Re: questions about cron.daily

[ken]

On 2023-04-07 05:20, davidson wrote:

 25 6* * *   roottest -x /usr/sbin/anacron || ( cd / && 
run-parts --report /etc/cron.daily )


Are the time format in /etc/crontab just random? why they are 6:25, 6:47 
etc?



--
Ken Peng
https://kenpeng.pages.dev/

Re: my immature thoughts on perl

[Ken Peng]

to...@tuxteam.de wrote:

I think watching carefully Tcl's evolution teaches a lot about
languages, the type of design decisions going into them and their
(changing) context.


I was surprised to see many people here still use TCL.
Many years ago I used this language for sysadmin jobs.
It was working just fine.


--
https://kenpeng.pages.dev/

Re: Is perl still the No.1 language for sysadmin?

[Ken Young]

Our production system is using heavily perl (many thousand lines of perl5 code) 
- it's mod_perl, but still perl.

Thanks.

service vs systemctl

[Ken Young]

Hello

For debian 11, service is just a wrapper to systemctl, is it right?
So for server management, both commands below have the same results.

sudo service nginx start
sudo systemclt start nginx

Which way is more prefered though?
Thanks.

-- 
Sincerely,
Ken Young

Re: what method do you prefer for data transfer between nodes?

[Ken Young]

For standalone backup I am using Liteserver's VPS, 512G disk for only 2.4
EUR/m.
Then I setup rsyncd on it, and rsync data from clients via crontab for
backup purpose.


Sincerely,
Ken Young


On Tue, Mar 7, 2023 at 9:21 AM Celejar  wrote:

> On Sun, 5 Mar 2023 14:05:37 +0100
> to...@tuxteam.de wrote:
>
> > On Sun, Mar 05, 2023 at 07:23:49PM +0800, Ken Young wrote:
> > > What provider do you recommend then?
> >
> > Around here (West Europe), for example, mailbox.org [1]. But they
> > are far from the only one (I'm not a customer, nor associated with
> > them, but I do know a few happy customers).
>
> €3 / month for 5 GB, €9 / month for 50 GB - that's rather overpriced
> for cloud storage:
>
> https://mailbox.org/en/services#tariffs
>
> > Whenever they don't cost any money you'll have to ask yourself
> > what their business model is.
>
> The big cloud providers (e.g. Backblaze) - whose business model does
> indeed involve charging money - charge as little as $5 / month per
> *terabyte* (plus egress), in addition to giving a few or few dozen GB
> free:
>
> https://www.backblaze.com/b2/cloud-storage-pricing.html
>
> I'm personally using Scaleway with rclone, which offers 75 GB free (but
> €12 / month per terabyte beyond that).
>
> Other prominent vendors (besides the obvious ones) include Wasabi and
> rsync.net. Pricing is tricky, as they break the fees into different
> categories (hot and cold storage, egress, etc.)
>
> --
> Celejar
>
>

Re: what method do you prefer for data transfer between nodes?

[Ken Young]

That's smart. Thanks

Sincerely,
Ken Young


On Mon, Mar 6, 2023 at 3:43 AM Linux-Fan  wrote:

> Ken Young writes:
>
> > Hello,[1;5B
> >
> >
> > The methods I know,
> >
> > 1. scp
> > pros: the native tool in the OS
> > cons: you will either input password or put key pairs into servers for
> > authentication.
>
> Works for simple cases.
>
> > 2. rsync
> > pros: it can transfer data by increasement
> > cons: you need to setup rsyncd server and make the correct authorization.
>
> Works for simple and complex cases.
>
> > 3. ftp/ftps
> > pros: easy to use
> > cons: need to setup ftpd server, and the way is not that secure?
>
> Whenever possible, I'd prefer 1 or 2 over this.
>
> > 4. rclone
> > pros:easy to use
> > cons: hard to setup (you may need a cloud storage for middleware).
>
> I only use rclone when I want to target a cloud storage.
> A „cloud storage for middleware” does not seem sensible to me when I can
> copy using methods 1 and 2 without using such a middleware.
>
> > For me I most often use scp + rsync. and what's your choice?
>
> These are my standard choices, too. In automated scenarios I often prefer
> rsync over scp due to more flexibility in configuration.
>
> My additional tools for special purposes:
>
> 5. lsyncd
> If you need to keep directories in sync continuously, there is a tool
> called
> `lsyncd` that automates repeated invocation of `rsync` in a smart way.
>
> 6. tar + netcat (or tar + ssh in very rare cases)
> Using tar sacrifices all the flexibility of rsync but may attain a
> significantly higher performance and does not need a lot of flags to do
> the
> right thing by default (i.e. preserve everything when acting as root). I
> prefer this variant when migrating to a new disk or PC because it seems
> to
> be the most efficient variant in a "local trusted network and no speedup
> from incremental copying" scenario.
>
> I documented my approach to this here:
> https://masysma.net/37/data_transfer_netcat_tar.xhtml
>
> HTH and YMMV
> Linux-Fan
>
> öö
>

Re: mount new block deivce in existing dir

[Ken Young]

The provider has separated block device service, like what Ceph does.
So I have to buy another 100GB block device as the secondary disk.

Sincerely,
Ken Young


On Mon, Mar 6, 2023 at 6:26 AM Andy Smith  wrote:

> Hi,
>
> On Mon, Mar 06, 2023 at 03:11:43AM +0800, Ken Young wrote:
> > Filesystem  Size  Used Avail Use% Mounted on
> > /dev/vda1   9.7G  1.5G  7.9G  16% /
> > ...
> >
> > So I want to add a block device /dev/vdb and fdisk/format it and mount it
> > as /home dir.
>
> Have you checked your VPS provider's documentation to see if you
> can simply increase the size of the vda disk and vda1 partition?
> This is very commonly available with VPS setups and can often even
> be done without you rebooting.
>
> Cheers,
> Andy
>
> --
> https://bitfolk.com/ -- No-nonsense VPS hosting
>
>

mount new block deivce in existing dir

[Ken Young]

Hello

My vps has limited volume for /. You can see it as follows.

root@nxacloud-bloghost:~# lsb_release -cd

Description: Debian GNU/Linux 11 (bullseye)

Codename: bullseye


root@nxacloud-bloghost:~# df -h

Filesystem  Size  Used Avail Use% Mounted on

udev472M 0  472M   0% /dev

tmpfs98M  496K   98M   1% /run

/dev/vda1   9.7G  1.5G  7.9G  16% /
...


So I want to add a block device /dev/vdb and fdisk/format it and mount it
as /home dir.
Since /home already exists (even have data in this path). can I mount it
without problem?


Sincerely,
Ken Young

Re: what method do you prefer for data transfer between nodes?

[Ken Young]

Thanks for the suggestion. I will give it a try on mailbox.org.


Sincerely,
Ken Young


On Sun, Mar 5, 2023 at 9:05 PM  wrote:

> On Sun, Mar 05, 2023 at 07:23:49PM +0800, Ken Young wrote:
> > What provider do you recommend then?
>
> Around here (West Europe), for example, mailbox.org [1]. But they
> are far from the only one (I'm not a customer, nor associated with
> them, but I do know a few happy customers).
>
> Whenever they don't cost any money you'll have to ask yourself
> what their business model is.
>
> Cheers
>
> [1] https://mailbox.org/en/
> --
> t
>

Re: alias in bash script issue

[Ken Young]

Is perl6 production ready?
I have not used perl for a long time.


Sincerely,
Ken Young


On Sun, Mar 5, 2023 at 7:28 PM Tom Browder  wrote:

> My take:
>
> I use aliases heavily in my shell (bash), but I rarely use bash scripting
> at all.
>
> For any serious scripting, since about 2016 I'ved used Raku (formerly Perl
> 6). Before that I used Perl. Both are much easier to use.
>
> -Tom
>

Re: what method do you prefer for data transfer between nodes?

[Ken Young]

What provider do you recommend then?

Sincerely,
Ken Young


On Sun, Mar 5, 2023 at 6:54 PM  wrote:

> On Sun, Mar 05, 2023 at 11:31:42AM +0100, Michel Verdier wrote:
> > Le 5 mars 2023 tomas a écrit :
> >
> > > The others I'd only use when forced to. Well, rclone, I think,
> > > I'd even refuse when forced to. Eek.
> >
> > Do you have an alternate solution for cloud storage like google one ?
>
> I don't "do" clouds. But around here, I'd recommend a provider who offers
> mail and nextcloud for a small monthly amount (1 EUR, I think).
>
> Cheers
> --
> t
>

what method do you prefer for data transfer between nodes?

[Ken Young]

Hello,

The methods I know,

1. scp
pros: the native tool in the OS
cons: you will either input password or put key pairs into servers for
authentication.

2. rsync
pros: it can transfer data by increasement
cons: you need to setup rsyncd server and make the correct authorization.

3. ftp/ftps
pros: easy to use
cons: need to setup ftpd server, and the way is not that secure?

4. rclone
pros:easy to use
cons: hard to setup (you may need a cloud storage for middleware).

For me I most often use scp + rsync. and what's your choice?

Regards,
Ken

Re: alias in bash script issue

[Ken Young]

On Sat, Mar 4, 2023 at 5:53 PM David  wrote:

> On Sat, 4 Mar 2023 at 19:30, Ken Young  wrote:
>
> > Do you know why my alias can't work in the bash script?
> > The info is as follows.
> >
> > 1)  this alias does exist
> > $ alias |grep 'k='
> > alias k='minikube kubectl --'
> >
> > 2) it also exists in .bash_profile
> > $ cat ~/.bash_profile |grep 'k='
> > alias k="minikube kubectl --"
>
> [...]
>
> > ./get.sh: line 5: k: command not found
>
> [...]
>
> > Please give suggestions.
>
> Hi. Friends don't help friends to do bad things :)
>
> Debian's default shell is 'dash'. Its manual, readable using 'man dash',
> says
>
>
>


Hi,

my debian 11's default shell is just bash.

root@nxacloud-bloghost:~# echo $SHELL

/bin/bash

root@nxacloud-bloghost:~# lsb_release -cd

Description: Debian GNU/Linux 11 (bullseye)
Codename: bullseye

Thanks

Re: besides ping/nslookup, any other tools to resolve a hostname?

[Ken Young]

Hi,

I just launched an ubuntu container in that namespace and installed
dnsutils/iputils-ping to resolve the hostname issue.

Thank you.


On Sat, Mar 4, 2023 at 1:11 PM Rodrigo Cunha 
wrote:

> I have bitnami/mysql container (debian OS) running on kubernetes.
>> This container seems too restricted for system software, these following
>> commands are removed:
>>
>> sudo (so I can't su to root for apt)
>> ping
>> dig
>> nslookup
>> net-tools
>>
> It is a good practice to build small images for pods without commands for
> the network. To solve this problem, you can do that:
> Create a pod in the same network with net tools.
> Or
> 1 .rebuild a new docker image with another version, with net-tools.
> 2. then change the current pod with edit in deployment.
>
>
>
> On Fri, Mar 3, 2023 at 9:59 PM Ken Young  wrote:
>
>> Hello,
>>
>> I have bitnami/mysql container (debian OS) running on kubernetes.
>> This container seems too restricted for system software, these following
>> commands are removed:
>>
>> sudo (so I can't su to root for apt)
>> ping
>> dig
>> nslookup
>> net-tools
>>
>> Do you know any other way to resolve a hostname by manual  on this
>> container?
>>
>> Thanks
>> Ken
>>
>
>
> --
> Atenciosamente,
> Rodrigo da Silva Cunha
> São Gonçalo, RJ - Brasil
>
>

alias in bash script issue

[Ken Young]

Hello,

Do you know why my alias can't work in the bash script?
The info is as follows.

1)  this alias does exist

$ alias |grep 'k='

alias k='minikube kubectl --'


2) it also exists in .bash_profile

$ cat ~/.bash_profile |grep 'k='

alias k="minikube kubectl --"


3) the content of bash script

$ cat get.sh

#!/bin/bash


source ~/.bash_profile

k get node


4) alias can't work

$ ./get.sh

./get.sh: line 5: k: command not found


The system is debian 11 and the shell is bash.

Please give suggestions. Thank you.


regards.

besides ping/nslookup, any other tools to resolve a hostname?

[Ken Young]

Hello,

I have bitnami/mysql container (debian OS) running on kubernetes.
This container seems too restricted for system software, these following
commands are removed:

sudo (so I can't su to root for apt)
ping
dig
nslookup
net-tools

Do you know any other way to resolve a hostname by manual  on this
container?

Thanks
Ken

Re: Cannot open *.asc files with gpg -- because no pinentry

[Ken Heard]

On 2023-01-06 22:48, Jerome BENOIT wrote:

Hi, I had some similar issue a long time ego.

So far I can remember the following page was very useful, in particular 
section 9.2:


https://wiki.archlinux.org/title/GnuPG

best,
Jerome


Thanks for the tip, it looks good.  When I have a chance to test it next 
week, I will let the list know the result.


Regards, Ken

Cannot open *.asc files with gpg -- because no pinentry

[Ken Heard]

When I try to do so I receive the following message:

[GNUPG:] ENC_TO 091EE1D5633F19F2 16 0
[GNUPG:] KEY_CONSIDERED 7D605C8CDA5240843A6B9A8394D949CCE9099937 0
[GNUPG:] KEY_CONSIDERED 7D605C8CDA5240843A6B9A8394D949CCE9099937 0
gpg: encrypted with 4096-bit ELG key, ID 091EE1D5633F19F2, created 
2007-12-05

"KenHeard (swimmerken) "
gpg: public key decryption failed: No pinentry
[GNUPG:] ERROR pkdecrypt_failed 67108949
[GNUPG:] BEGIN_DECRYPTION
[GNUPG:] DECRYPTION_FAILED
gpg: decryption failed: No secret key
[GNUPG:] END_DECRYPTION

After an internet search I found one responder who suggested that 'the 
issue might be in the configuration of your gpg-agent'.  So I checked 
the contents of the gpg-agent.conf file.  There were only two lines:


default-cache-ttl 300
max-cache-ttl 99

At this point I ran command gpg-agent which returned 'gpg-agent running 
and available'.


Another responder said that running the following two commands worked 
for him:


pkill gpg-agent
gpg-agent --pinentry-program=/usr/bin/pinentry-gtk-2 --daemon

Before running those two commands I found that I had the following 
pinentries:


/etc/alternatives/pinentry*
/usr/bin/pinentry-curses*
/usr/bin/pinentry-tgt*
/etc/alternatives/pinentry-x11*

I then ran those two commands twice, for -curses and -tgt.  It may have 
worked for him but did not for me.


I finally tried to add to the gpg-agent.conf file the second of those 
two commands, also twice, but before trying to decrypt an *anc file I 
ran file pkill gpg-agent.  That arrangement did not work either.  By now 
I had run out of suggestions.


I would consequently appreciate any advice as to how to decrypt my *.anc 
files.


By the way, I was able to run from an Xterm in succession the commands 
gpg -k and -K, which -- without the use of any passkey -- returned 
information about my public and secret keys in succession.


Other maybe useful information: I am using Bullseye and had added to my 
file .bashrc the following two lines:


GPG_TTY=$(tty)
export GPG_TTY

Re: Will my reconstructed fstab work?

[Ken Heard]

On 2022-11-04 01:52, mick.crane wrote:

On 2022-11-03 04:52, Ken Heard wrote:

A few days ago using vim I added to my desktop fstab file a line for a
new portable storage device.  in the process I somehow managed to
screw up fstab.  Unfortunately I saved the screwed up version of fstab
before I noticed the damage done to it.




01 -e8b57fb2ac09static file system information.
02 -e8b57fb2ac09
03 -e8b57fb2ac09lkid' to print the universally unique identifier for a
04 -e8b57fb2ac09vice; this may be used with UUID= as a more robust way
to name devices
05 -e8b57fb2ac09at works even if disks are added and removed. See 
fstab(5).


It is likely unrelated but I had a thing a while ago while editing files 
with vim/nano that random text from within the file would be randomly 
inserted elsewhere in the file which was incredibly disturbing.

mick.


It's what I call the inherent inconstancy of computers; you never know 
what they are going to do!


Ken

Will my reconstructed fstab work?

[Ken Heard]

7fb2ac09020-1029  /boot/efi   vfatumask=0077  0   1
14-e8b57fb2ac09ot was on /dev/sda1 during installation
15 -e8b57fb2ac09020-1029  /dosvfatutf80   0
16 -e8b57fb2ac09os2 was on /dev/sdb1 during installation
17-e8b57fb2ac09AF2-0A16  /dos2   vfatutf80   0
18 -e8b57fb2ac09mapper/Morcom-HOME_crypt /home   ext4 defaults 
   0   2

19 -e8b57fb2ac09mapper/Morcom-VAR /varext4defaults  02
20 -e8b57fb2ac09mapper/Morcom-SWAP_crypt noneswapswap	 0 
  0

21 -e8b57fb2ac09r0  /media/cdrom0   udf,iso9660 user,noauto 0   0
22 -e8b57fb2ac09tmpfs   nodev,nosuid,size=20%   0   0
23 -e8b57fb2ac0944477-7f18-4443-a77b-c5827f977449 /media/fdr ext2 		 
user,noauto,noatime 0	0
24 -e8b57fb2ac09cebfb3-b568-493f-853b-e1b7ca5cc3a2 /media/fde ext2 
user,noauto,noatime 0 0

25 -e8b57fb2ac09/media/ssda ext4 user,noauto,noatime 0 0
26 -e8b57fb2ac099167-8497-4471-ae0c-e8b57fb2ac09 /media/phda ext4 
user,noauto,noatime 0 0
27 UUID=0fee2d01-2441-4699-a4ae-bb45c417b8ee /media/ssda ext4 
user,noauto,noatime 0 0
28 UUID=e26255ab-e6c5-4bcd-941c-7378b7cf4083 /media/ssdb ext4 
user,noauto,noatime 0 0

29 UUID=3DB1-1700   /media/fdg  vfatuser,noauto,noatime 0   0
30 UUID=5966-5502   /media/fdp  vfatuser,noauto,noatime 0   0
31 UUID=1170-1657   /media/hca  vfatuser,noauto,noatime 0   0
32 UUID=0E0A-0F26   /media/hcb  vfatuser,noauto,noatime 0   0
33 UUID=1E82-122E   /media/hcc  vfatuser,noauto,noatime 0   0
34 UUID=1D1F-1032  /media/xca   exfat   user,noauto,noatime 0   0
35 UUID=1909-1458   /media.xcb  exfat   user,noauto,noatime 0   0
36 UUID=6238-3434   /media.xcc exfatusr,noauto,nonatime 0   0
37 UUID=206F-163F   /media/xcd  exfat   user,noauto,noatime 0   0
38 UUID=3630-6530   /media/xce  exfat   user,noauto,noatime 0   0
39 UUID=3065-6630   /media/xcf  exfat   user,noauto,noatime 0   0
40 LABEL=ostree /media/phdc ext4user,noauto,noatime 0   0

The screw up started a few days ago when I entered into the file a new 
portable storage device, line 26 above.  That entry itself was 
successful.  In addition however the beginning of the text of all lines 
1-26 was overwritten by -e8b57fb2ac09, which happens to be the last 12 
digits, plus the preceding hyphen, of the UUID of the added storage 
device on line 26.  Lines 27-40 were spared the overwrites.  I have no 
idea how these overwrites happened; it certainly was not the result of 
any conscious action on my part.


