly or
> monthly ones. The drawback is that git rebase changes commit hashes.
git rebase is too inefficient for that kind of use.
Regards,
--
Nicolas George
h, the original data is
not actually deleted.
I asked a question and somehow it has become my responsibility to
explain things…
Regards,
--
Nicolas George
am looking for.
Regards,
--
Nicolas George
etwork would get
de-configured if somebody pulled the ethernet cable (and that happens a
lot here!) and rebooted the machine while they were at it, and the NIS
and NFS do not appreciate it much.
I switched to ifupdown and everything worked nicely.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
somebody give me the hints that will save me some times in
trial and error to get my head around the logic?
Thanks in advance.
--
Nicolas George
signature.asc
Description: PGP signature
Maybe some parameter for diff or rsync that I missed or another utility?
I doubt it, it is too specific a use case.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
Max Nikulin (12023-11-01):
> I am curious if ConfigureWithoutCarrier= and IgnoreCarrierLoss= from
> systemd.network(5) may help.
>
> NetworkManager has a similar option, unfortunately it may be changed
> per-device, not per-connection.
I do not have the time to test it anew. IIRC, when I tried, I
y...@vienna.at (12023-11-02):
> Why not try ARMbian?
I ear with ARMbian you have to use Docker to build a kernel. That makes
a pretty strong “why not ARMbian”.
Regards,
--
Nicolas George
s and who knows what else.
More probably because you tried to use it on a non-Unix filesystem (or
even non-Unix system).
Regards,
--
Nicolas George
y...@vienna.at (12023-11-02):
> > I ear with ARMbian you have to use Docker to build a kernel. That makes
> > a pretty strong “why not ARMbian”.
> No
No what? No Docker is not necessary or no it is not a reason to ditch
armbian?
--
Nicolas George
d “folder” belongs in the “computers are too complicated
for you so we'll make them guess what you want to do and pretend they're
easy but it's normal if they crash” mentality where we cannot get the
computer to do what you want because it is second-guessing our
instructions: vague and not suited for technical communication.
Regards,
--
Nicolas George
Alexander V. Makartsev (12023-11-02):
> It could be also a limitation or bug of overlayfs since it doesn't have
> locale/iocharset/codepage parameters for mount.
No, it is not: Unix file names are made of octets, not characters.
Regards,
--
Nicolas George
directories are: they do not
contain anything except a list of names and numbers.
Folder is the metaphor, directory is the fact.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
sical
world, I have *sometimes* seen a folder inside a folder, but a folder
inside a folder inside a folder inside a folder…
Regards,
--
Nicolas George
Alexander V. Makartsev (12023-11-03):
> Personally, I don't see the problem, because I was talking to people not
> Unix user interfaces.
You talk TO people ABOUT Unix user interface.
--
Nicolas George
people who are dumbing down are the same as people who
are using the metaphoric word instead of the precise one.
--
Nicolas George
ul for performances, but when it comes to I/O, they
them harder.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
erfect solution.
So all we have to do is trust that the original poster has made the
cost-benefit analysis according to one's own priorities and not try to
convince otherwise.
Regards,
--
Nicolas George
signature.asc
Description: PGP signature
al means”. Anyway, I am not interested
in splitting hair about stylistic devices.
--
Nicolas George
801 - 900 of 1329 matches
Mail list logo