Regards, Ken Heard

Re: printing pages in reverse order not possible in LO 7.2.5.2.0+

[Ken Heard]

On 2022-02-13 18:09, Ken Heard wrote:
In order to print documents on both sides of the paper using a printer 
without collating ability, I always printed first in reverse order the 
even pages.  I then  print in numerical order the odd pages on the back 
side of the paper on which the even pages are printed.  In this way the 
pages do not need to be collated manually.


While this version of LO still allows printing separately the odd and 
even pages, I discovered to my horror that it no longer allows printing 
of pages in reverse order.


The help page however still says that all one has to do to print in 
reverse order is to


1.    Choose File - Print.
2.    Click the General tab.
3.    Choose Print in reverse page order.
4.    Click Print.

In view of the foregoing it would appear that disappearance of the 
reverse print option was unintended.  In that case when can we helpless 
users expect this error to be corrected?


I don't know how I managed to miss the 'More' which causes the 'Print in 
reverse order' option to emerge.  In fact there are two 'Mores', one for 
'Range and Copies' and the other for 'Page Layout' .  The first one 
opens more options most of which I cannot use because the printer I am 
using is a simplex.  The one option I can use and want to use is "Print 
in reverse order'.  (I am not too sure what additional
options the second 'more' opens; they mostly seem to relate to 
collations which a simplex printer cannot do.)


I must however in my defence say that in the help entry for reverse 
printing there is no reference to the first 'more'.  I naturally 
inferred that an intermediate step was not necessary, as it probably was 
at some point in the dim and distant past.  I assume it likely that help 
revisions lag behind changes in the software, understandable.  The 
general layout of the print window could nevertheless stand improvement.


In any event many thanks to all for your help, in particular for Dave 
Barton's screen shot which made everything crystal clear.


Regards, Ken

Re: Debian 11 xfce

[Ken Cunningham]

> On Jan 12, 2022, at 9:58 AM, to...@tuxteam.de wrote:
> 
> On Wed, Jan 12, 2022 at 09:52:01AM -0600, c. marlow wrote:
> 
> [...]
> 
>> I honestly don't see what people see in XFCE
> 
> Uh, oh. This is asking for trouble ;)
> 
>> To me, XFCE looks VERY DATED and out of the 90s. 
> 
> perhaps...
> 
>> YMMV, I guess.
> 
> That's *exactly* the point. Mileage varies wildly, especially in the
> user interface department. There are even people who don't want a
> desktop environment *at all*, imagine that :^)
> 
> The real challenge would be a way for all of us to get along together
> despite such preferencial differences :-)
> 
> Cheers
> -- 
> t

On a debian 11 ppc64 system running on an Apple Dual G5 PPC 970 with 3.5G of 
Ram, xfce turns out to the most responsive environment of the ones that will 
work there. So that is what I use on that system.

On the other hand, on an Apple Intel Macbook circa 2006 with a Core2 processor 
and 4G of ram, Mate is very nice, and I prefer that there.

Once the applications are running it matters little however, of course.

I’m happy there are choices.

Ken

Re: Non-working CPU cores showing up

[Ken Cunningham]

hey, you're right!

free upgrade :)

K

On Tuesday, November 30, 2021, Dirk Neumann  wrote:

> On Tue, 30 Nov 2021 08:52:59 -0800
> Ken Cunningham  wrote:
>
>
> >
> > Some software, like ninja etc, use that information to decide how many
> parallel jobs to set up. On my systems (2 processors, 6 CPUs on each, each
> with two threads per core = 12 parallel build processes) that works out
> well it seems.
>
> I would have expected 24 parallel build processes on your machine
>
>
>

Re: Non-working CPU cores showing up

[Ken Cunningham]

> On Nov 30, 2021, at 8:18 AM, Paul M. Foster  wrote:
> 
> On 11/30/21 6:24 AM, Dan Ritter wrote:
>> Paul M. Foster wrote:
>>> Folks:
>>> 
>>> Here's a curious thing. I have a 10th gen Intel i3 CPU with four cores. When
>>> I look at /proc/cpuinfo, it actually shows eight cores. There's a line in
>>> the output of each core which is
>>> 
>>> cpu cores   : 4
>>> 
>>> But there are outputs for each of eight cores, numbered 0 through 7.
>>> 
>>> Is it possible that there were eight cores on this CPU, and four of them
>>> were non-working (I know it's typical to have non-working cores on a die),
>>> and this file shows all the original cores?
>>> 
>>> Or does someone have a better explanation?
>>> 
>> Try lscpu. Useful lines:
>> CPU(s):  12
>> On-line CPU(s) list: 0-11
>> Thread(s) per core:  2
>> Core(s) per socket:  6
>> Socket(s):   1
>> So this machine has one socket, 6 cores in the socket, 2 threads
>> per core, which looks like 12 CPUs.
>> You should find that your has one socket, 4 cores in the socket,
>> 2 threads per core, which looks like 8 CPUs.
>> -dsr-
> 
> It appears you are correct. lscpu shows this CPU has 4 cores, and 2 threads 
> per core. But it shows 8 CPUs. Silly.
> 
> Paul
> 

Some software, like ninja etc, use that information to decide how many parallel 
jobs to set up. On my systems (2 processors, 6 CPUs on each, each with two 
threads per core = 12 parallel build processes) that works out well it seems.

Ken

Re: Debian 11 on old Macbook

[Ken Cunningham]

Well if MacOSX is no longer booting immediately, then it appears installing
rEFInd solved your initial problem, so that is good.

rEFInd is quite configurable via refind.conf, but you may find it is
simpler to just ignore the Windows icon rather than try to force it not to
appear. The hairy details are here:

https://www.rodsbooks.com/refind/configfile.html

And rEFInd has it's own mailing list for more user help here:

https://sourceforge.net/p/refind/discussion/general/

K



On Saturday, November 27, 2021,  wrote:

> Hello,
> thanks for the help and suggestions:
> I don't need to press alt (option) to view the operating systems because I
> have rEFInd installed.
> However, 2 Debian icons appear, one OS X, and one Windows.
> If I click on the latter, a white writing on a black background appears
> with: MBR 12: and a flashing underscore.
> But I can't write anything.
> The Debian 2 takes me to Debian and OS X to the Apple operating system.
> Assuming I keep the 2 Debian, how do I get rid of the Windows symbol, since
> it doesn't lead anywhere, so it results useless?
> Thanks again
> Francesco
>

Re: Debian 11 on old Macbook

[Ken Cunningham]

> On Nov 24, 2021, at 11:11 AM, David Wright  wrote:
> 
> On Wed 24 Nov 2021 at 14:59:09 (+0100), fran...@libero.it wrote:
>> I installed Debian 11 (386) on a 2009 Macbook Pro 13 "(5.5).
>> 
>> The installation did not give me any problems except it did not detect wifi 
>> card and touchpad, but I was connected with ethernet and used an external 
>> mouse, so the whole process ended.
>> 
>> During the installation phase of Grub I only chose the hd that appeared in 
>> the window and did all the installer.
>> 
>> I enclose photos of the subdivision that the installer did. I only chose to 
>> install Debian on a partition that I had left empty choosing partitions / 
>> and home Unfortunately Debian does not appear on reboot and neither does 
>> Grub, but Mac OS (Snow Leopard) starts immediately How can I solve this 
>> problem?
>> 
>> If I reinstall using AMD64 instead, what can be the right suggestions to 
>> install Grub in the right place to reboot with it?
> 
> You don't mention anything about how you boot. From my great
> experience of Macs (watching people use them in the last
> century), I'm guessing you might have to hold down some key
> while you boot. That's not just for dual-booting (certainly
> not, 30 years ago), but for doing various Mac-ish things,
> so it should be documented somewhere.
> 
> Also there were threads here, in late August, about booting Macs.
> 
> Cheers,
> David.
> 


I can’t see your photos, but I am currently dual-booting between Ubuntu 21.10 
(64 bit) and MacOSX 10.7 on a MacBook 2,1 without troubles.

Although you can usually hold down the option key during initial boot-up to 
bring up the Mac’s BIOS boot selector, and then select either the MacOSX 10.7 
partition or the GRUB/linux partition, to make things simpler, most 
walkthroughs recommend installing rEFInd 
> which gives you a nice graphical 
menu of your available bootable systems to choose from. 


There are various walkthrus available, and all of them seem to recommend rEFInd 
as well — not doubt I followed one of these several years ago

https://www.makeuseof.com/tag/install-linux-macbook-pro/ 


https://www.lifewire.com/dual-boot-linux-and-mac-os-4125733 


https://www.maketecheasier.com/install-dual-boot-ubuntu-mac/ 


HTH,

K

Dcopserver not loading

[Ken Heard]

Very soon after I started to use the TDE 14.0.11 version I began
receiving DCOP error messages, having the effect of preventing use of
various applications like Firefox and LibreOffice. After online research
and experimentation on my part I discovered that I could solve such
preventions if before opening TDE I ran in a tty as root the following
three commands:
'dcopserver -- serverid', which returned nothing, then 'dropserver' and
finally again 'dcopserver -- serverid'.  This time however that command
returned something like the following:
'[dcopserver] local/Morcom:/tmp/.ICE-unix/dcop5698-1636014592'

>From this point on there were no longer any more DCOP error messages. I
had full use of the TDE and various applications until when I next
closed completely the computer.

It is consequently my understanding that running the 'dcopserver'
command is presumably required as part of the initial boot-up.  If such
is usually the case I would appreciate knowing what I need to do to have
my computer, named Morcom, do so as well.  Can anyone tell me?

Regards, Ken

Access to files in wiki.debian.org.

[Ken Heard]

Does anybody know when access to wiki.debian.org will be opened?  There are
two files there, UEFI and GrubEFIReinstall, that I need in order to find
out how to install a boot loader in my desktop.  On 2021-09-11 I was able
to read these files on another computer; but not thinking that it would be
necessary, I did not download them then.  Would they be available from
another source?  I did a Google search but could not find another another
one.

As suggested in a post to the debian-user list last August I sent an email
to w...@debian.org with my 1P address and the names of the files I need.  I
do however find it passing strange that while the wiki had apparently been
locked down since mid-August or earlier,  in mid-September I did have
access on wiki to both but not since.

Regards, Ken

Bullseye installation failure because boot loader not installed

[Ken Heard]

 I recently upgraded a desktop computer by replacing major parts in it
including the mainboard and CPU. I now want to install in it a more up to
date operating system than Wheezy which was the one used before the
upgrade. Having noted that the release of Bullseye is imminent I decided to
use the RC2 iteration of Bullseye rather than Buster.
I consequently downloaded file 'Debian-bullseye-DI-rc2-amd4-netinst.iso'
and also the 2021-07-18 version of the 'Debian GNU/Linux Installation
Guide".

I had considerable difficulties with the installer, to the effect that it
took me about 27 hours off and on over 10 days with nine false starts. I
managed to complete the installation but with one crucial exception. In due
course I want to impart to you my complete installation experience -- but
not before that exception is rectified. I refer to the last item on the
installation menu, 'Install the boot loader' -- it was never installed. The
verbatim transcript of the messages received from the installer at this
point and my answers thereto follow.

Install the GRUB boot loader. First message from the installer:
It seems that this new installation is the only operating system on this
computer. If so, it should be safe to install the GRUB boot loader to your
primary drive (UEFI partition/boot record).
I gave my consent to have the GRUB boot loader to my primary drive. Second
message:
You need to make the newly installed system bootable, by installing the
GRUB boot loader on a bootable device. The usual way to do this is to
install GRUB to your primary drive (UEFI partition/boot record). You may
instead install GRUB to a different drive (or partition), or to removable
media,

Device for boot loader installation:
Enter device manually
/dev/sda (ata-ST2000DM008-2FR102_ZFL3PHLG
/dev/sdb (ata-ST2000DM001-1CH164_Z340HH9V

I chose to install it on /dev/sda.

The installer however did not like the answer and sent me a third message:
It seems that this computer is configured to boot via EFI, but maybe that
configuration will not work for booting from the hard drive. Some EFI
firmware implementations do not meet the EFI specification (i.e. they are
buggy!) and do not support proper configuration of boot options from system
hard drives.

A workaround for this problem is to install an extra copy of the EFI
version of the GRUB boot loader to a fall back location, the “removable
media path”. Almost all EFI systems, no matter how buggy, will boot GRUB
that way.

Force GRUB installation to the EFI removable path?  or 
Once again I did not really have much choice in the matter; so I chose
 and pressed 'Enter', but the installer did not seem to like that
answer either.
Next a series of messages crossed the screen too quickly for me to read
them, and then the screen went blank. Not only was the ‘Install the boot
loader’ aborted but also was aborted whatever would follow – if anything.

I would consequently be very grateful if someone could tell me what needs
to be done to provide the boot loader and how to do it. I am quite eager to
start using this computer as soon as possible.

By the way I was able to find a way into the computer by doing a Knoppix
live installation. I examined the files in /boot/grub, the one belonging to
the computer, not to Knoppix. I discovered that in directory /boot/grub
there is no grub.cfg file. Also there is no directory /boot/efi. Finally I
noted that from the installer start page grub commands are accessible.

On 2021-08-12 I sent this message to sub...@bug.debian.org; but I was told
that my message will be ignored unless I identify the package to which it
relates and its version. My situation is such that i don't think i can
provide such information.

For the record the mainboard is a Gigabyte B450 I Aorus pro wifi. The CPU
is a 4 core AMD RyZen3 3200G with Radeon Vega 8 Graphics. I will not be
using a separate GPU. There are two 2 tb hard drives for a RAID1, with LVM.

Regards, Ken

Re: Acer Aspire 3 A315-21 laptop has mono sound only?

[Ken Heard]

Hello,

On 19/03/20 05:44 PM, Ken Heard wrote -- my original post on the subject:

I think that this particular laptop has mono sound.  I have been using 
it temporarily to stream content by HDMI to a LG HD monitor and then by 
fibre optical cable to a NED D3020 V2 hybrid amplifier and finally to a 
pair of bass reflect speakers.  Unfortunately sound comes out of only 
one speaker.


After extensive tests I have determined to my satisfaction that the 
problem is in the laptop.  I can test the two speakers built in the 
laptop separately, but I have no way of knowing whether the sound so 
produced is mono or stereo.  I could find no specs for this laptop with 
details about the sound card.



First, my apologies to all for my tardiness in not answering sooner the 
responses I received about the problem explained in my original post. 
All were helpful.  My tardiness was caused by problems with my desktop 
which had to take priority, such as a failed hard drive and for a few 
days every boot's starting by opening the BIOS with the message "BIOS 
has been reset – please decide how to continue". After dealing with 
these and some other problems I resumed efforts on my part to solve the 
problem described in my original post quoted above.


In the second quoted paragraph I claimed that I had made extensive 
tests.  Unfortunately they were not extensive enough. This time I did 
the sensible thing, I started by testing the Acer laptop alone. I 
connected a pair of ear buds to the laptop 3.5 mm earphone jack and 
streamed something from YouTube which had sound.  What I heard in *both* 
ears was sound!  So much for my nonsense of only mono sound; the laptop 
was doing perfectly what it should.


I then repeated with more care the tests previously done as to why there 
is no sound to the left speaker when connected by an HDMI cable from the 
laptop to the monitor.  To start these tests I used camcorder to provide 
sound imput.  The HDMI connecting cable in this case has a mini HDMI 
plug at the camcorder end and a standard size plug at the other. By 
disconnecting the amp from the monitor, plugging the buds directly into 
the monitor headphone jack and plugging the HDMI cable successively into 
the two HDMI input jacks in the monitor I heard sound in both ears in 
both cases.


I then connected the monitor to the amplifier and repeated the same 
tests described in the previous paragraph. There was sound from both 
speakers in both cases.


The next series of tests involves connecting the two HDMI cables I have 
between the laptop and the monitor.  Both have standard size plugs at 
each end. On is a cheap 3 metre one; the other is 5 metres long and of a 
higher quality.  For these tests each cable was plugged successively to 
both jacks in the monitor, and both cables were tried in each direction 
between the laptop and the monitor.


I first disconnected the amplifier from the monitor and listened on the 
buds plugged into the monitor 3.5 mm headphones jack.  I repeated these 
tests with the amplifier connected by a fibre optic cable to the monitor.


The result for all the tests described in the last two paragraphs was 
the same.  In every case there was sound from only one channel. (I also 
made sure that each speaker could work if plugged into the right speaker 
jacks.  They both did.)


To me it seems improbable that both HDMI cables would be faulty at the 
same time; that surely is too much of a coincidence, especially as one 
is supposed to be a good one. If that assumption is correct then the 
HDMI jack on the laptop must be at fault.  At the moment however I only 
have one source to test both cables, only the laptop.


I do have however a desktop computer which I purchased in 2015 and 
installed Debian Wheezy in it.  This computer is the one I have normally 
used ever since, the laptop since 2018 being the backup.


In the past four years however -- in spite of much on line research and 
numerous posts back and fourth on the Debian user list -- I was never 
able to get sound to work, .  It is now my intention – once I have a two 
month window in which to do so – to replace in the desktop Wheezy with 
Buster.  With any luck I may have only sound working on it.


On that assumption I will have another device to test both HDMI standard 
cables. If only one of them works in these tests then the question to 
answer is why it did not work for the laptop.  If both work on this 
device then the fault must be the HDMI jack in the Acer laptop.


Regards, Ken

Re: Acer Aspire 3 A315-21 laptop has mono sound only?

[Ken Heard]

On 19/03/20 09:10 PM, Joe wrote:




If you're getting convincing stereo, then the problem is elsewhere in
the path. I always suspect HDMI of anything that involves it. Yes, it
can do many channels, and something with 'HDMI' stamped on it should do
at least stereo, but that may be the weak link in the chain. You should
have a headphone jack on any modern screen which handles audio, try
your cheap earphones in that, with sound via HDMI. Again, VLC provides a
quick mono/stereo switch, but many sound hardware drivers can also do
that.


I did what you suggested, and only one ear received any sound. The next 
step is to try another HDMI cable.  Thanks all for the suggestions.


Regards, Ken

Acer Aspire 3 A315-21 laptop has mono sound only?

[Ken Heard]

I think that this particular laptop has mono sound.  I have been using 
it temporarily to stream content by HDMI to a LG HD monitor and then by 
fibre optical cable to a NED D3020 V2 hybrid amplifier and finally to a 
pair of bass reflect speakers.  Unfortunately sound comes out of only 
one speaker.


After extensive tests I have determined to my satisfaction that the 
problem is in the laptop.  I can test the two speakers built in the 
laptop separately, but I have no way of knowing whether the sound so 
produced is mono or stereo.  I could find no specs for this laptop with 
details about the sound card.


Lack of such specs indicates to me that Acer does not want people to 
know that this laptop is mono only. Is mono sound the norm for low 
market laptops such as this one and notebooks?


Ken Heard

Re: Script does not do what it is told to do

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2019-10-26 9:55 p.m., The Wanderer wrote:
> On 2019-10-26 at 21:33, Ken Heard wrote:
> 
>> The tar scripts now look like this. 
>> --- 
>> #!/bin/bash # Script to back up all browser files in directory
>> /home/ken/mozilla. echo here tar -cpzf
>> /media/fde/backups/kbrowsers.tgz  --exclude-caches --wildcards -T
>> docs/tarlists/kbrowsers.lst 
>> --- This one
>> exceptionally has 'echo here" above the tar command at the 
>> suggestion of Kushal Kumeran.
> 
> Where is this script located (full path and filename)?
> 
> If I'm reading things right, you also might want to either put the
> 'cd /home/ken' line back (no need for the other cd), or use the
> full path to kbrowsers.lst instead of a relative path; otherwise
> this won't work right if you run it from any directory which
> doesn't have that relative path under it.
> 
> (If kbrowsers.lst has relative paths in it, this might not work 
> correctly from another location without the cd anyway, even if you
> do use a full path on the command line. I don't have experience
> with that tar feature myself.)
> 
>> Also at his request I ran the following two commands: 'which
>> tarkbrowsers' which returned '/home/ken/bin/tarkbrowsers', and 
>> 'cat $(which tarkbrowsers)' which returned 
>> --- # Script
>> to back up all browser files in directory /home/ken/mozilla. 
>> CURPWD=$PWD cd /home/ken tar -czf
>> /media/ssda/backups/kbrowsers.tgz  --exclude-caches --wildcards
>> -T docs/tarlists/kbrowsers.lst cd $CURPWD 
>> --- As this
>> return includes the change of directories commands it is clear 
>> that somewhere there is a series of extraneous scripts but the 
>> software is being coy about telling me where they are.
> 
> Eh? No, it isn't. It already told you clearly that this latter
> script is '/home/ken/bin/tarkbrowsers'.
> 
> Your mail doesn't (seem to?) give us enough information for us to 
> identify where the former of the two given scripts is, but
> presumably you know where to find that one, since you were able to
> give its contents here.
> 
> If you don't want to run this latter version of the script, either
> move it out of its current directory (which is apparently in your
> PATH), or replace it with the version you do want to run.
> 
> 
> (I'm presuming that the newline between '--exclude-caches' and 
> '--wildcards' in each script is an artifact of the mail process,
> and not present in the actual on-disk script. If it *is* present
> there, then these are both going to break and not do what you
> expect.)
> 
Thank you for your prompt response.  I understand what you mean about
full and relevant paths.  I will have to look into this issue; it may
affect the outcome of the script regardless of the computer as both
the Toronto and Thailand computers are set up the same way, with the
same paths, etc.

Regards, Ken

-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXbWk7AAKCRCU2UnM6QmZ
N/WlAJ4+Nso8uo4X/wLXmj5TMvqSO0uMggCdEphMTf/xIiDxtAJbnQyUV/l/8vA=
=7+2G
-END PGP SIGNATURE-

Re: Script does not do what it is told to do

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I want at this time to thank all of you for taking the time to provide
various suggestions as to how to fix the problems I describe in my
original post and in my previous thread 'Signs of hard drive failure?'
My reason for the thank you is that early Monday morning 28 October I
am leaving for Thailand and will not be back in Toronto until
mid-April 2020 and so will not have time for any follow up before I
leave.

Consequently when my Toronto computer is closed tomorrow it will not
be opened or used for the next six months approximately. For those six
months I will be using my desktop computer in Thailand which will have
been idle from April 2019 to November 2019. I will however summarize
here what I have found out so far in Toronto.

With regard to possible hard disk failures, with the help of some of
you I found out that the internal two hard drives have some life left
in them.  What had failed however some of my backup portable devices,
a 1 tb portable USB hard drive and two of my six high capacity (32-64
tb) USB flash drives, all of which I use for back-ups.  What I think
had not failed was a SanDisk Extreme 500 (500 gb so they say, but
somewhat less) solid state USB portable drive.

So for my working files which I will be taking with me I am using one
of the 'good' 64 tb USB flash drives mount on /media/fde, and maybe
the the SanDisk on /media/ssda. My working files will be compressed
into some twenty tarballs written to the fde drive, but not on the
ssda for reasons explained below. I am also storing those tarballs in
Dropbox.

The scripts for these two devices follow.  Both are encrypted.
- --
#!/bin/bash
# Script to open and mount fde

sudo cryptsetup luksOpen UUID=26bb9cea-4ac0-47fa-838b-067baeb1936f fde
mount /media/fde
- --
#!/bin/bash
# Script to open and mount ssda

sudo cryptsetup luksOpen UUID=347eec80-ede8-4b10-bef6-24065ced3e97
ssda mount /media/ssda
- ---
As I had found that all these scripts could operate without changing
the working directory back and forth from /home/ken I removed from
them all the lines changing the working directories which shocked,
horrified and amazed many of you.

The first script above works; it can open the fde drive.  The second
one does not work, but it will work if the tar command is extracted
from the script and run independently in a virtual terminal.  (By the
way I have sudo set up to allow me to open every file without
password, as this is the only computer on the premises.)

The tar scripts now look like this.
- ---
#!/bin/bash
# Script to back up all browser files in directory /home/ken/mozilla.
echo here
tar -cpzf /media/fde/backups/kbrowsers.tgz  --exclude-caches
- --wildcards -T docs/tarlists/kbrowsers.lst
- ---
This one exceptionally has 'echo here" above the tar command at the
suggestion of Kushal Kumeran.  Also at his request I ran the following
two commands:
'which tarkbrowsers' which returned '/home/ken/bin/tarkbrowsers', and
'cat $(which tarkbrowsers)' which returned
- ---
# Script to back up all browser files in directory /home/ken/mozilla.
CURPWD=$PWD
cd /home/ken
tar -czf /media/ssda/backups/kbrowsers.tgz  --exclude-caches
- --wildcards -T docs/tarlists/kbrowsers.lst
cd $CURPWD
- ---
As this return includes the change of directories commands it is clear
that somewhere there is a series of extraneous scripts but the
software is being coy about telling me where they are.  As about a
fortnight ago all these scripts worked as they should have, it may be
possible that the ones which do not work now are those which I did not
change in the meantime.

So here is where the matter stands now.  I will be using these scripts
in Thailand and determine whether these idiosyncrasies occur there,
where I will have a buster machine instead of the stretch one I am
still using in Toronto.  So I may be on the thread again in Thailand,
and or possibly again in Toronto in April 2020.

Regards, Ken Heard







-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXbTz8gAKCRCU2UnM6QmZ
N1ZGAJ9oiCkQsjTrdBa3HC5p5+CqybdO4ACeIex9comH1W7x1H3mbt7N1Tdm0pI=
=VKWL
-END PGP SIGNATURE-

Re: Firefox Seems to Have a Mind of It's Own

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Spelling error in the title.  'Its' as a possessive adjective does not
have an apostrophe.  "It's" with the apostrophe is only used as a
contraction of 'it is' or 'it has'.

Regards, Ken
-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXbJkzwAKCRCU2UnM6QmZ
N/RLAJ9qwCulXA35j0dR5BBckI679sBiNwCfUZiwY/8tWEgIXz6iN6JfMGgZPVg=
=0IuP
-END PGP SIGNATURE-

Re: Script does not do what it is told to do

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2019-10-24 8:40 p.m., deloptes wrote:
> Ken Heard wrote:
> 
>> Whenever I run this script -- or several others like it
>> 
>> #!/bin/bash # Script to back up all browser files in directory
>> /home/ken/mozilla. STARTDIR=$PWD cd /home/ken tar -czf
>> /media/fde/backups/kbrowsers.tgz  --exclude-caches \ --wildcards
>> -T docs/tarlists/kbrowsers.lst cd "$STARTDIR"
>> 
>> it returns the following:
>> 
> 
> this is not possible because in the script you tell tar to create
> archive in /media/fde/backups/

Yes, that is the location where I want the archive kbrowsers.tgz to be
put.

>> ken@Turing:~$ tarkbrowsers tar (child):
>> /media/ssda/backups/kbrowsers.tgz: Cannot open: No such file or
>> directory tar (child): Error is not recoverable: exiting now tar:
>> /media/ssda/backups/kbrowsers.tgz: Cannot write: Broken pipe tar:
>> Child returned status 2 tar: Error is not recoverable: exiting
>> now
>> 
> 
> and here you are failing to write
> /media/ssda/backups/kbrowsers.tgz

/media/ssda/ .. does not exist, neither as a directory or in /etc/fstab

>> It is trying to save the tarball to a different portable storage 
>> device, ssda, which had failed and consequently is no longer in 
>> /etc/fstab.  How could such a thing happen?
>> 
> 
> obviously it is another script or whatever

But how?

>> However, if I pull the line with the tar command out of the
>> script and running it independently it does what it is told.  If
>> there is something wrong with the script I can't find it.
>> 
>> I also tried the script with the second line reading
>> STARTDIR="$PWD"; but that change made no difference, the result
>> was the same.
>> 
> 
> The question is where do you want to write the data - if on media
> attached then create entry in /etc/fstab so that you can mount it
> in consistent way, change the script and the problem will be gone.

/media/fde is the mount point for the device to which will be written
the archive.  It is properly mounted, and I can write files to it and
read those files.  As I said in my original post I am able to write
the archive successfully to /media/fde/backups/ by copying the tar
command line to a console and read the contents of the tarball.

> I have seen all kind of stupid and ugly admin scripts, where the
> author himself does not know how or why it is or it is not working
> at the end.
> 
> regards

Regards, Ken



-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXbJjwQAKCRCU2UnM6QmZ
NzIJAJ9Ng8JazKDR52yEBcyEoatQBu7PQACcCT5xNctrkAiobqpPGG1Soj9lmCI=
=fv+M
-END PGP SIGNATURE-

Re: Script does not do what it is told to do

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2019-10-24 7:12 p.m., Anuradha Weeraman wrote:

> On Thu, Oct 24, 2019 at 06:37:08PM -0400, Ken Heard wrote:
>> tar -czf /media/fde/backups/kbrowsers.tgz  --exclude-caches \ -
>> --wildcards -T docs/tarlists/kbrowsers.lst
> 
> Try removing the single '-' at the beginning of the second line...

In this instance it does not matter whether that '-' is there or not.
 Pulling that line out of the script and running it independently
works both with and without it.

Regards, Ken



-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXbJHlQAKCRCU2UnM6QmZ
N9CyAJ0V2UNi7X6qEj+oTe24JWYyksnm/gCff2SSNSfy8oFTt6wS9R+8f5HupZY=
=nMke
-END PGP SIGNATURE-

Script does not do what it is told to do

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Whenever I run this script -- or several others like it

#!/bin/bash
# Script to back up all browser files in directory /home/ken/mozilla.
STARTDIR=$PWD
cd /home/ken
tar -czf /media/fde/backups/kbrowsers.tgz  --exclude-caches \
- --wildcards -T docs/tarlists/kbrowsers.lst
cd "$STARTDIR"

it returns the following:

ken@Turing:~$ tarkbrowsers
tar (child): /media/ssda/backups/kbrowsers.tgz: Cannot open: No such
file or directory
tar (child): Error is not recoverable: exiting now
tar: /media/ssda/backups/kbrowsers.tgz: Cannot write: Broken pipe
tar: Child returned status 2
tar: Error is not recoverable: exiting now

It is trying to save the tarball to a different portable storage
device, ssda, which had failed and consequently is no longer in
/etc/fstab.  How could such a thing happen?

However, if I pull the line with the tar command out of the script and
running it independently it does what it is told.  If there is
something wrong with the script I can't find it.

I also tried the script with the second line reading STARTDIR="$PWD";
but that change made no difference, the result was the same.

Help please!

Regards, Ken


-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXbInjgAKCRCU2UnM6QmZ
N4QyAJkBqb+i4+4VRA38OWfWHV1GE15AmQCffAk/LjD8QAk48ElZPDk2ReI5RJc=
=tfcB
-END PGP SIGNATURE-

Re: Signs of hard drive failure?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

At about 04:35 EDT today 2019-10-22 Tuesday I ran 'sudo smartctl -t
long /dev/sdb'.  At 12:48 I ran the command in the next line.  The
result is below after the next paragraph.

Based on the results below relating to /dev/sdb and the results
relating to /dev/sda quoted in my previous email today, I conclude
that while smartctl considers my two hard drives three years old and
consequently prone to failure -- what else really is new, all devices
are prone to failure -- my current backup problems are not related to
failure of either or both of these drives.  As I stated in my original
post, they have been in use for only 1.75 years.  I shall consequently
continue to examine at my scripts for errors in creating tarballs for
my backups.

Regards, Ken
- -------

ken@SOL:~$ sudo smartctl -a /dev/sdb
smartctl 6.6 2016-05-31 r4324 [x86_64-linux-4.9.0-9-amd64] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke,
www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family: Seagate Barracuda 7200.14 (AF)
Device Model: ST2000DM001-1ER164
Serial Number:Z4Z0TKY8
LU WWN Device Id: 5 000c50 0796a3479
Firmware Version: CC25
User Capacity:2,000,398,934,016 bytes [2.00 TB]
Sector Sizes: 512 bytes logical, 4096 bytes physical
Rotation Rate:7200 rpm
Form Factor:  3.5 inches
Device is:In smartctl database [for details use: -P show]
ATA Version is:   ACS-2, ACS-3 T13/2161-D revision 3b
SATA Version is:  SATA 3.1, 6.0 Gb/s (current: 3.0 Gb/s)
Local Time is:Tue Oct 22 12:45:42 2019 EDT
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

General SMART Values:
Offline data collection status:  (0x82) Offline data collection activity
was completed without error.
Auto Offline Data Collection:
Enabled.
Self-test execution status:  (   0) The previous self-test routine
completed
without error or no self-test
has ever
been run.
Total time to complete Offline
data collection:(   80) seconds.
Offline data collection
capabilities:(0x7b) SMART execute Offline immediate.
Auto Offline data collection
on/off support.
Suspend Offline collection
upon new
command.
Offline surface scan supported.
Self-test supported.
Conveyance Self-test supported.
Selective Self-test supported.
SMART capabilities:(0x0003) Saves SMART data before entering
power-saving mode.
Supports SMART auto save timer.
Error logging capability:(0x01) Error logging supported.
General Purpose Logging supporte
d.
Short self-test routine
recommended polling time:(   1) minutes.
Extended self-test routine
recommended polling time:( 219) minutes.
Conveyance self-test routine
recommended polling time:(   2) minutes.
SCT capabilities:  (0x1085) SCT Status supported.

SMART Attributes Data Structure revision number: 10
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME  FLAG VALUE WORST THRESH TYPE
UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate 0x000f   115   099   006Pre-fail
Always   -   90023624
  3 Spin_Up_Time0x0003   096   096   000Pre-fail
Always   -   0
  4 Start_Stop_Count0x0032   100   100   020Old_age
Always   -   557
  5 Reallocated_Sector_Ct   0x0033   100   100   010Pre-fail
Always   -   0
  7 Seek_Error_Rate 0x000f   073   060   030Pre-fail
Always   -   24992023
  9 Power_On_Hours  0x0032   096   096   000Old_age
Always   -   3834
 10 Spin_Retry_Count0x0013   100   100   097Pre-fail
Always   -   0
 12 Power_Cycle_Count   0x0032   100   100   020Old_age
Always   -   556
183 Runtime_Bad_Block   0x0032   100   100   000Old_age
Always   -   0
184 End-to-End_Error0x0032   100   100   099Old_age
Always   -   0
187 Reported_Uncorrect  0x0032   100   100   000Old_age
Always   -   0
188 Command_Timeout 0x0032   100   100   000Old_age
Always   -   0 0 0
189 High_Fly_Writes 0x003a   096   096   000Old_age
Always   -   4
190 Airflow_Temperature_Cel 0x0022   070   065   045 

Re: Signs of hard drive failure?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

As instructed at approximately 19:00 on 2019-10-21 Monday I ran first
'sudo smartctl -t long /dev/sda' which returned the following:

ken@SOL:~$ sudo smartctl -t long /dev/sda
smartctl 6.6 2016-05-31 r4324 [x86_64-linux-4.9.0-9-amd64] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke,
www.smartmontools.org

=== START OF OFFLINE IMMEDIATE AND SELF-TEST SECTION ===
Sending command: "Execute SMART Extended self-test routine immediately
in off-line mode".
Drive command "Execute SMART Extended self-test routine immediately in
off-line mode" successful.
Testing has begun.
Please wait 212 minutes for test to complete.
Test will complete after Mon Oct 21 22:38:34 2019

Use smartctl -X to abort test.

Without aborting the test, at approximately 03:20 on 2019-10-22
Tuesday I ran 'sudo smartctl -a /dev/sda which produces the following
result:

ken@SOL:~$ sudo smartctl -a /dev/sda
smartctl 6.6 2016-05-31 r4324 [x86_64-linux-4.9.0-9-amd64] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke,
www.smartmontools.org

=== START OF INFORMATION SECTION ===
Model Family: Seagate Barracuda 7200.14 (AF)
Device Model: ST2000DM001-1ER164
Serial Number:Z4Z0WAAC
LU WWN Device Id: 5 000c50 07969f376
Firmware Version: CC25
User Capacity:2,000,398,934,016 bytes [2.00 TB]
Sector Sizes: 512 bytes logical, 4096 bytes physical
Rotation Rate:7200 rpm
Form Factor:  3.5 inches
Device is:In smartctl database [for details use: -P show]
ATA Version is:   ACS-2, ACS-3 T13/2161-D revision 3b
SATA Version is:  SATA 3.1, 6.0 Gb/s (current: 3.0 Gb/s)
Local Time is:Tue Oct 22 03:41:10 2019 EDT
SMART support is: Available - device has SMART capability.
SMART support is: Enabled

=== START OF READ SMART DATA SECTION ===
SMART overall-health self-assessment test result: PASSED

General SMART Values:
Offline data collection status:  (0x82) Offline data collection activity
was completed without error.
Auto Offline Data Collection:
Enabled.
Self-test execution status:  (   0) The previous self-test routine
completed
without error or no self-test
has ever
been run.
Total time to complete Offline
data collection:(   80) seconds.
Offline data collection
capabilities:(0x7b) SMART execute Offline immediate.
Auto Offline data collection
on/off support.
Suspend Offline collection
upon new
command.
Offline surface scan supported.
Self-test supported.
Conveyance Self-test supported.
Selective Self-test supported.
SMART capabilities:(0x0003) Saves SMART data before entering
power-saving mode.
Supports SMART auto save timer.
Error logging capability:(0x01) Error logging supported.
General Purpose Logging supporte
d.
Short self-test routine
recommended polling time:(   1) minutes.
Extended self-test routine
recommended polling time:( 212) minutes.
Conveyance self-test routine
recommended polling time:(   2) minutes.
SCT capabilities:  (0x1085) SCT Status supported.

SMART Attributes Data Structure revision number: 10
Vendor Specific SMART Attributes with Thresholds:
ID# ATTRIBUTE_NAME  FLAG VALUE WORST THRESH TYPE
UPDATED  WHEN_FAILED RAW_VALUE
  1 Raw_Read_Error_Rate 0x000f   117   099   006Pre-fail
Always   -   165044768
  3 Spin_Up_Time0x0003   096   096   000Pre-fail
Always   -   0
  4 Start_Stop_Count0x0032   100   100   020Old_age
Always   -   558
  5 Reallocated_Sector_Ct   0x0033   100   100   010Pre-fail
Always   -   0
  7 Seek_Error_Rate 0x000f   074   060   030Pre-fail
Always   -   29287810
  9 Power_On_Hours  0x0032   096   096   000Old_age
Always   -   3826
 10 Spin_Retry_Count0x0013   100   100   097Pre-fail
Always   -   0
 12 Power_Cycle_Count   0x0032   100   100   020Old_age
Always   -   557
183 Runtime_Bad_Block   0x0032   100   100   000Old_age
Always   -   0
184 End-to-End_Error0x0032   100   100   099Old_age
Always   -   0
187 Reported_Uncorrect  0x0032   100   100   000Old_age
Always   -   0
188 Command_Timeout 0x0032   100   100   000Old_age
Always   -   0 0 0
189 High_Fly_Writes 0x003a   097   097   000   

Re: Signs of hard drive failure?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Today Bob Weber wrote, in part:


> I would first check if the raid was working.  Use "cat
> /proc/mdstat". You will see something like this for each raid drive
> configured:
> 
> md0 : active raid1 sdb1[3] sda1[2] 28754230 blocks super 1.2 [2/2]
> [UU]
> 
> Make sure that both U's are there.  If not be careful because the
> raid is operating on one disk.  Before you reboot copy all the
> important data from that raid drive.

I had tried that, and both U's were there.

> Next use smartctl to do a long self test.  Use "smartctl -t long 
> /dev/sda".  You can still use the machine but it will slow the
> test down.  The tests take a long time and smartctl will estimate
> how long. Then do the second drive "smartctl -t long /dev/sdb".

I will do these tonight while I watch the results of the Canadian
election.

> If these pass then you could try booting with a system rescue CD.
> First check what drive names it has used by running "ls /dev/md*".
> You will see something like /dev/md0 or /dev/md123. Now check the
> filesystem on the raid drive with "fsck -f /dev/mdx" replacing x
> with what you found in the previous command.

To be kept in reserve; I have used Knoppix before.

> That should keep you bust for a while.

Indeed.  The other three responders also found problems with my
scripts; so I will have to look into those as well.

> Let the list know what you found.

Of course.

Regards, Ken




-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXa44BAAKCRCU2UnM6QmZ
N/OGAJ9n/eEBs3TjdrBz7x5VRr4qJ4jilgCeKnzk4FHtXNkFdM/x2tajyXOh3WI=
=Q0kb
-END PGP SIGNATURE-

Signs of hard drive failure?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

In the past week or so some in my computer procedures have become
sluggish, and some others have not worked at all.

For example the following script works:

#! /bin/bash
CURPWD=$PWD
cd /home/ken
tar -czf /media/fde/backups/kfinancescurrent.tgz --wildcards\
- --exclude-from=docs/tarlists/kfinancesarchive.lst docs/01-kens/Finances
cd $CURPWD

Whereas this one does not work now but did two weeks ago:

#!/bin/bash
# Shell script to create a tgz file for the contents of the
# /home/ken/docs and the /usr/local/ directories,
# minus the files in file /home/ken/docs/tarlists/kexcludedocs.lst
# This script may be run from any directory to which the user has
write # permission.

# Start by creating a variable with the current directory.
CURPWD=$PWD
# Change directory to /
cd /
# Create the tarball.
tar -czpf media/fde/backups/kdocsfull.tgz  -X
/home/ken/docs/tarlists/kdocsfullexclude.lst -T
/home/ken/docs/tarlists/kdocsfullinclude.lst
# Return to the starting directory.
cd $CURPWD

Now when I try to run it it returns the following:

ken@SOL:~$ tar -czpf media/fde/backups/kdocsfull.tgz  -X
/home/ken/docs/tarlists/kdocsfullexclude.lst -T
/home/ken/docs/tarlists/kdocsfullinclude.lst
tar (child): media/fde/backups/kdocsfull.tgz: Cannot open: No such
file or directorytar: home/ken/docs: Cannot stat: No such file or
directory
tar: usr/local: Cannot stat: No such file or directory
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now

All the files/directories which this script cannot stat do in fact
exist, proven by the fact that the first script uses the same
directories, but different files in those directories.

As these symptoms can indicate I think hard drive failure or potential
failure I am trying to explore this possibility.

I am using Stretch and TDE with two 2 TB Seagate Barracuda hard drives
in a RAID 1 configuration.  Both drives were purchased at the same
time and were installed in the box on 2016-05-30.  Although hree and
one half years ago, this particular box is only used six months out of
twelve.  I would not have thought that drives -- if they last the
first year -- would not show signs of failure after 1.75 years.

In any event, I ran smartctl -a on both drives.  For both "SMART
overall-health self-assessment test result [was] 'PASSED'"
Nevertheless for all the specific attributes, identical for both
drives, three of them had the indication 'Pre-fail' and 'Old-age' for
the other nineteen.

I also ran 'badblock -v'.  Both had 1953514583 blocks.  The test for
/dev/sda was interrupted at block 738381440, and for /dev/sdb at block
42064448.

I am not sure what all these test results mean, or how accurate they
would be for indicating if or when a failure would occur. It did occur
to me that if after copying all my data files to an external hard
drive I could replace the /dev/sdb device with a new one and copy all
the data in /dev/sda on the assumption with a new and presumably
pristine drive the OS given the choice would access the data it wanted
from the drive which responded the quicker.

If that approach worked I could replace the other drive in another
year or two (really one year of use) so that both drives in the RAID 1
would not be of the same age.

Comments and advice as to the best way of getting this computer back
to 'normal' to the extent that such a state could ever be 'normal'.

Regards Ken

-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXa0WDwAKCRCU2UnM6QmZ
N6elAJ0dWU0ElkIqvRebe8xGCrg77Tl0IQCeIj94dVV7aeBfjBq6Mpna/Jol/J0=
=PrGY
-END PGP SIGNATURE-

Re: USB flash drive opens read only -- how to fix?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2019-10-02 2:51 a.m., to...@tuxteam.de wrote, in part:

> Yep. Never forget -- there's a whole computer with its own OS in 
> your flash drive. That "write protect" (sometimes) available as a 
> physical switch is just communicated to your drivers via some 
> protocol over USB.

Interesting point.  So it is likely a permanent hardware failure after
all.

Ken

-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXZSvcQAKCRCU2UnM6QmZ
N8P0AJ0bUEKWkFA3+jK7BDN5nOIEo9TwlgCfZJdf7QYj2vVh449m9F0HZnqmTiI=
=JF3Z
-END PGP SIGNATURE-

Re: USB flash drive opens read only -- how to fix?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2019-10-01 11:03 p.m., David Christensen wrote:

> I have read that some USB flash drives will revert to read-only
> mode when they detect an internal error.

Makes sense I suppose, but in a negative way. I did not know that
flash drives (some? all?) do have this feature.

> This gives the user a chance to copy out their data.

I would have had to do so if the data there I needed to keep.

> My limited experience with failing USB flash drives is that they go
> from read-write to read-only to useless in a matter of minutes or
> seconds.

Strange.  I had two other flash drives fail at about the same time;
both I was able to resuscitate with luksformat. Neither of those had
the ro marker.

> What happens if you insert the USB drive into a Windows machine and
> try to format it (cancel out before doing the actual format)?
> macOS? FreeBSD?  Other?

I will have to try using it with other versions of the Turing machine
as soon as I have access to same.

Ken



-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXZSu1wAKCRCU2UnM6QmZ
N1n+AJ9Nixu4KhokPJNYbDES+fGakzDCiQCfWYa9G038uQMwB81/rD1f0BpQdOE=
=ml+2
-END PGP SIGNATURE-

Re: USB flash drive opens read only -- how to fix?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2019-10-01 5:33 p.m., Pascal Hambourg wrote:
> Le 01/10/2019 à 23:09, Ken Heard a écrit :
>> 
>> - - after unmounting and closing encryption running as root
>> 'wipefs -a -f /dev/sdd' returns 'wipefs: error: /dev/sdd: probing
>> initialization failed: Read-only file system'.
> 
> The USB flash drive is probably physically write-protected. Check
> the kernel logs juste after inserting it for confirmation.

Here are the kern.log entries created the most recent time this drive:

Oct  1 19:25:21 SOL kernel: [28463.821015] usb 9-4: new SuperSpeed USB
device number 15 using xhci_hcd
Oct  1 19:25:21 SOL kernel: [28463.844007] usb 9-4: New USB device
found, idVendor=0951, idProduct=1666
Oct  1 19:25:21 SOL kernel: [28463.844010] usb 9-4: New USB device
strings: Mfr=1, Product=2, SerialNumber=3
Oct  1 19:25:21 SOL kernel: [28463.844012] usb 9-4: Product:
DataTraveler 3.0
Oct  1 19:25:21 SOL kernel: [28463.844014] usb 9-4: Manufacturer: Kingst
on
Oct  1 19:25:21 SOL kernel: [28463.844016] usb 9-4: SerialNumber:
000AEB91EC07BD41230C0068
Oct  1 19:25:21 SOL kernel: [28463.845810] usb-storage 9-4:1.0: USB
Mass Storage device detected
Oct  1 19:25:21 SOL kernel: [28463.845973] scsi host9: usb-storage 9-4:1
.0
Oct  1 19:25:23 SOL kernel: [28464.861194] scsi 9:0:0:0: Direct-Access
Kingston DataTraveler 3.0 PMAP PQ: 0 ANSI: 6
Oct  1 19:25:23 SOL kernel: [28464.861817] sd 9:0:0:0: [sdd] 61226496
512-byte logical blocks: (31.3 GB/29.2 GiB)
Oct  1 19:25:23 SOL kernel: [28464.862088] sd 9:0:0:0: [sdd] Write
Protect is off
Oct  1 19:25:23 SOL kernel: [28464.862090] sd 9:0:0:0: [sdd] Mode
Sense: 45 00 00 00
Oct  1 19:25:23 SOL kernel: [28464.862248] sd 9:0:0:0: Attached scsi
generic sg4 type 0
Oct  1 19:25:23 SOL kernel: [28464.862365] sd 9:0:0:0: [sdd] Write
cache: disabled, read cache: enabled, doesn't support DPO or FUA
Oct  1 19:25:23 SOL kernel: [28465.042599] sd 9:0:0:0: [sdd] Write
Protect is on
Oct  1 19:25:23 SOL kernel: [28465.042604] sd 9:0:0:0: [sdd] Mode
Sense: 45 00 80 00
Oct  1 19:25:23 SOL kernel: [28465.043086] sd 9:0:0:0: [sdd] Attached
SCSI removable disk
Oct  1 19:25:37 SOL kernel: [28479.639563] EXT4-fs (dm-7): mounting
ext2 file system using the ext4 subsystem
Oct  1 19:25:37 SOL kernel: [28479.642175] EXT4-fs (dm-7): mounted
filesystem without journal. Opts: (null)

One entry says that Write protect is off, but a later entry says that
is is on.

How could this drive be physically protected from writing to it when
there is no switch on the drive itself to unable (or release) such
physical protection?

Regards, Ken




-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXZPiFQAKCRCU2UnM6QmZ
N5LyAJ90ESCk5g0fJa6tnqPeCClmjhXX+wCfb7feHPTKEm6dkAxOpb1yT8nU5VA=
=3RCx
-END PGP SIGNATURE-

USB flash drive opens read only -- how to fix?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Messieurs et mesdames:

Command 'mount' returns among other things the following two lines:

/dev/mapper/fde on /media/fde type ext2\
(rw,nosuid,nodev,noexec,noatime,block_validity,barrier,user_xattr,acl)

/dev/mapper/fdc on /media/fdc type ext2\
(ro,nosuid,nodev,noexec,noatime,block_validity,barrier,user_xattr,acl)

The first line above (divided into two here) is what 'mount' shows for
all my encrypted USB flash drives when opened -- they are all opened
with 'rw'.

The second lino above is what is shown for only one of my incrypted
USB flash drives; it opens with 'ro'.  When this drive is opened and
mounted the message returned is 'mount: /dev/mapper/fdc is
write-protected, mounting read-only'.

This situation is very strange.  Something somewhere causes this drive
to be opened 'ro' instead of the 'rw' specified for this mount point
in file /etc/fstab.

Not only can I not find out where I cannot find a way make this drive
'rw'. It is consequently useless.  Among the workarounds I tried are:

- - the commands 'sudo chmod' to change permissions.  (I own the
  directories and files in the drive.)

- - after dismounting the drive but keeping the encryption open running
 'sudo mount -o remount,rw /media/fdc' which returns the same message
  mentioned above: 'mount: /dev/mapper/fdc is write-protected, mounting
  read-only'.

- - after unmounting and closing encryption running as root 'wipefs -a -f
  /dev/sdd' returns 'wipefs: error: /dev/sdd: probing initialization
  failed: Read-only file system'.

- - 'sudo luksformat -t ext2' /dev/sdd after going through the
  preliminaries returns 'Could not create LUKS device /dev/sdd
  at /usr/sbin/luksformat line 91,  line 30'

- - 'sudo mkfs.ext2 /dev/sdd' after asking if I really want to go ahead
   with this command returns '/dev/sdd: Read-only file system while
   setting up superblock'.

After some on line searches it appears that to regain use of this
flash drive (it contains no essential data) I will have to go through
several steps to erase LUKS and then use the 'dd' command to wipe the
drive clean.  I am still afraid to use that command because of the
potential damage that can be caused by one or more bad 'if' and 'of'
settings if there is not an easier way.  Does anyone here know of one?
 In any event I would surely know how this situation would arise in
the first place.

Regards, Ken Heard

-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXZPAiAAKCRCU2UnM6QmZ
N5HJAJ9S/haC6nE8/EkuhXtCETZimi8mzQCeO7BmV2pKWO4TuPMkcJTH/71nMDY=
=04oO
-END PGP SIGNATURE-

Cannot connect to HP printer

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I am presently using stretch.  I have a USB connection between the
computer and the HP LaserJet Pro MFP M475dw.  Command lsusb shows the
connection.

Command HP-setup finds the printer, but the last line says:

QXcbConnection: XCB error: 3 (BadWindow), sequence: 1323, resource id:
25167914, major code: 40 (TranslateCoords), minor code: 0

The result is that I cannot print.  Also neither xsane nor vuescan
will open.

I can however connect to both the printer and xsane via the ethernet,
but not vuescan.  As I prefer vuescan and can use it via the USB I
will use USB for the printer for both printing and scanning.

Advice and help appreciated.

Regards, Ken
-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQR9YFyM2lJAhDprmoOU2UnM6QmZNwUCXXPMkwAKCRCU2UnM6QmZ
N5PiAJ9DeHNYoxDJYqXuLpFokF0k4RvmUwCfR6tA7N9iTwzRUICO29fRNYnGgns=
=4auH
-END PGP SIGNATURE-

Date format for Thunderbird 60.7 -- partial success in changing it

[Ken Heard]

The latest version of Thunderbird for Debian Stretch, 70.7 which I now
use, still allows only the US date format, MM-DD-, but for me at
least expresses the time as HH:MM (24 hour clock).  In a partially
successful attempt to change the date format I did the following.

1. Ran update-locales 'LC_Time=en_DK.UTF-8'; file /etc/default/locale
now reads as follows:
LANG="en_CA.UTF-8"
LANGUAGE="en_CA:en"
LC_TIME=en_DK.UTF-8,
hereby confirming that en_DK.UTF-8 is my system wide time default.

2. 'Date and Time Formatting' options at 'Edit->Preferences->Advanced'
originally read as follows:

o Application locale: English (United States)
o Regional settings locale: English (Denmark)

The second of the two options was the default selection.  I tried both
options and then rebooted.  The result however the same; the date/time
format was MM-DD- HH:MM, the useual US date format, but the time in
24 hour format.

3.  Next, in file prefs.js (Edit->Preferences->Advanced->Config Editor)
I changed line 'intl.accept.languages' to read 'en-DK,en' instead of
'en-US,en'. Those two lines then read as follows:

o Application locale: und [whatever that means]
o Regional settings locale: English (Denmark) [no change]

Both options changed the desired date format to the SI standard
-MM-DD. However the time instead, of being HH:MM as before (24 hour
clock), was now a twelve hour clock format, complete with a.m. and p.m.
after the time as appropriate.

So, by exchanging the date format to the one I wanted I am forced to
have to put up with a retrograde format for the time.  While this result
is not completely satisfactory I reluctantly find the the price for the
SI date format in exchange for the retrograde time format worth paying.

In order to have both the date and time formats conform to the SI
standard, I tried one other change in the pref.js file, this time in
'extensions.superDateFormat.dateFormat" which had the following
date/time setup: "%Y-%m-%d %R', This format would be the correct one for
both time and date: -MM-DD 00:00, the latter in 24 hour format.  As
%r however is the symbol for 12 hour clock time, I exchanged the %R for
%H:%M.  The time still stayed at 12 hour.

Can anyone advise me as to how to have both the date and time formats
conform to the SI standard, or any other way to work around this
particular fault of Thunderbird 60.7?

Ken





signature.asc
Description: OpenPGP digital signature

Re: USB hard drives -- recommendations?

[Ken Heard]

On 26/01/19 03:06 AM, Celejar wrote:

On Fri, 25 Jan 2019 09:24:56 -0800
"James H. H. Lampert"  wrote:


Fellow List members:

Would anybody care to voice an opinion on USB external hard drives in
the 2 terabyte size range, for automated backup purposes?

We've been looking at the Seagate "Expansion" and the WD "Elements";
I've noticed that on Amazon, both have a fair number of negative reviews
citing reliability issues. (We recently discovered that our current
Seagate had apparently failed on us.)

Any opinions? Seagate? WD? Toshiba? Something else?


I so far am not having any trouble with something called "Seagate Backup 
Plus Portable Drive" -- "product of China (HDD), assembled in Thailand" 
as proudly proclaimed on the cases. These two I purchased in Thailand 
and am using them there. The spec sheet claims that they work with both 
USB 2.0 and 3.0, but likely with a lower transmission speed with 2.0. I 
am using them only on 3.0, not having tried either on 2.0.(1)


They are usable 'right out of the box' with Mac computers and also with 
Micro$oft, complete with built-in 'hardware' encryption if you choose to 
use it.  I of course chose not to use any of it and so reformatted both 
with ext4 and luks encryption.


They also come with a 'limited' three year warranty. I have only had 
them for a year; so far I have had no need to test those 'limitations'. 
Finally they come in six decorator colours, light and dark blue, red, 
gold, silver and black, useful to tell them apart if you have more than 
one.  I stuck to black for both of them as I have another way to tell 
them apart -- relative depth.


The height,width and depth of the 1 TB model are respectively 114, 77 
and 11 mm.  The height and width of the 4 TB model are the same; the 
depth however is 20 mm.  In view of the height and width the 
electro-mechanical drives used in these drive must be the 63.5 mm size. 
In view of the depth difference I suspect that there is one such drive 
in the

1 TB model, but two such drives in the 4 TB model.  Besides
the 1 and 4 TB models, also available are 2 and 5 TB models.

(1) https://www.seagate.com/as/en/consumer/backup/backup-plus/

Ken

Re: Thunderbird 60 ignores LC_TIME environment variable

[Ken Heard]

On 2018-10-02 3:07 p.m., Greg Wooledge wrote:

On Tue, Oct 02, 2018 at 02:47:52PM -0400, Ken Heard wrote:

$ LC_TIME=de_DE.utf8 thunderbird


This command doesn't work for me.


You have to generate the locales that you want to use.

dpkg-reconfigure locales >
Check it by running "locale -a" to see which locales are available on
your system.


I had long since already done that.  Among the available locales was and 
still is en_DK.utf8.


Ken

Re: Thunderbird 60 ignores LC_TIME environment variable

[Ken Heard]

On 2018-08-20 3:27 p.m., Stefan Pietsch wrote:

On 20.08.18 11:57, Curt wrote:

On 2018-08-20, Stefan Pietsch  wrote:

Dear list,

after the upgrade from Thunderbird 52.9.1 to 60.0 (Debian sid) it seems
for me that the environment variable LC_TIME is ignored.

I used the following command to set 24h time format in Thunderbird.

$ LC_TIME=de_DE.utf8 thunderbird


This command doesn't work for me.


With Thunderbird 60 setting the LC_TIME or LANG variables will not help.
Can someone confirm this?


I can't confirm anything but this bug report

https://bugzilla.mozilla.org/show_bug.cgi?id=1426907

looks both pertinent, and, after a cursory read (recent posts re 60 toward the
bottom), discouraging concerning your plight (at least for the moment).


Thanks for the hint to that bug id.

I had to set the "Date and Time Formatting" option (exists since TB
version 56) in the Advanced Preferences AND the variable LC_TIME when
starting Thunderbird to get it working.



Regards,
Stefan


The date and time option for me has two choices:

Application locale: English (United States)
Regional settings locale: English (Canada)

The result for me is the same for both. It is the ISO 8601 format for 
the date, but for time the format is 12 hour plus AM/PM.


How did you manage to the ISO 8601 time format be setting the LV_TIME 
format?


Regards, Ken

Calculator with "tapes"

[Ken Heard]

Does Debian have a calculator package which has the equivalent of the 
tape produced by mechanical machines to show the entire calculation.  I 
find such "tapes" essential when for example I am adding a long list of 
numbers and need to check after the addition is done to verify that all 
the numbers were entered correctly.


Regards, Ken

Display full date in Thunderbird

[Ken Heard]

I start Thunderbird with the following script:

#! /bin/bash
export LC_TIME=en_DK.utf8 && thunderbird "$@"

For emails sent or received today only the time is shown, e.g., 13:17.
For emails sent or received more than a week ago, the year, month, day 
and time are shown, e.g., 2018-06-13 07:43.


For emails sent and received a week after today only the year, month and 
time are shown, e.g., 2018/06 19:53.


How do I get Thunderbird to use the same format for dates/times more 
than a week old?


This problem appeared after upgrading from Jessie to Stretch with 
Thunderbird 52.8.0 in Stretch.  I used in Jessie the same script quoted 
above, where all dates after today were 2017-03-28 23:19.  (I do not 
remember which version of Thunderbird I had in Jessie.)


Regards, Ken

Re: I do not want to install Linux

[Ken Heard]

On 2018-02-08 15:57, Michael Fothergill wrote:

> Actually Kali Linux [1] is pretty cool and definitely worth a look.
> And, as a Debian derivative, it is a wonderful illustration of the
> things the Debian culture makes possible (another being Ubuntu, of
> course).
> 
>> ​I guess I could give it a try.  I have run Bodhi Linux.  That is quite
>> good.
> 
I would rather wait for Brahma and try to avoid Shiva, but I may have to
settle for Vishnu.

Ken




signature.asc
Description: OpenPGP digital signature

Cannot read my draft emails.

[Ken Heard]

In my Jessie box I recently converted from Icedove to Thunderbird
52.2.1.  As part of that conversion I also upgraded Enigmail to version
2:1.8.2-4~deb8u1 which is in the Thunderbird extensions list.

Before the conversion I never encrypted my emails, but I did digitally
sign them.  I expected to be able to do the same with Thunderbird and
presumably so far I nave done so.

I now in Drafts emails which I now want to edit and send.  When I try to
open them for editing in the text section nothing appears except this
message:

This is an encrypted OpenPGP message.
In order to decrypt this mail, you need to install an OpenPGP add-on.

In addition to Enigmail I also have  gnupg 1.4.18-7+deb8u3,
gnupg-agent 2.0.26-6+deb8u1 and gnupg2 installed, as well as six
packages with pgp in their names.  I find it however strange that
apt-cache no longer finds those packages. I also find it strange none of
the options in the Thunderbird Enigmail menu is operative.  If something
else is needed, what would it be?

In short, can anyone tell me what is going on?  I want my email client back!

Regards, Ken Heard

Re: Installed Stretch, but no command line

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2017-07-23 19:08, Felix Miata wrote:
> Ken Heard composed on 2017-07-23 18:25 (UTC-0400):
> 
>> Felix Miata wrote:
> 
>>> Ken Heard composed on 2017-07-23 12:21 (UTC-0400):
> 
>>>> Ctrl-alt-F1 shows the blank screen which appears at the end
>>>> of the boot.  I think the problem is that what appears on
>>>> opening tty2 to tty5 should appear in tty1 but for some
>>>> reason does not -- instead the blank screen.
> 
>>> /etc/systemd/system/getty.target.wants/getty@tty1.service
>>> probably says
> 
>>> TTYVTDisallocate=yes
> 
>>> Changing to
> 
>>> TTYVTDisallocate=no
> 
>>> should on subsequent boots leave init messages on tty1, but 
>>> plymouth may need to be purged for that to happen.
> 
>> I tried that, but after I entered my /home password the boot
>> continued and then stopped.  After waiting fifteen minutes or so
>> I concluded that the machine hanged.
> 
> All my installations have that set to no, so it's very surprising
> to me that this could happen.
> 
>> I tried using the recovery mode from GRUB, but that hahged the
>> machine as well.  I then tried rescue mode from the installation
>> disk.  If it is possible to use that to put the Yes back, I could
>> not figure it out.
> 
>> What do I do now?
> 
> If you had made a copy before your edit it would be a simple matter
> to rename it. :-(
> 
> That said, that file is a plain text file. Any editor available in
> rescue mode can put it back once the Stretch / partition is
> mounted. Try nano first, which for most people is much more
> intuitive than traditional Linux editors like vi.
> 
> Something to try first: append plymouth.enable=0 to the kernel
> cmdline for an otherwise normal boot. Also try appending a 3, which
> should block startup of the GUI login manager, but also should make
> normal ttys 1-6 available.
> 
All your suggestions require that I be able to boot the computer and
log on as root from a tty, but as already explained I cannot do that.
 For rescue mode I need access to the original /etc directory.  I
looked for a way to do that but could not find one.


-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAll1Z4kACgkQlNlJzOkJmTcEBwCfVxnS23IHBAGapqU101rroo1x
kz8An0btbyvXRwRIgG5MM7P9EW6wap5s
=mJ6U
-END PGP SIGNATURE-

Re: Installed Stretch, but no command line

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2017-07-23 16:12, Felix Miata wrote:
> Ken Heard composed on 2017-07-23 12:21 (UTC-0400):
> 
>> Ctrl-alt-F1 shows the blank screen which appears at the end of
>> the boot.  I think the problem is that what appears on opening
>> tty2 to tty5 should appear in tty1 but for some reason does not
>> -- instead the blank screen.
> /etc/systemd/system/getty.target.wants/getty@tty1.service probably
> says
> 
> TTYVTDisallocate=yes
> 
> Changing to
> 
> TTYVTDisallocate=no
> 
> should on subsequent boots leave init messages on tty1, but
> plymouth may need to be purged for that to happen.

I tried that, but after I entered my /home password the boot continued
and then stopped.  After waiting fifteen minutes or so I concluded
that the machine hanged.

I tried using the recovery mode from GRUB, but that hahged the machine
as well.  I then tried rescue mode from the installation disk.  If it
is possible to use that to put the Yes back, I could not figure it out.

What do I do now?

Ken


-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEUEARECAAYFAll1IkwACgkQlNlJzOkJmTdyUACfR77Krz5cbuUvDQVRsr1wosfR
9H8AmMfXIjk7JFrg2i8J6TrSM4xNkik=
=HXHr
-END PGP SIGNATURE-

Re: Installed Stretch, but no command line

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2017-07-23 07:43, songbird wrote:
> Ken Heard wrote:
>> 
>> Just finished installing Stretch, using a RAID1, LVM and
>> encryption for home (passkey) and swap (random key).  It is at
>> the moment a basic installation with X-server-Xorg, but no DE
>> yet.
> 
> if you use the control-alt F1 - F? sequence do you get a login
> prompt?  (check F7 and F8 too)

I should have thought of that, but when I wrote my original post I was
tired and frustrated.  Thanks for the reminder.

This morning I did as you suggested.  Yes, ctrl-alt-F2 to F6 does open
a console where I can log in one of them as my user and in another as
root.  So far I seem to be able to execute all the commands I need to
or want to in both.

For example, in my user I was able to run startx, which prodices a
bluish hue screen with the Debian swirl. I can't go any further
because I do not have a desktop environment yet.  My next step is to
install one and see what happens.

Ctrl-alt-F7 and F8 open nothing, not surprising as F7 and above are
normally reserved for graphical user interfaces (GUI), which I have
not yet installed.

Ctrl-alt-F1 shows the blank screen which appears at the end of the
boot.  I think the problem is that what appears on opening tty2 to
tty5 should appear in tty1 but for some reason does not -- instead the
blank screen.
> 
>> As part of the boot it accepts my /home password, but I do not
>> get the command line. I end up with a blue-grey screen. No input
>> from mouse or keyboard is accepted.  To close the computer I have
>> to use the power switch.
> 
> oh, this doesn't sound good, but since you don't describe your
> keyboard or mouse type i'm not sure if you need to change something
> in the bios or what for them to show up.  check the bios if you
> have settings that might help.  use a different keyboard and/or
> mouse to check that something shows up.

Keyboard is fine; it has worked on another computer,  I have nothing
yet installed which uses the mouse.

> your X setup may need adjusting.  check /var/log/Xorg.0.log and
> X*.log.old  if you have no log files install rsyslog.

Probably not necessary, but I shall see after the DE is installed.

>> I can open it in recovery mode, and after I enter the root
>> password I can get the command line as root. But I don't know
>> what to do next to effect a rescue.  From the root command line I
>> can however close the computer properly with "shutdown -h now".
>> 
>> I did notice when the computer was booting normally -- not in
>> rescue mode -- the following lines loop many times until the
>> computer gives up:
>> 
>> Running /scripts/local-block Failed to connect to lvmetad Failed
>> to find logical volume SOL1/swap_crypt
>> 
>> According to several posts on line, this behaviour is a nuisance
>> but not fatal.  As some people suggested on line I tried to get
>> rid of this nuisance by changing in /etc/lvm/lvm.conf
>> use_lvmetad=1 to =0, but to no effect.
>> 
>> If I have to I can live with the nuisance, but I cannot live
>> without the command line.
>> 
> i do know that at times some partitions are set up but may not be
> found due to various issues.  i make life simpler by using volume
> labels and adjusting /etc/fstab to use those instead of UUIDs.

Probably a good idea; I shall try it.

> the problem i had was that the reformat of the swap partition got
> rid of my label.  so now if i do anything that touches the swap
> partition i reset the label.

Regards, Ken


-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAll0zREACgkQlNlJzOkJmTekcgCfeFjlPRPXaxhhJq4TPwvLfl6K
0WwAmwb+xo26JnNg0/j4m40IL+ZJe7Hn
=2QF1
-END PGP SIGNATURE-

Installed Stretch, but no command line

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Just finished installing Stretch, using a RAID1, LVM and encryotion
for home (passkey) and .  It is ant the moment a basic installation
with X-server-Xorg, but no DE yet. As part of the boot it accepts my
/home password, but I do not get the command line. I end up with a
blue-grey screen. No input from mouse or keyboard is accepted.  To
close the computer I have to use the power switch.

I can open it in recovery mode, and after I enter the root password I
can get the command line as root. But I don't know what to do next to
effect a rescue.  From the root command line I can however close the
computer properly with "shutdown -h now".

I did notice when the computer was booting normally -- not in rescue
mode -- the following lines loop many times until the computer gives up:

Running /scripts/local-block
Failed to connect to lvmetad
Failed to find logical volume SOL1/swap_crypt

According to several posts on line, this behaviour is a nuisance but
not fatal.  As some people suggested on line I tried to get rid of
this nuisance by changing in /etc/lvm/lvm.conf use_lvmetad=1 to =0,
but to no effect.

If I have to I can live with the nuisance, but I cannot live without
the command line.

Help please, anyone.

Regards, Ken

-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAll0LEYACgkQlNlJzOkJmTfDZQCeLFHAuLm4OIeuMk3kVJH089PG
OGQAn2Zhgpm3Oyb5wPwkSFnOG5c5opxX
=IRii
-END PGP SIGNATURE-

Stretch installer: cannot encrypt /home

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I want to encrypt my /home directory but the Stretch installer will
not let be do it.  I got so far as to create a RAID1 and then lvm2 to
create several logical volumes, but I could not encrypt the logical
volume for the /home directory.

In the overview of my partitions the one for my home directory reads
as follows:

LVM VG SOL1, LV home 2.0 TB lINUX device-mapper  (linear)
#1  2.0 TBK crypto (SOL1-home_crypt)
Encrypted volume (SOL1-home_crypt) - 2.0 TB Linux device-mapper (crypt)
#1  2.0 TBK   lvm

What I think I should be able to do now is to select the last line
above so I can enter the file system. the password and the mount point
/home.  When I select that line nothing happens,

If I select "configure encrypted volumes", I am asked to save my
partitioning scheme, After doing so another window appears with two
choices, "Finish" and "Create encrypted volumes".  Selecting the
latter asks me to select the devices to encrypt, but the one I want to
encrypt does not appear there.

Either way installer will not let me go any further.

Have others had this problem?  If anyone found a solution I would
appreciate knowing how you did it.

Regards, Ken Heard





-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAllIXiUACgkQlNlJzOkJmTft3ACdGhTQ0iWyx3g1FKQCenD4Hz3W
hTAAn1tnH16y78wlc19PpvsMhTwQdvXN
=VR1d
-END PGP SIGNATURE-

Re: why can't I visit this web site

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 06/06/17 08:44 AM, rhkra...@gmail.com wrote:
> Just as a data point, it seems to work for me in Wheezy / Iceweasel
> / Firefox ESR--I see lots of content, but all in a foreign language
> using--what's the right word?--pictographs?--those "picture
> characters" used in several far east languages (e.g., Chinese).

I think the word you want is one of logogram, logograph or
grammalogue, meaning in all three cases a word represented by a single
sign.

Ken


-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAlk3GSoACgkQlNlJzOkJmTcy6gCffYjIKtMZwmgMIhTtm71NsB2C
HGoAnAq34TyLVeH+FRE1OkNmlK3doom4
=4UlJ
-END PGP SIGNATURE-

Re: Skype

[ken]

On 02/07/2017 03:25 PM, Anonymous wrote:

Skype on Linux is terrible and crashes randomly, even after fixing the
Pulseaudio issue. No other software I've ever used on Linux causes as
many problems as Skype. Obviously this problem ultimately comes from
Microsoft, but nevertheless many of us need Skype for communications
with colleagues running a non-Linux OS since it is the overwhelming
market leader in VOIP; this is especially common in a business environment.

Skype needs to run on Linux with the same stability that it does on
Windows and MacOS, WITHOUT resorting to installing WINE, and preferably
without relying on WebRTC.

If you could develop a non-Skype Linux client with the ability to easily
communicate on the Skype platform --- something open source but which
has compatibility with Skype, coordinated with Microsoft --- that would
be ideal. But relying on Microsoft to develop good software for Linux
has proven to be a losing proposition and I'm sure is stifling the
adoption of Linux by a wider audience.

Having frequent connection drop-outs with Skype on different Linux and 
Android platforms, I switched over to Hangouts, a Google product.  It's 
better, buy still not as good as I'd like.  Now I'm trying to migrate 
over to Signal, a Whisper Systems product, but as with all of them, 
people on both ends need to use the same product and there aren't yet a 
lot of people using Signal.  More info about it at 
http://www.linux-magazine.com/Online/Features/Signal-Private-Messenger 
and https://whispersystems.org/

Re: What file system to use?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2017-02-03 03:03, Ben Caradoc-Davies wrote:
> On 02/02/17 20:58, Ken Heard wrote:
>> I use USB flash drives and 480 GB SanDisk portable drives for
>> some backups. I started by using ext4 on one of the flash drives.
>> The next day I tried to boot the computer, but found that the
>> computer hung. Only at this point did I find out that those
>> drives are not capable of journalling.
> 
> I use USB flash drives with ext4 for backups and have not had any 
> problems with journalling. I have had at least one successful
> journal recovery, after an inadvertent disconnection before
> unmount. What problems did you encounter?
> 
> I optimised my ext4 backup filesystems for a small number of large
> files (compressed archives) with:
> 
> mkfs.ext4 -J size=4 -m 0 -T largefile4 -O "^resize_inode"
> /dev/sdb1 tune2fs -c 0 -i 0 -L Backup /dev/sdb1
> 
> My /etc/fstab contains:
> 
> LABEL=Backup /media/backup ext4
> noatime,noauto,user,errors=remount-ro 0 0
> 
> Kind regards,
> 
First I should say that the result I described in my original post
occurred in a computer with Debian Wheezy as the OS.  I do not know
whether the result would be the same with Jessie, and frankly I am not
really anxious to find out.

Second, all these drives were encrypted with cryptsetup.  Encryption
plus ext4 with journalling may have been too much to ask of the
operating system.

Third, the last thing I did the evening before the morning when the
computer crashed was to transfer a series of uncompressed jpg files to
a USB flash drive.  I assumed that the transfer was finished when the
bang reappeared on the command line; so I shut down the computer. I
did not unmount the drive before shutdown.  It may be that the
journalling requirements for the transfer were never finished when the
bang appeared, and were still unfinished at shutdown, thereby causing
the computer not to boot the next morning. I will never know now.  I
did of course have to reinstall the OS starting on that day.

Finally, I discovered that trying to open an encrypted USB flash drive
in Dolphin caused Dolphin to crash.  I had to use Konqueror instead.
I filed a bug report for the crash. I have never checked to see
whether the bug was fixed.

All the events described in my original post and in this one took
place in December 2014.  I continue since then to use Konqueror for
such encrypted drives with ext2.  This arrangement works for me, and
my curiosity does not extend so far as to find out whether I still
need to use it three years later.

Regards, Ken Heard



-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAliUBsUACgkQlNlJzOkJmTdZegCePvnEqZGsP+old2vKryH+Z60C
UzEAmwfpk6kBwCj7upU/uEfV4KF9GxjC
=K72B
-END PGP SIGNATURE-

Re: Android phone access in Dolphin does not work

[ken]

On 01/26/2017 11:42 PM, Gary Dale wrote:

On 25/01/17 01:27 PM, Tony Baldwin wrote:



On 01/25/2017 12:58 PM, ken wrote:

On 01/24/2017 02:38 PM, Gary Dale wrote:

I'm running stretch/AMD64 and I'm having extreme problems accessing my
Samsung S5 phone from my Plasma5 desktop.

When I plug the phone in, I get the notification with 2 suggested
actions. They both read "Open with File Manager" but the first one
opens Dolphin with "camera:..." while the second opens Dolphin with
"mtp:..." in the location bar. When I actually try to navigate to the
folder with my photos, I get a notification on the phone telling me a
device is attempting mtp access... When I click "Allow", I get another
notification pop-up on my desktop with the same two options...

When I actually try to access the photos, I frequently get "process
died" or "I/O error" messages, although sometimes I can actually see a
photo (not just the file names) - persistence pays off. Moreover the
phone will lock when connected, even if a file transfer is underway.
This kills the transfer.

To make matters worse, I have photos and videos that I have taken
which show up in the Gallery on the phone and which I can view on the
phone and see when I remove the SD card and read it directly but which
don't show up in Dolphin.

Clearly this is not the way the process is supposed to work. I can
understand the phone requesting a verification for the connection, but
everything on the desktop side seems wrong:
- the two actions should be distinguishable by their description,
- the process shouldn't die and if it does, it should relaunch,
- the connection should block the phone from locking if locking will
stop file transfers,
- there should be a record of the error in a log somewhere. As it is,
all I can find is a lot of systemd journal entries like "Jan 24
14:01:33 transponder kernel: usb 4-5: usbfs: process 20598 (mtp.so)
did not claim interface 0 before use" which seem to occur before the
device is actually recognized.
- all the photos should show up


It sounds like you want to offload files from your android. For a few
years now I've been using Software Data Cable to do this, free app, no
cable needed.  Launch the app and it sets up an ftp server on the
phone.  Then from my laptop I log in, navigate, up- and download files.
When done, I shut down the ftp server on the phone. Easy-squeezy.




Just in case the above doesn't work out for you (and I'm going to 
look into that one, myself), because,  in my experience, this 
disaster called MTP is simply a headache...
I'm going to suggest what I do to access storage on my Motorola Droid 
II.

Install Dropbox.
Alternatively, if you have access to any remote server running an 
ftpd, AndFTP is an option. I've done both, and currently find the 
dropbox most convenient.


Tony


I'm going to hazard a guess that this problem is Linux specific since 
the hordes of Android users would rebel if this was happening on 
Windows. However I can't understand why it should be such an issue 
since the MTP is an open specification that has been around since 2008 
and Google is heavily involved with it.


The Dropbox idea sounds reasonable if I don't mind putting massive 
amounts of data onto the Internet. I think I'll try the software data 
cable instead.


Thanks.

Just a quick tip for using Software Data Cable:  When you fire it up on 
your android, you'll get a screen helpfully telling you the ftp command 
to run on the other end... but the command they give isn't right... 
well, maybe it is for Windows, I don't know. What's worked on Linux for 
me dozens of times is this:


$ ftp [phone's IP address] [port number]

Software Data Cable uses a non-standard port for its ftp service... a 
good thing, since ftp isn't the most secure protocol in town.  Note that 
there's a space between the ip and port#... also the first arg is just 
numbers.


I wish there was an ssh-based app for android... I've looked for such, 
haven't found it.  Is that a Windows thing?

Re: What file system to use?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2017-02-02 08:47, Gary Dale wrote:

> There shouldn't be any problem with ext4. It's been the Linux
> standard for years now. I can't even imagine what the backup issue
> would be unless they were trying to make some kind of sparse backup
> using FS internals.

I use USB flash drives and 480 GB SanDisk portable drives for some
backups. I started by using ext4 on one of the flash drives. The next
day I tried to boot the computer, but found that the computer hung.
Only at this point did I find out that those drives are not capable of
journalling.

So I either had to use ext4 with the journalling feature disabled or
revert to ext2 for these drives.  I chose the latter, as simpler and
less susceptible to errors.  I have had no trouble with them since.

Regards, Ken Heard





-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAliS5pkACgkQlNlJzOkJmTcGcgCfXV7MzOXuKKGfjPYeiGlyqzTl
HOkAnjRazo1ab744SF3PPNdLO1V8sJKm
=Dyv9
-END PGP SIGNATURE-

Re: Android phone access in Dolphin does not work

[ken]

On 01/24/2017 02:38 PM, Gary Dale wrote:
I'm running stretch/AMD64 and I'm having extreme problems accessing my 
Samsung S5 phone from my Plasma5 desktop.


When I plug the phone in, I get the notification with 2 suggested 
actions. They both read "Open with File Manager" but the first one 
opens Dolphin with "camera:..." while the second opens Dolphin with 
"mtp:..." in the location bar. When I actually try to navigate to the 
folder with my photos, I get a notification on the phone telling me a 
device is attempting mtp access... When I click "Allow", I get another 
notification pop-up on my desktop with the same two options...


When I actually try to access the photos, I frequently get "process 
died" or "I/O error" messages, although sometimes I can actually see a 
photo (not just the file names) - persistence pays off. Moreover the 
phone will lock when connected, even if a file transfer is underway. 
This kills the transfer.


To make matters worse, I have photos and videos that I have taken 
which show up in the Gallery on the phone and which I can view on the 
phone and see when I remove the SD card and read it directly but which 
don't show up in Dolphin.


Clearly this is not the way the process is supposed to work. I can 
understand the phone requesting a verification for the connection, but 
everything on the desktop side seems wrong:

- the two actions should be distinguishable by their description,
- the process shouldn't die and if it does, it should relaunch,
- the connection should block the phone from locking if locking will 
stop file transfers,
- there should be a record of the error in a log somewhere. As it is, 
all I can find is a lot of systemd journal entries like "Jan 24 
14:01:33 transponder kernel: usb 4-5: usbfs: process 20598 (mtp.so) 
did not claim interface 0 before use" which seem to occur before the 
device is actually recognized.

- all the photos should show up


It sounds like you want to offload files from your android.  For a few 
years now I've been using Software Data Cable to do this, free app, no 
cable needed.  Launch the app and it sets up an ftp server on the 
phone.  Then from my laptop I log in, navigate, up- and download files.  
When done, I shut down the ftp server on the phone.  Easy-squeezy.

Wheezy no longer accepts my passwords to open /home and /mnt partitions on boot

[Ken Heard]

In my Wheezy box I have two encrypted hard drive partitions, 
/dev/mapper/md07_crypt for /home and /dev/mapper/md05_crypt for /mnt. (Mnt is 
no longer used. That partition was originally for /tmp; in a weak moment I 
persuaded myself that I needed to encrypt /tmp. I have since changed the mount 
point for /tmp to tmpfs and consequently changed the mount point of 
/dev/mapper/md05_crypt to /mnt pending its deletion. In the meantime it still 
has to be opened.)

This morning, when I tried to boot this box, I found that the passwords for 
those two partitions were not accepted. Finally, after entering the correct 
passwords numerous times, the following messages were received. 

[info] Loading kernel module loop.
[info] Loading kernel module coretemp.
[info] Loading kernel module it87.
[ok] Activating lvm and md swap ... Done
[] Checking file systems ... Fsck from util-linux 2.20.1
BOOT was not cleanly unmounted, check forced.
BOOT: 245/120960 files (20.4% non-contiguous) 58505/241664 blocks
Fsck.ext4 No such file or directory while trying to open /dev/mapper/md07_crypt.
Possibly non-existent device?
VAR: recovering journal
Fsck died with exit status 9
Failed (code 9).
[FAIL] File system check failed. A log is being saved in /var/log/fsck/checkfs 
if that location is writable.
{My note: that location was not writable; so no such log was created.}
Please repair the system manually ... Failed!
[warn] A maintenance shell will now be started. CONTROL-D will terminate the 
shell and resume system boot ... (warning)
Give root password form maintenance
(or type CONTROL-D to continue):

After entering the root password the root bang appeared.

At this point I did not know what to do by way of reparing the system manually. 
I consequently closed the box, first by running "shutdown now". After various 
messages flashed by on the monitor I was asked once more to give the password 
for maintenance or type CONTROL-D again. After typing CONTROL-D again my user 
login appeared, but of course I could not log in. At that point I was finally 
able to close the computer with the Alt-S command.

I hope that somebody or bodies can tell me what to do to make this computer 
usable again -- short of having to do a completely new installation.

Regards, Ken Heard

-- 

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

DBus error: how to fix.

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I last upgraded my Wheezy installation on 2016-11-11. Ever since,
whenever I run "apt-get update" or "apt-get upgrade", at the end the
following message appears:

Error: GDBus.Error:org.freedesktop.DBus.Error.Spawn.ExecFailed: Failed
to execute program /usr/lib/dbus-1.0/debus-daemon-launch-helper: Success

That directory exists and the script file (if that is what it is) is
also exists. Is there something that script file should do but cannot
do for some reason?

Ken Heard


-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlgxc20ACgkQlNlJzOkJmTeX6ACfdoxwiKVBWSb8PDhVMuPFokD2
x24AnifiEwAXyHboKORmCWV64cBeKCir
=KumC
-END PGP SIGNATURE-

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

Extending file system after enlarging a logical volume

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

In my jessie box I need to increase the non-encrypted root partition
(/) size from 14G to 19G.  I was able successfully to increase the
logical volume (and decrease the space available in the volume group
accordingly) but am not sure how to do the next step.

The following command produced the result shown.

root@BDS:~# fsadm resize /dev/BDS1/root
resize2fs 1.42.12 (29-Aug-2014)
Filesystem at /dev/mapper/BDS1-root is mounted on /; on-line resizing
required
old_desc_blocks = 1, new_desc_blocks = 2
resize2fs: Read-only file system While checking for on-line resizing
support
fsadm: Resize ext4 failed

Does the phrase "on-line resizing required" mean that I have to
unmount /dev/BDS1/root, then run the fsadm command and finally remount
/dev/BDS1/root?  Is unmounting the basic partition (/) safe while
presumably some of the files there are being used by the computer?

I think I read somewhere that to resize an ext4 partition did not
require unmounting it.  Perhaps it it is required.

Regards, Ken Heard
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAlfyvooACgkQlNlJzOkJmTcX4ACcCD/QlTflgyvQv1dQ/NAKdYHT
CZkAn039JkY+ZoJeOxi1QQre8gx7xRqI
=E0Wk
-END PGP SIGNATURE-

Re: Typing Cyrillic script with a UK keyboard in an en-gb setting

[ken]

On 09/24/2016 10:10 AM, Lisi Reisz wrote:

My husband has just asked to do this.  His system is vanilla from this point
of view.  (Mine is in a mess, with a messed-up scim and no foreign
fonts "working", but that is another story.)

Advice please on the best way to achieve this for him.  I.e., what do those of
you doing this or similar find works comfortably.

Thanks,
Lisi

This may be of some help: 
http://linuxmanagers.blogspot.com/2016/09/using-multiple-language-keyboards-in.html

Mount /tmp on tmpfs jessie and stretch -- howto?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I would like in my jessie and stretch boxes (one of each, both with
systemd) to mount /tmp on tmpfs instead of a hard drive partition tmp
or /dev/mapper/SOL1-tmp.  I assumed that to do so I could not have
either of those partitions; so I unmounted them and then deleted them
as a partition or as part of a LVM as the case may be.

I then created a created a line in /etc/fstab as follows:
tmpfs   /tmp/   tmpfs   nodev,nosuid,size=20%   0   0.  When I rebooted I
received dependency errors.

In addition to the line above /etc/fstab still had a line
/dev/mapper/SOL1-tmp/tmpext2defaults0   2.

However, df shows the first line (tmpfs /tmp etc.), but not the second
(/dev/mapper/SOL1-tmp etc.)

Obviously something is wrong but what?  I know that file
/etc/default/tempfs cannot be used in systemd, but I am at a loss to
know what if anything replaces it in systemd.  Apparently it is not
enough to add the tmpfs /tmp tmpfs etc line in /etc/fstab or run
"systemctl enable tmp.mount".
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAlfGHiUACgkQlNlJzOkJmTceiACcCZfxiTfMOQ1i8n/gm1ZBQvTB
rf8AnjBT3QkxzIk85aLQmZc9Yw3wMoJR
=wWwi
-END PGP SIGNATURE-

Re: Failure messages relating to strict.pm -- shutdown

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-08-25 16:31, Ken Heard wrote:

> Shutdown however does not seem to work.  Selecting turning off the
> computer from the DE closes the DE and opens tty1 which is ready
> for a login.  Logging in as root and running "shutdown now" has no
> effect.  I have to shut down by holding the start button down for
> four seconds.

I did determine that package systemd-sysv, which contains the command
/sbin/shutdown, is installed, but for some reason /sbin/shutdown does
not work.

> Another thing I find strange -- after booting and opening my user
> only tty1 and tty7 are available.  Tty2 through tty6 do not exist.
> 
> Regards, Ken
> 
> 

-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAle/YuUACgkQlNlJzOkJmTdc9QCffiRhzKCLTS+5sGlFw2WcCDYc
Q94Ani6C1bj2DZDx0g3i5bV8+jrNNTRY
=zbgB
-END PGP SIGNATURE-

Re: Failure messages relating to strict.pm

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-08-25 07:39, Kushal Kumaran wrote:

> Looks like you have a local (broken) perl installation in
> /usr/local that is interfering with the
> /usr/bin/deb-systemd-helper.
> 
> deb-systemd-helper has a shebang line of #!/usr/bin/env perl, which
> is picking up the first perl available in $PATH, which is 
> /usr/local/bin/perl in your case.
> 
> Remove /usr/local/bin from your PATH and try again.  If you have
> the PERLLIB or PERL5LIB environment variables set, unset them as
> well.
> 
> # export PATH=/usr/bin:/usr/sbin:/bin:/sbin

Did that.

> # unset PERLLIB # unset PERL5LIB

Neither of these variables existed, but I unset them anyway.

Running apt-get update && apt-get dist-upgrade still produces the same
error messages quoted in my original post.

It seems that the downloaded packages are not unpacked or installed.
The first package in the list to be unpacked is apt; it seems to be
the one which causes the error messages. Since this package cannot be
unpacked the whole process is aborted -- none of them is unpacked or
installed.

On the other hand the DE now appears to be stable.  Shutdown however
does not seem to work.  Selecting turning off the computer from the DE
closes the DE and opens tty1 which is ready for a login.  Logging in
as root and running "shutdown now" has no effect.  I have to shut down
by holding the start button down for four seconds.

Another thing I find strange -- after booting and opening my user only
tty1 and tty7 are available.  Tty2 through tty6 do not exist.

Regards, Ken

-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAle/VZkACgkQlNlJzOkJmTcODwCdFH8L7Ci6+5/O54yfxqPvOl3A
w6YAnjYPxHTDBJMjniN9+NCEdALRgAWg
=4RoE
-END PGP SIGNATURE-

Failure messages relating to strict.pm

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

At the end of a package update and upgrade today of Stretch the
following messages appeared.

apt-listchanges: Mailing root: apt-listchanges: news for SOL
debconf: Perl may be unconfigured (Can't locate strict.pm:
/usr/local/lib/x86_64-linux-gnu/perl/5.22.2/strict.pm: Permission
denied at (eval 1) line2.
BEGIN failed--compilation aborted at (eval 1) line 2.) -- aborting
(Reading database ... 76727 files and directories currently installed.)
Preparing to unpack .../libapt-pkg5.0_1.3~rc2_amd64.deb ...
Unpacking libapt-pkg5.0:amd64 (1.3~rc2) over (1.3~rc1) ...
Setting up libapt-pkg5.0:amd64 (1.3~rc2)
(Reading database ... 76727 files and directories currently installed.)
Preparing to unpack .../libapt-pkg5.0_1.3~rc2_amd64.deb ...
Unpacking libapt-pkg5.0:amd64 (1.3~rc2) over (1.3~rc1) ...
Can't locate strict.pm:
/usr/local/lib/x86_64-linux-gnu/perl/5.22.2/strict.pm: Permission
denied at /usr/bin/deb-systemd-invoke line 52.
BEGIN failed--compilation aborted at /usr/bin/deb-systemd-invoke line 52
.
dpkg: warning: subprocess old pre-removal script returned error exit
status 13
dpkg: trying script from the new package instead ...
Can't locate strict.pm:  Permission denied at
/usr/bin/deb-systemd-invoke line 52.
BEGIN failed--compilation aborted at /usr/bin/deb-systemd-invoke line 52
.
dpkg: error processing archive
/tmp/apt-dpkg-install-GpvFkF/1-apt_1.3rc2_amd64.deb (--unpack):
   subprocess new pre-removal script returned error exit status 13.
Can't locate strict.pm:
/usr/local/lib/x86_64-linux-gnu/perl/5.22.2/strict.pm: Permission
denied at /usr/bin/deb-systemd- helper line 83.
BEGIN failed--compilation aborted at /usr/bin/deb-systemd- helper line 8
3.
Can't locate strict.pm:
/usr/local/lib/x86_64-linux-gnu/perl/5.22.2/strict.pm: Permission
denied at /usr/bin/deb-systemd- helper line 83.
BEGIN failed--compilation aborted at /usr/bin/deb-systemd- helper line 8
3.
Can't locate strict.pm:
/usr/local/lib/x86_64-linux-gnu/perl/5.22.2/strict.pm: Permission
denied at /usr/bin/deb-systemd-helper line 83.
BEGIN failed--compilation aborted at /usr/bin/deb-systemd- helper line 8
3.
Can't locate strict.pm:
/usr/local/lib/x86_64-linux-gnu/perl/5.22.2/strict.pm: Permission
denied at /usr/bin/deb-systemd-invoke line 52.
BEGIN failed--compilation aborted at /usr/bin/deb-systemd-invoke line 52
.
Errors were encountered while processing:
  /tmp/apt-dpkg-install-GpvFkF/1-apt_1.3rc2_amd64.deb
E: Sub-process /usr/bin/dpkg/ returned an error code (1)

Can anyone tell me what the foregoing means?  If it is of any help, I
do have two problems with Stretch.  First I cannot close the computer
using shutdown -- I have to close it by holding down the start button
for four seconds.

Second, after opening my user in the DE, the DE remains on the monitor
for only a few seconds.  The monitor then reverts to the tty1 command
line and asks me to log in.  I can do so either as root or my user but
am restricted to use of the usual commands.

Regards< Ken Heard

-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAle92+IACgkQlNlJzOkJmTeh5ACfQTV1v1+PHiE1odUMGo+WUTpf
PcEAnRva7BRqOoNR+B20/TbsrzrQ8rhQ
=pkJe
-END PGP SIGNATURE-

*ERROR* Unable to find select subconnector property

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

When I opened my Jessie box today I for once ran dmesg.  The last ten
lines thereof read as follows:

[   45.377328] NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state
recovery directory
[   45.384608] NFSD: starting 90-second grace period (net
818ba200)
[   53.192364] [drm:select_subconnector_show] *ERROR* Unable to find
select subconnector property
[   53.232210] [drm:subconnector_show] *ERROR* Unable to find
subconnector property
[   68.261234] [drm:select_subconnector_show] *ERROR* Unable to find
select subconnector property
[   68.295298] [drm:subconnector_show] *ERROR* Unable to find
subconnector property
[   74.927462] [drm:select_subconnector_show] *ERROR* Unable to find
select subconnector property
[   74.958852] [drm:subconnector_show] *ERROR* Unable to find
subconnector property
[   80.850145] [drm:select_subconnector_show] *ERROR* Unable to find
select subconnector property
[   80.885065] [drm:subconnector_show] *ERROR* Unable to find
subconnector property

Are these errors something about which I need to concern myself?

Regards, Ken Heard



-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAle7GdYACgkQlNlJzOkJmTeupgCePlwbc0byNQeHUK8Hrlk3Yjy1
1MUAn14NjJUzQdLF6LEFrWmlKi9vG+DU
=VT3D
-END PGP SIGNATURE-

Re: Using sudo in a script.

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-08-03 15:16, Jude DaShiell wrote:
> Why not try a -H switch on the sudo line in your script and see
> what happens?

I did.  The answer was nothing.

Regards, Ken

-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAleiYfsACgkQlNlJzOkJmTcD7wCbB/PgXclQY8xzuUAov+I37DaR
2qUAnj8XHWQADZzDmIyB6QopXDRsMYaO
=6LWe
-END PGP SIGNATURE-

Re: Using sudo in a script.

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-08-03 12:21, Ken Heard wrote:
> I would like to create a short script to mount and unmount SCXC
> cards. For these exfat cards to do either in Jessie has to be done
> as root.
> 
> I have set up sudo so that my user can run any command as root.
> the following command will mount the card without asking for my
> password.
> 
> sudo mount -U 6238-3434 /media/xca
> 
> (My file /etc/fstab contains the following line:
> 
> UUID-6238-3434  /media/xca  exfat   rw,users,noauto,noatime 0
> 0)
> 
> In order to automate the process I wrote the following script named
> mxca :
> 
> #!/bin/bash # Script to mount xca from any directory. CURPWD=$PWD 
> cd /home/ken sudo mount -U 6238-3434 /media/xca cd $CURPWD
> 
> Running mxca alone does not produce the required result.  Running
> sudo mxca together however does, even though the sudo command is
> also in the script.
> 
> I can unmount the card by running command sudo umount /media/xca,
> but I would like a similar script named uxca to do the same thing
> for unmounting.
> 
> Is there a way to make sudo do its thing in these scripts so that
> I can mount and unmount the SDXC card by running the script
> commands mxca and uxca by themselves?

I have since discovered a command "/sbin/mount.fuse" and "fusermount
- -u" which are supposed to allow a non-privileged user to mount and
unmount exfat devices; but I haven't figured out how to use them yet.
 A simple "/sbin/mount.fuse /media/xca" or "fusermount -u /media/xca"
do not work.  Other than fixing the format of the UUID device indicate
from UUID- to UUID= I have yet to experiment with Don Armstrong's
other suggestions.  I thing I would rather get /sbin/mount.fuse and
fusermount -u to work instead.

Regards, Ken


-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAleiYSoACgkQlNlJzOkJmTe4fwCcDvq8eBcq0I7guGdmmhN23sgP
3KEAn0LJKJzJUlIJTD0jMMt3T4CyXWAw
=lclv
-END PGP SIGNATURE-

Using sudo in a script.

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I would like to create a short script to mount and unmount SCXC cards.
 For these exfat cards to do either in Jessie has to be done as root.

I have set up sudo so that my user can run any command as root.  the
following command will mount the card without asking for my password.

sudo mount -U 6238-3434 /media/xca

(My file /etc/fstab contains the following line:

UUID-6238-3434  /media/xca  exfat   rw,users,noauto,noatime 0 0)

In order to automate the process I wrote the following script named mxca
:

#!/bin/bash
# Script to mount xca from any directory.
CURPWD=$PWD
cd /home/ken
sudo mount -U 6238-3434 /media/xca
cd $CURPWD

Running mxca alone does not produce the required result.  Running sudo
mxca together however does, even though the sudo command is also in
the script.

I can unmount the card by running command sudo umount /media/xca, but
I would like a similar script named uxca to do the same thing for
unmounting.

Is there a way to make sudo do its thing in these scripts so that I
can mount and unmount the SDXC card by running the script commands
mxca and uxca by themselves?

Regards, Ken Heard


-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iEYEARECAAYFAleiGgQACgkQlNlJzOkJmTeE9ACfaJZ3Fk27+a9YPhHtk040cCpF
D/QAnA7KuGGRMlkP3AIpHqUiTgtyaaq2
=cxii
-END PGP SIGNATURE-

pppoe setup <- Re: Testing AT&T, so routing setup

[ken]

On 05/06/2016 04:43 AM, Pascal Hambourg wrote:

Le 05/05/2016 21:05, ken a écrit :


Working on the Pi just from the bash prompt as root, how do I set the
routing table (etc.) to connect directly to the DSL modem?  The routing
table on my router currently shows:

# route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric RefUse
Iface
77.219.160.10.0.0.0 255.255.255.255 UH 0 00 ppp0
192.168.0.0 0.0.0.0 255.255.255.0   U 0 00 br0
169.254.0.0 0.0.0.0 255.255.0.0 U 0 00 br0
127.0.0.0   0.0.0.0 255.0.0.0   U 0 00 lo
0.0.0.0 77.219.160.10.0.0.0 UG 0 00 ppp0

Here's the Pi's current routing table:

Destination Gateway Genmask Flags Metric RefUse
Iface
0.0.0.0 192.168.0.1 0.0.0.0 UG0 0 0 eth0
10.10.0.0   0.0.0.0 255.255.255.0   U 0 0 0 wlan0
192.168.0.0 0.0.0.0 255.255.255.0   U 0 0 0 eth0

To plug the Pi directly into the modem, I think I'm going to need to run
ppp and then run some commands to temporarily set the IP address of the
ethernet port (eth0) and the routing table.  I much prefer to do all
this without changing config files so that when I reboot the Pi, the
settings go back to their original.  What do I need to 'apt-get' for the
ppp executables?


The interface type is PPP so you need the package "ppp" providing pppd.

The DSL provider probably uses PPP over Ethernet (PPPoE). You can 
check on your router. pppd can handle PPPoE direclty through a plugin 
and a kernel module or in userland though the package "pppoe" (more 
overhead and more options). The extra package "pppoeconf" eases the 
creation of a PPPoE configuration file in /etc/ppp/peers. Once 
created, the connection can be started and stopped manually with "pon" 
and "poff".


Some DSL providers may use the PPTP protocol instead. Then you need 
the package "pptp-linux".


Thanks for the very helpful reply.

The router is a dd-wrt, set up years ago and I don't recall anymore how 
I set it up.  So I called the ISP: they said they use pppoe.  I 
installed that on the Pi... along with ppp.






And... I'm more accustomed to using the 'route'
command to manually set up a routing table, but have used 'ip' commands
too, if that's more familiar to you.


You should not need to issue any commands. Just make sure the PPP 
config file has the options "defaultroute" to set the default route 
through the PPP connection and "replacedefaultroute" to override the 
existing default route.


I *want* to use commands.  As said in my original (& probably seemingly 
wordy, hence redacted) email, this is going to be temporary use of this 
Pi, so I don't want to set up a new configuration on it.  I want the 
default networking configuration to remain as is, plug it into a 
different connection, run some commands to change the networking to run 
ppp, and then, when I'm done with this new connection, reboot and go 
back to the previous networking config which doesn't even run ppp.


Reading "man pon" I found I can do "pon testconfig" and pon will load 
the specified "testconfig" settings file  instead of the default 
config... and then after a reboot, go back to the previous network 
settings (with no ppp running and the default route to what it was 
previously).


Said another way, this Pi is currently just part of my LAN and is 
configured to connect to a router (192.168.0.1).  I'm going to unplug it 
from that, connect it directly to the DSL modem (temporarily) to test my 
ISP.  Then, when I'm done with the test, I'll unplug it, put it back 
onto the LAN, and boot into it's previous network config.


So I expect that, after the Pi is connected directly to the modem, I'll 
need to do "ifdown eth0" first, then do "pon testconfig".  The 
/etc/ppp/peers/testconfg file is going to be somewhat different from the 
default one... on these items:


1) I have static ip addresses (2 of them), not dynamic.  So where, if 
anywhere, do I need to specify one?  And what other setting(s) to 
specify a static IP?


2) My setup needs to specify a user and password.  I'm guessing the 
'user "myusername@realm"' line should serve for that, yes?


3) The Pi has two network interfaces (shown above, in my original 
post).  How do I tell ppp to use eth0 and leave wlan0 alone?



This is actually a cool, little project... might be helpful for other 
folks who want to run tests on their ISP isolated from their LAN.  
Again, thanks for your help.

Testing AT&T, so routing setup

[ken]

Test context:

Having problems with my internet connection, technicians came out, said 
they fixed the problem, and left.  The connection was a bit better 
(fewer lost ping packets), but nowhere close to where it should be.  
They say they see no other problem with their equipment, the problem is 
probably with my equipment.  Repeat that exact same story two more times 
in the following week.


Here's what mtr shows me:

HOST: yes.mynet.net Loss%   Snt   Last   Avg Best  Wrst StDev
  1.|-- 192.168.0.1  0.0%   1000.4   0.8 0.3  20.1   2.4
  2.|-- 77.219.160.153.0%   100   71.3 100.7 9.1 321.5  90.4

So what I'm thinking is to take an RPi and my DSL modem and plug it in 
directly to the box outside (on the customer side, of course) and test 
it from there.  If I get the same results from there, I've eliminated 
everything on my side as the source of the problem except the DSL modem 
and the RPi.  Since, however, I get the same results on four other nodes 
inside the house as I get from the Pi, the problem is likely not the Pi's.



Setting up for testing:

Working on the Pi just from the bash prompt as root, how do I set the 
routing table (etc.) to connect directly to the DSL modem?  The routing 
table on my router currently shows:


# route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric RefUse 
Iface

77.219.160.10.0.0.0 255.255.255.255 UH 0  00 ppp0
192.168.0.0 0.0.0.0 255.255.255.0   U 0  00 br0
169.254.0.0 0.0.0.0 255.255.0.0 U 0  00 br0
127.0.0.0   0.0.0.0 255.0.0.0   U 0  00 lo
0.0.0.0 77.219.160.10.0.0.0 UG 0  00 ppp0

Here's the Pi's current routing table:

Destination Gateway Genmask Flags Metric RefUse 
Iface

0.0.0.0 192.168.0.1 0.0.0.0 UG0 00 eth0
10.10.0.0   0.0.0.0 255.255.255.0   U 0 00 wlan0
192.168.0.0 0.0.0.0 255.255.255.0   U 0 00 eth0

To plug the Pi directly into the modem, I think I'm going to need to run 
ppp and then run some commands to temporarily set the IP address of the 
ethernet port (eth0) and the routing table.  I much prefer to do all 
this without changing config files so that when I reboot the Pi, the 
settings go back to their original.  What do I need to 'apt-get' for the 
ppp executables?  And... I'm more accustomed to using the 'route' 
command to manually set up a routing table, but have used 'ip' commands 
too, if that's more familiar to you.


Thanks.

Re: Fwd: kencunninghan_Pedophile_alert_in_your_area_

[Ken Cunningham]

What do you want me to do with this
On May 1, 2016 8:19 AM,  wrote:

>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
> * Dear {kencunninghan} You Are Receiving This Email Because There May Be A
> Risk Of Sex Offender Activity In Your Aera. Here Is A Link To A Great
> Website Called Kids Live Safe.  The Website Show
> You All The Offenders That Live Nearby. You Can Do A Quick Search
> . It Only Takes A Few Seconds Not Only Can You See
> Who They Are And Where They Live, You Also Get Email Alerts When A New Sex
> Offender Moves Close To Your Home. It Also Has A Brunch Of Tools To Help
> Keep Our Kids Safe. Search Your Neighborhood  *
> */*/*/*/*/*
>
>

Re: change hostname & static IP & ???

[ken]

On 04/12/2016 03:27 AM, to...@tuxteam.de wrote:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On Mon, Apr 11, 2016 at 08:31:16PM -0500, David Wright wrote:

On Mon 11 Apr 2016 at 19:02:45 (-0400), ken wrote:

This is on Wheezy for a Raspberry Pi.

Using dd, I copied the SD card for one (nicely working and
configured) system onto another SD card [...]

.ssh/known_hosts [...]

Also, your new host will have the same ssh host private key. This
might or might not annoy you.


You mean those /etc/ssh/ssh_host_*key* files, yes?



Another thing to look out for:

Bits and pieces of your /etc might end up in your initramfs,
namely those needed in early boot (e.g. things needed to
bring up the root file system). The static IP address might
be needed if you have your root on NFS, for example. Or
the /etc/crypttab, if you're running with encrypted root
(I don't think so, but I lost myself some hair over this
one).

So after you have "fixed up" your SD card clone (IPs and all
that things) you might want to run update-initramfs for those
bits to find their way into the initramfs.


Thanks for letting me know.  Though I've set up a few encrypted file 
systems and compiled lots of kernels in the past, still I'm a 
feets-pointing-backwards idiot when it comes to initramfs, never needed 
to know much about it,


It sounds like "update-initramfs -v -u all" should be the last thing I 
do before the initial reboot, yes?




Beware that now you have two different file systems with
the same UUID. This might confuse some all too credulous
programs that rely on (re-) seeing the same medium if the
UUID didn't change.


Very good to know... and I didn't know that at all.  A quick look around 
found me dbus-uuidgen, a command to create a new uuid.  It's stored in 
/var/lib/dbus/machine-id says the manpage... and confirmed.  Is that the 
UUID you're referring to?


The manpage makes this one seem easy...

cd [the SD card's / mount point]
dbus-uuidgen > var/lib/dbus/machine-id


There are also things called UUIDs, or 'labels', for devices and for 
partitions and for swap(s).  Do I need to do something with them as well?

Clone SD card for new system <- Re: change hostname & static IP & ???

[ken]

On 04/11/2016 09:31 PM, David Wright wrote:

On Mon 11 Apr 2016 at 19:02:45 (-0400), ken wrote:

This is on Wheezy for a Raspberry Pi.

Using dd, I copied the SD card for one (nicely working and
configured) system onto another SD card.  I want to use this second
card for another system, so need to change the hostname and (static)
IP address.  Yesterday I grepped the whole system for one and then
the other and where I found either, I made appropriate changes
(i.e., previous hostname to new hostname, previous IP address to new
IP address).  I also zeroed out all text-based files in /var/log/
(e.g., >messages), just to start the new system with clean logs.

Both the previous (still running) system and the new one (not yet
running) are to be on the same LAN, so I'm pretty certain that no
other networking configuration-- like broadcast or netmask-- needs
to be touched (how could it need any?).

Neither of these systems has any public-facing services currently running.

My concerns are these:

Only files which grep could find were amended. (Note that grep was
run on the card when it was mounted on another system, not when the
system on that card was booted and running.)  Is there a file of
some kind, perhaps, which encodes either the hostname and/or IP
address in such a way that would hide it/them from grep?

If you regenerate ssh keys, your .ssh/known_hosts files will have
hostnames and IP numbers you might want to remove. They're encrypted.
You could blank them like the log files.

The key pairs themselves only have visible hostnames in the public
part's file.


Ah, yes!  Thanks.  Some brain cells knew that, but were sleeping.  But 
actually I want to keep that ~/.ssh/known_hosts so I can ssh from this 
new RPi to the other machines on my LAN without having to reconfigure 
all that ssh stuff... one of the several time-saving reasons I'm 
(quasi-)duplicating the SD card.


But you bring up a very good point: I don't want the new system to have 
the same /etc/ssh/ssh_host_*key* files that the previous system has.  
Though it might work okay, still I think the sshd host keys are meant to 
be unique across all LAN'd systems (ideally unique in the universe).  So 
I think I need to do "rm etc/ssh/ssh_host_*key*" so that sshd knows 
they're gone; in the past IIRC these host keys were automatically 
created, probably done in the sshd script under init.d;  I hope that 
it's the same under the new-fangled systemd code.




Since people are supposed to be able (I think, never done) to take
an SD out of one RPi and put and boot it in another RPi (with the
same hardware), I didn't make any attempt to find and change any
hardware designations, like the MAC address of the NIC.  Is that
okay?

Your router may know IP#/MAC combinations for dhcp.


I'm not using dhcpd on the router, so that won't affect me. But for 
those who do, if you can do "arp -a" on your router, that will show 
which MAC addresses are cached; "arp -d address" will delete that 
address's entry in the arp table, allowing/obligating the system to 
generate a/the new entry.





Also IPv6 link-local addresses are generated from the MAC address
(in case you've told them to any of your other machines).


Again, not using that, but others may, so find the info useful.  Thanks.




Cheers,
David.

change hostname & static IP & ???

[ken]

This is on Wheezy for a Raspberry Pi.

Using dd, I copied the SD card for one (nicely working and configured) 
system onto another SD card.  I want to use this second card for another 
system, so need to change the hostname and (static) IP address.  
Yesterday I grepped the whole system for one and then the other and 
where I found either, I made appropriate changes (i.e., previous 
hostname to new hostname, previous IP address to new IP address).  I 
also zeroed out all text-based files in /var/log/ (e.g., >messages), 
just to start the new system with clean logs.


Both the previous (still running) system and the new one (not yet 
running) are to be on the same LAN, so I'm pretty certain that no other 
networking configuration-- like broadcast or netmask-- needs to be 
touched (how could it need any?).


Neither of these systems has any public-facing services currently running.

My concerns are these:

Only files which grep could find were amended. (Note that grep was run 
on the card when it was mounted on another system, not when the system 
on that card was booted and running.)  Is there a file of some kind, 
perhaps, which encodes either the hostname and/or IP address in such a 
way that would hide it/them from grep?


Since people are supposed to be able (I think, never done) to take an SD 
out of one RPi and put and boot it in another RPi (with the same 
hardware), I didn't make any attempt to find and change any hardware 
designations, like the MAC address of the NIC.  Is that okay?


Is there anything I'm missing or forgetting?


tia for fullohelp tips.

Re: [ A little off topic] Best e-mail client for Android

[ken]

On 04/11/2016 08:08 AM, Hans wrote:

Am Montag, 11. April 2016, 08:02:13 schrieb German:

I wonder what Debian users use on their phone/tablet.

Thanks

K9-Mail


+1

Re: iceape availability

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-02-26 16:33, Bret Busby wrote:
> On 26/02/2016, Ken Heard  wrote:
>> -BEGIN PGP SIGNED MESSAGE- Hash: SHA1
>> 
>> On 2016-02-25 22:46, Bret Busby wrote:
> 
> 
> 
>>>> 
>>> 
>>> I currently have and use, seamonkey v 2.29 (.1, I think), as
>>> the latest version of which I have been aware, has been
>>> available as a .deb package.
>> 
>> Is that package available for Debian, or only for the Ubuntu
>> offshoot? I ran Lisi's URL for =Seamonkey instead of =Iceweasel
>> but received a nil return.  Since the demise of Iceape I am using
>> Seamonkey, presently version 2.31, directly from Mozilla, with
>> Wheezy.
>> 
>> Ken
>> 
> 
> It is a .deb package, and, from memory, runs on Debian as easily as
> on Ubuntu.
> 
> Try 
> https://sourceforge.net/projects/ubuntuzilla/files/mozilla/apt/pool/ma
in/s/seamonkey-mozilla-build/
>
>  which includes 2.39.
> 
Thank you, Bret, I was not aware of the Ubuntuzilla project.  I will
have to try that version of Seamonkey, and perhaps use it instead of
the "pure" one I got from Mozilla.

Regards, Ken

-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlbQWHoACgkQlNlJzOkJmTezFwCfXguHfA/uL/Uh5Iep4G9Jidwu
qcQAn2Od6HB0UJbIz32HXOjb7jZidS/4
=kgQ4
-END PGP SIGNATURE-

Re: iceape availability

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-02-25 22:46, Bret Busby wrote:
> On 25/02/2016, Richard Owlett  wrote:
>> On 2/25/2016 7:38 AM, Lisi Reisz wrote:
>>> On Thursday 25 February 2016 12:59:29 Siard wrote:
>>>> On Thu, 25 Feb 2016 14:08:20 +0800, Bret Busby wrote:
>>>>> I searched for a .deb package, for iceape, so that I could
>>>>> download the package for the iceape suite, to try to
>>>>> install it.
>>>> 
>>>> AFAIK, Iceape, being the Debian version of Seamonkey, has
>>>> been discontinued for quite some time now. There is only an
>>>> old version for squeeze (oldoldstable) in the repos.
>>> 
>>> https://packages.debian.org/search?suite=default§ion=all&arch=an
y&searchon=names&keywords=Iceape
>>>
>>>
>>> 
Lisi
>>> 
>>> 
>> 
>> That has *SECURITY* patches. But what upstream version is it 
>> based on? SeaMonkey is currently at 2.39 .
>> 
>> 
>> 
> 
> I currently have and use, seamonkey v 2.29 (.1, I think), as the 
> latest version of which I have been aware, has been available as a 
> .deb package.

Is that package available for Debian, or only for the Ubuntu offshoot?
 I ran Lisi's URL for =Seamonkey instead of =Iceweasel but received a
nil return.  Since the demise of Iceape I am using Seamonkey,
presently version 2.31, directly from Mozilla, with Wheezy.

Ken

-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlbQAmcACgkQlNlJzOkJmTerfgCeNH045cqeG11DmOrCKwYHatqR
TkkAniPkAFDvEOQPniULor3FgEKd2574
=b6HA
-END PGP SIGNATURE-

Re: Posting a message on all desktops

[ken]

On 01/30/2016 02:33 PM, Renaud (Ron) OLGIATI wrote:

Is there a way for a system admin to post a message on the desktops of all the 
machines on a LAN ?

Cheers,

Ron.



If all the desktops are just terminals (text), you could use "wall". 
But they're probably not.  (?)


You could put the desired announcement in /etc/motd and they would see 
it when they log in.


If they all have Tk installed, you could write a bit of code to pop a 
window on each machine on the lan, but you would also need access to all 
of them.


But I'm going to guess everyone is running GUIs on their desktops and 
you don't want to write code, so all I can think of-- and what we've 
done at every place I've worked-- is just to send everyone an email.

Re: updating debian 8 (jessie) stable to permanent testing

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-01-25 09:29, Adam Wilson wrote:



> Something else would be better- not using jessie-backports. If
> you're already using testing, enabling jessie-backports is
> pointless and will put you halfway into FrankenDebian territory.
> Beware.
> 
> I would do something like this:
> 
> deb http://http.debian.net/debian testing main deb
> http://http.debian.net/debian testing-updates main deb
> http://security.debian.org testing/updates main

Is it possible at this stage to use stretch instead of testing?  Then
when testing becomes stable it will not be necessary to change the
sources.list again -- unless of course at that point you want to go
directly to the new testing.

Regards, Ken

-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlalrekACgkQlNlJzOkJmTeocACdGKFvpscnn7zcDAmYsxRSbdHp
+Q0An1spB9zflEgCY1CRjAfrXjCofYGU
=iv0C
-END PGP SIGNATURE-

Re: updating debian 8 (jessie) stable to permanent testing

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2016-01-25 10:34, Francis Gerund wrote:
>>> 
>>> -- or, would something else be better?
>>> 
>> 
>> Something else would be better- not using jessie-backports. If
>> you're already using testing, enabling jessie-backports is
>> pointless and will put you halfway into FrankenDebian territory.
>> Beware.
>> 
>> I would do something like this:
>> 
>> deb http://http.debian.net/debian testing main deb
>> http://http.debian.net/debian testing-updates main deb
>> http://security.debian.org testing/updates main
>> 
>> 
> 
> 
> 
> Hi.
> 
> I am not using testing - yet.  Here is the current
> /etc/apt/sources.list :
> 
> # deb cdrom:[Debian GNU/Linux 8.2.0 _Jessie_ - Official amd64
> NETINST Binary-1 20150906-11:09]/ jessie main
> 
> deb  http://ftp.us.debian.org/debian/ jessiemain 
> contrib non-free deb-src  http://ftp.us.debian.org/debian/ jessie
> main contrib non-free
> 
> deb  http://security.debian.org/  jessie/updatesmain 
> contrib non-free deb-src  http://security.debian.org/
> jessie/updatesmain contrib non-free
> 
> # jessie-updates, previously known as 'volatile' deb
> http://ftp.us.debian.org/debian/ jessie-updatesmain contrib
> non-free deb-src  http://ftp.us.debian.org/debian/ jessie-updates
> main contrib non-free
> 
> # jessie-backports, previously on backports.debian.org deb
> http://ftp.us.debian.org/debian/ jessie-backports  main contrib
> non-free deb-src  http://ftp.us.debian.org/debian/ jessie-backports
> main contrib non-free
> 
> 
> 
> 
> ---  So, how about this proposed /etc/apt/sources.list :
> 
> 
> # deb cdrom:[Debian GNU/Linux 8.2.0 _Jessie_ - Official amd64
> NETINST Binary-1 20150906-11:09]/ jessie main
> 
> deb  http://ftp.us.debian.org/debian/ testingmain 
> contrib non-free deb-src  http://ftp.us.debian.org/debian/ testing
> main contrib non-free
> 
> deb  http://security.debian.org/  testing/updatesmain 
> contrib non-free deb-src  http://security.debian.org/
> testing/updatesmain contrib non-free
> 
> # jessie-updates, previously known as 'volatile' deb
> http://ftp.us.debian.org/debian/ testing-updatesmain contrib
> non-free deb-src  http://ftp.us.debian.org/debian/ testing-updates
> main contrib non-free
> 
> # jessie-backports, previously on backports.debian.org # deb
> http://ftp.us.debian.org/debian/ jessie-backports  main contrib
> non-free # deb-src  http://ftp.us.debian.org/debian/
> jessie-backports  main contrib non-free
> 
> Is this about right, to convert from stable to permanent testing?
> Or has having backports enabled already made it not doable?

Adam Wilson did say _not_ to use jessie-backports, as it would be
"pointless".  You could I suppose use testing-backports instead, but
there will be nothing in it until after testing becomes stable.

Regards, Ken



-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlalrSYACgkQlNlJzOkJmTdh/wCeNDdpPBzJAw0vZmBcaYt9Zl/6
LicAn2RpbBq5KyVHq8bu9RRkoKY+nfuB
=JT6z
-END PGP SIGNATURE-

Konqueror default view mode -- how to make it ‘really’ permanent?

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

My default view profile for Konqueror is ‘File management’.  In that
profile I have set the ostensible default view mode to ‘detailed list’.
However, during a Konqueror session it does not always stay that way;
sometimes it reverts to ‘icon view’.

For example, if I open a text file from Konqueror, the file will open
in the application I have specified in ‘file associations’:
LibreOffice for odt files, Kedit for txt files, etc.  After such files
are closed from Konqueror, the file list reverts to ‘detailed list’.

If however I open any other file type, such as a jpg file, after such
file is closed the file list reverts to ‘icon view’ instead of the
ostensible default ‘detailed list’ file view.  How do I cause
Konqueror in such situations to revert to that default?

In file .kde/share/config/konquerorrc there is the following entry:

[ModeToolBarServices]
external=konq_detailedlistview

Is that entry related to the problem I am experiencing?

Regards, Ken
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlahpBUACgkQlNlJzOkJmTcX5QCfaadz/NG6vzIqtXMbeN3wp4od
dVsAmgKo4xHS+CwsCL0CXMwEf0reNtHX
=5YXu
-END PGP SIGNATURE-

Re: problems running backup

[ken]

On 01/21/2016 12:33 AM, Bob Holtzman wrote:

On Wed, Jan 20, 2016 at 04:59:05PM -0500, ken wrote:

On 01/20/2016 04:33 PM, Bob Holtzman wrote:

2) Target partition was mounted BUT had quietly been auto-renamed by

my system k/t that another partition was already bearing the same
label name. Fodder for another eventual thread and why I love and
advocate UUIDs over labels or other.

1) Target was external HD and was mounted.


The simple matter of permissions is often the problem.  This kind of problem
has cropped up for me before over an upgrade (not yet with Debian though).

Also, what exactly is the error message?  Can you increase its verbosity
so's to examine that as well?


In case you deleted my original post, the error message is:

  rsync: mkdir "/media/cf0a98ed-3c11-4107-b61e-f5139d024396/Jessie-laptop"
  failed: No such file or directory (2)
  rsync error: error in file IO (code 11) at main.c(674) [Receiver=3.1.1]

I already had verbosity turned on except I believe that refers to the list of 
files as they're
backed up. Not sure how to expand an error message, or even if that's possible.

Permissions are 700 for the executable file.

I'm snowed. Any other ideas?



Can you do at the cli

'mkdir "/media/cf0a98ed-3c11-4107-b61e-f5139d024396/Jessie-laptop"'

as the same user as runs rsync?

Re: problems running backup

[ken]

On 01/20/2016 04:33 PM, Bob Holtzman wrote:

2) Target partition was mounted BUT had quietly been auto-renamed by
>my system k/t that another partition was already bearing the same
>label name. Fodder for another eventual thread and why I love and
>advocate UUIDs over labels or other.

1) Target was external HD and was mounted.


The simple matter of permissions is often the problem.  This kind of 
problem has cropped up for me before over an upgrade (not yet with 
Debian though).


Also, what exactly is the error message?  Can you increase its verbosity 
so's to examine that as well?

Re: Need a calculator that knows about coulombs

[ken]

On 12/28/2015 01:59 AM, Gener Badenas wrote:



On Wed, Dec 23, 2015 at 9:09 AM, Gene Heskett mailto:ghesk...@shentel.net>> wrote:

Greetings all;

Do we have such a beast?

Thanks.




A quasi-cli utility called "units" does that.

Sound from computer to TV monitor via HDMI -- how?

[Ken Heard]

onf.d/asoundrc.conf:

Pcm.! Default   {
type hw
card 0
device 7
}

(b) Adding the following to /usr/share/alsa/alsa.conf.d/asoundrc.conf:

defaults.ctl.card 1
defaults.pcm.card 1
defaults.timer.card 1

I tried both, and also tried various combinations of card and device
numbers.  None had the desired effect.  In fact when I tried any of
these and left the other entries in that file open, access to the
analogue speakers was cut off.

So where do I go now?  I will be thankful for any help received from
other users on this list.

Regards, Ken Heard






-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.19 (GNU/Linux)

iEYEARECAAYFAlZ2g5wACgkQlNlJzOkJmTdcpwCdE4C9ltf2gAMTorTXruD0VYJ9
7AEAn1mtPmLqmI57m1sXtOIZnt+NS+C3
=gVtS
-END PGP SIGNATURE-

Re: sudo does not respond to settings in /etc/sudoers

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2015-11-16 19:02, Andreas Henriksson wrote:

> For what it's worth, if you want to avoid logging out/in you can 
> temporarily join the group in your shell by using "newgrp sudo".

Thank you for that tip; it is useful to know.

Ken


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlZJ6QYACgkQlNlJzOkJmTd7XQCfVkYMZawUyHDxy7A/O96OYmaF
SfIAmwTd50WNppet3r1P9DknHJtILdGv
=JG+B
-END PGP SIGNATURE-

Re: sudo does not respond to settings in /etc/sudoers

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

On 2015-11-15 23:26, Teemu Likonen wrote:

> Maybe you added the user to the "undo" group

Surely you mean the "sudo" group?

> but forgot to test it with a new login session. Group changes
> don't affect the current login session (unless newgrp is used).

My eureka moment! I forgot to logout and in again as my user before
testing the various options. I tested the two options to do what I
wanted to do as described in my original post, but not both on, or
open at the same time. I found that adding to the "user privilege
specification" section in the sudoers file the following line DOES work:

ken ALL=NOPASSWD: ALL

However, removing that line, and in the TDE User Manager (Kuser)
simply adding my user to the "sudo" group does NOT work. That option
DOES nevertheless work in my Jessie box. Strange.

Thanks guys for collectively putting me on the right track.

Ken




-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlZJfKEACgkQlNlJzOkJmTfY6ACeOijk3QTayP+2rIpUJ0N9Zj2x
sSYAn01UPKZyBd6QAa1t0GbcT+9d7hF3
=hCiX
-END PGP SIGNATURE-

sudo does not respond to settings in /etc/sudoers

[Ken Heard]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

In my Wheezy box I want to be able to run any root command as my user
without having to enter a password to do so.  I assume that there are
two ways to do so: either make my user a member of the sudo group, or
add a line to sudoers giving such privileges to my user.  Neither worked.

My sudoers file reads as follows:

#
# This file MUST be edited with the 'visudo' command as root.
#
# Please consider adding local content in /etc/sudoers.d/ instead of
# directly modifying this file.
#
# See the man page for details on how to write a sudoers file.
#
Defaultsenv_reset
Defaultsmail_badpass
Defaults
secure_path="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"

# Host alias specification

# User alias specification

# Cmnd alias specification

# User privilege specification
rootALL=(ALL:ALL) ALL
# ken   ALL=NOPASSWD: ALL
# martinAll=(ALL:ALL) ALL
# Allow members of group sudo to execute any command
%sudo   ALL=(ALL:ALL) ALL

# See sudoers(5) for more information on "#include" directives:

#includedir /etc/sudoers.d

I first entered the line under "user privilege specification"
beginning with "ken" -- without the "#" of course -- it did not work.
 Next I commented out that line and added to the sudo group "ken" my
user.  That did not work either.

The command "visudo -c" returns the following:

/etc/sudoers: parsed OK
/etc/sudoers.d/README: parsed OK
/etc/sudoers.d/tde-secure-path: parsed OK

Strangely enough I have the same sudoers file in another computer
using Jessie.  On that computer both the options I tried for the
wheezy computer worked.

Is this a bug for the Wheezy computer, or is there something I did not
do or did wrong to prevent me from achieving the result I desired?

Regards, Ken


-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.12 (GNU/Linux)

iEYEARECAAYFAlZG7JkACgkQlNlJzOkJmTd6LACeMoTBd6V5VLmOMfsyg3p1DuXI
rxEAnRfp5FdfLZwxiNB4fZHYi9MMxJx1
=PwCP
-END PGP SIGNATURE-

Re: Installing mSATA Solid State Drive in Dell Precision M3800

[ken]

On 11/03/2015 01:28 AM, David Christensen wrote:

On 11/02/2015 07:49 PM, ken wrote:

I'm considering getting a Dell Precision M3800 Mobile Workstation, would
want to install my own mSATA SSD in it.  Does anyone here have one of
these machines?  If so, is an SSD a user-installable component?  And if
so again, how much needs to be taken apart?


Download the service manual from Dell.  It should provide instructions
for replacing the drive(s) -- HDD/ SSD, "mobility", etc..  It's nice to
pick a model that you can get to the drive(s) by removing some screws
and sliding a carrier out, rather than having to take off the bottom
cover, etc..

David


John and David,

Thanks for replying.  Yes, I had a look at that Dell Service Manual 
before posting.  Judging from that manual, it looks simple enough.  But 
then my experience with service manuals of all kinds has shown me that 
they often leave out or gloss over steps, which is why I was hoping to 
get a response from someone who's actually done this.  The other issue 
regards the warranty: That service manual says only that replacing or 
installing some parts may void the warranty; it doesn't get any more 
specific than that.  So that was the other thing I was hoping to hear 
first-hand experience about.  And too, sometimes screws are painted with 
some kind of colored faux-glue, not locktite, to tell any authorized 
service person that that screw has been tampered with, voiding the 
warranty.  Again, not something a service manual would warn about.

Installing mSATA Solid State Drive in Dell Precision M3800

[ken]

I'm considering getting a Dell Precision M3800 Mobile Workstation, would 
want to install my own mSATA SSD in it.  Does anyone here have one of 
these machines?  If so, is an SSD a user-installable component?  And if 
so again, how much needs to be taken apart?


Thanks much.